Skip to main content

Full text of "Persia past and present; a book of travel and research, with more than two hundred illustrations and a map"

See other formats


From apainting by Jay Httmbidge. Courtesy of the Century Co. 

KING DARIUS OFFERS SACRIFICE UNTO ORMAZD 
(See page 808) 



PERSIA PAST AND PRESENT 

A BOOK OF TRAVEL 
AND RESEARCH 

WITH MORE THAN TWO HUNDRED ILLUSTRATIONS AND A MAF 



BY 



A. V. WILLIAMS JACKSON 

PROFESSOR OF INDO-IRANIAN LANGUAGES, AND SOMETIME 

ADJUNCT PROFESSOR OF THE ENGLISH LANGUAGE 

AND LITERATURE IN COLUMBIA UNIVERSITY 



OF THF 

UNIVERSITY 

OF 




THE MACMILLAN COMPANY 

LONDON: MACMILLAN & CO., LTD. 
190G 

All rights reserved 






' \ 
? - 



\-JBEHEBM. 



COPYRIGHT, 1906, 
BY THE MACMILLAN COMPANY. 

Set up and electrotyped. Published September, 1906. Reprinted 
December, 1906. 



NortoooH 

J. 8. Gushing & Co. ik-rwick A Smith Co. 
Norwood, Muss., U.S.A. 



TO MY MOTHER 

More is thy due than more than all can pay.' 

Sll AKSPERE. 






PREFACE 

THE preparation of this volume has been a work after my 
own heart for the past three years, and I am now almost sorry 
that it is finished, although it has been carried forward amid 
many distractions of an official and public nature besides my 
regular duties at the University. I have entitled it a book of 
travel and research, having essayed the somewhat difficult task 
of combining these two themes into a union which I hope will 
be found true, at least in spirit, even if imperfect when judged 
by the standards that I should like to uphold. 

I was tempted at first to label some of the chapters, like 
those on Takht-i Suleiman, Kangavar, Ragha, and a part of 
the section on the Old Persian Inscriptions, with a warning, 
4 this chapter is dedicated to the student,' and to prefix to 
other chapters, like those on the different cities, By Caravan 
and Cavalcade, Pasargadse, and Persepolis, a prefatory line, 
'dedicated to the general reader.' I decided against such a 
procedure, because I believe that the interests of both are ulti- 
mately one, and if the general reader enjoys a comfortable forty 
winks while certain technical matters are being discussed, he 
will awaken refreshed to resume his interest at a point where 
the specialist may begin to nod. But these very comments 
may have in themselves a somnolent effect and defeat their 
own end by superinducing the results they are seeking to avert. 
For that reason I shall refrain from adding others, and shall 
proceed rather to bring out the points which I wish to 
emphasize. 

By hard work during my stay in Persia, I succeeded in 
seeing a good deal of the country and observing Persian life 

vii 



viii PREFACE 

with that keenness of interest which enthusiasm for the sub- 
ject produces and that closeness of application which years of 
preparation tend to cultivate. I could have wished that the 
time and means at my disposal had been more abundant, but 
4 half a loaf is better than no bread at all,' and years of expe- 
rience in travel have taught me that impressions gained in a 
few months are often more vivid than when the stay in the 
country is prolonged indefinitely. Of course, there is the 
well-recognized danger of receiving wrong impressions and of 
generalizing from insufficient data, but I have tried to reduce 
the factor of error as far as possible by seeing as much of the 
country as I could, so as to be able to distinguish between 
local and general conditions, and by subjecting my observations 
to the test of comparison with the history of the country from 
the earliest times. How far I have succeeded or failed must 
be decided by those who are competent to judge. I can only 
plead in my own behalf that I have worked conscientiously 
throughout to make the book as good as I was able, and one 
example of this will be found in the attention given also to 
illustration. The illustrations that are ordinarily presented 
I have supplemented by pictures of subjects that are rarely 
chosen, and I have done this mainly from a collection of photo- 
graphs taken by myself on the journey or by friends in Persia. 
How little can be done without the aid of one's predecessors, 
when writing a book, will be best appreciated by those who 
have aimed, like myself, at adding a mite to the body of knowl- 
edge already existing. On that account I have constantly 
and without hesitation consulted the works of my fellow- 
laborers in the field during the past three thousand years, from 
Zoroaster and Herodotus, through the Arab writers, to Ker 
Porter, Browne, and Curzon. I have been careful everywhere 
to quote my sources for any special piece of information and 
to acknowledge my indebtedness without hesitation. If the 
footnotes seem more abundant than usual, it is because of 
this desire faithfully to provide the reader with the means of 



PREFACE IX 

consulting the material used at each particular point or to 
enable him to call into question a view I have expressed, be- 
cause I have frankly stated the fact when the authorities are at 
variance. 

There are several respects in which I should have liked to 
be able to bring the book nearer to completeness, but I think 
I may say with all modesty that some compensation for this 
will be found in the suggestiveness of the volume because of 
the light it throws upon certain historical points which were 
not previously clear or even known before ; and I also believe 
that many readers will welcome a new presentation of matters 
that were already familiar to them. The student of Zoroas- 
trianism, moreover, will find many points to consider in con- 
nection with his own work, especially the numerous identifica- 
tions suggested for ancient sites, or explanations proposed for 
doubtful passages in the literature, as I have always kept that 
in view. 

I have spoken of my indebtedness to those of the past; I 
have also an obligation to the present which I deeply feel. 
First of all, I am indebted to President Nicholas Murray 
Butler and the Trustees of Columbia University for granting 
me leave of absence to visit Iran. Then, I am grateful to the 
many friends, as well as officials, who gave me suggestions and 
aid in regard to my journey. The Christian missionaries at 
home were especially helpful in this respect, and in Persia 
they welcomed me at their homes with a hospitality for which 
I shall ever be grateful. 

In the preparation of the book on my return, I received 
many serviceable hints from Mr. Frank E. Morgan, of Ben- 
nington, Vermont, and I owe a special debt of gratitude to 
two of my student friends. To my pupil, Dr. Louis H. Gray, 
former Fellow in Indo-Iranian Languages at Columbia, I wish 
to express my sincere thanks for generous assistance rendered 
in reading over the manuscript of a large part of the book 
before it went to the press and for his counsel with regard 



X PREFACE 

to matters of general presentation. To my younger pupil, 
Mr. George C. O. Haas, A.M., Fellow in Indo-Iranian, I am 
especially grateful for help from the beginning. Hardly a 
day has passed since the first chapter was written without 
his giving me some aid, either by going through the entire 
manuscript with an eye to matters of detail, or by correcting 
with me every page of proof as it came from the compositor, 
or by preparing the Index, the value of which the reader will 
easily recognize. There are many other friends whose kind- 
ness I recall, but if their names be missing in print, it is not 
because they are forgotten, but because they are 4 registered 
where every day I turn the leaf to read them,' including among 
this number the publishers, whose generous attitude toward 
the subject has enabled me to present the work in a way that 
I hope may find favor with the reader, general and special, 
for whom the book is written. 

A. V. WILLIAMS JACKSON. 
COLUMBIA UNIVERSITY, 
May 7, 1906. 



CONTENTS 

PAGE 

PREFACE vii 

LIST OF ILLUSTRATIONS xiii 

LIST OF WORKS OF REFERENCE xxvi 

LIST OF ABBREVIATIONS xxx 

CHAPTEB 

I. EN ROUTE FOR THE LAND OF THE LION AND THE SUN . 1 

II. TIFLIS, THE CAPITAL OF TRANSCAUCASIA .... 7 
III. ERIVAN, MOUNT ARARAT, AND THE ROAD TO THE PERSIAN 

BORDER 16 

^ IV. PERSIA, THE LAND AND ITS HISTORY, AND OUR INTEREST 

IN THE COUNTRY 24 

V. THROUGH THE SNOW FROM THE ARAS TO TABRIZ . . 33 
VI. TABRIZ, THE RESIDENCE OF THE CROWN PRINCE ... .39 

VII. ZOROASTER AND THE AVESTA 57 

VIII. AROUND LAKE URUMIAH 70 

IX. URUMIAH, A SUPPOSED EARLY HOME OF ZOROASTER . . 87 

X. BY CARAVAN AND CAVALCADE 108 

XL TAKHT-I SULEIMAN, AN ANCIENT FORTIFIED TOWN IN 

RUINS 124 

XII. HAMADAN, THE ANCIENT ECBATANA 144 

XIII. THE ROCK INSCRIPTIONS OF THE GREAT PERSIAN KINGS . 175 

XIV. THE GREAT BEHISTAN ROCK AND AN ASCENT TO READ THE 

CUNEIFORM RECORDS OF KING DARIUS . . . .186 

XV. TAK-I BOSTAN AND KERMANSHAH 213 

XVI. THE GREAT RUINED TEMPLE OF THE PERSIAN DIANA AT 

KANGAVAR 234 

xi 



xii CONTENTS 

CHAPTER PAGE 

XVII. FROM HAMADAN TO THE RUINED FIRE-TEMPLE NEAR 

ISFAHAN 245 

XVIII. ISFAHAN, THE FORMER CAPITAL 262 

XIX. ANCIENT PASARGAD.E AND THE TOMB OF CYRUS THE 

GREAT 278 

XX. PERSEPOLIS AND ITS MONUMENTS 294 

XXI. SHIRAZ, THE HOME OF THE PERSIAN POETS . . . 321 

XXII. FROM SHIRAZ TO YEZD 339 

XXIII. THE ZOROASTRIANS OF YEZD 353 

XXIV. ZOROASTRIAN RELIGIOUS CUSTOMS AT YEZD . . . 378 

XXV. FROM YEZD TO TEHERAN 401 

XXVI. TEHERAN, THE MODERN CAPITAL OF PEUSIA . . . 418 

XXVII. RUINS OF REI, OR ANCIENT RAGHA 428 

XXVIII. THROUGH MAZANDARAN TO THE CASPIAN SEA . . 442 

GENERAL INDEX 449 

INDEX OF PASSAGES 469 

MAP OF PERSIA 472 



LIST OF ILLUSTRATIONS 

KING DARIUS OFFERS SACRIFICE UNTO ORMAZD . . Frontispiece 
From a painting by Jay Hambidge. 

PAQB 

OLD FORTIFICATIONS AT ELIZABETPOL, THE ANCIENT GANJAH . 3 
From a photograph. 

THE OLD CITADEL OF TIFLIS 5 

From a print. 

GENERAL VIEW OF TIFLIS 5 

From a print. 

THE MICHAELIS BRIDGE OVER THE KURA AT TIFLIS ... 7 
From a print. 

THE CAUCASIAN MUSEUM, TIFLIS 7 

From a print. 

RUINED TOWER AND AQUEDUCT OVERLOOKING THE BOTANICAL 

GARDEN 10 

From a print. 

THE TARTAR MEIDAN AT TIFLIS 10 

From a print. 

YEZIDIS, OR SO-CALLED DEVIL- WORSHIPPERS, IN TIFLIS ... 12 
From a photograph. 

TYPES OF YEZIDIS, OR DEVIL-WORSHIPPERS, AT TIFLIS ... 14 
From a photograph. 

TIFLIS CITADEL (Another View) . . . . . . . .14 

From a print. 

MOUNT ARARAT, NEAR ERIVAN ........ 17 

From a print. 

RUSTOM, MY GEORGIAN GUIDE TO THE FRONTIER . . . .17 
From a photograph by the author. 

ERIVAN 19 

From a print. 

ARMENIAN CHURCH AND MONASTERY AT ECHMIADZIN ... 19 
From a print. 

xiii 



xiv LIST OF ILLUSTRATIONS 

PAGE 

ECHMIADZIN IN THE SNOW . .21 

From a photograph by the author. 

BAZAAR OF ERIVAN 21 

From a print. 

A CARAVAN IN THE SNOW 23 

From a photograph by the author. 

THE DIEULAFOY FRIEZE OF ARCHERS FROM SUSA .... 24 
From Perrot and Chipiez, Histoire de I' Art, 5. 868. 

DARIUS III DEFEATED BY ALEXANDER IN THE BATTLE OF ISSUS . 26 

From the Pompeian Mosaic in the Naples Museum. 

YAZDAGARD III SLAIN IN A MILL AT MERV 26 

From the Columbia University Manuscript of the Shah Namah. 

THE PORTAL OF XERXES AT PERSEPOLIS 28 

From a photograph by A. Sevruguin. 

His ROYAL HIGHNESS MUZAFFAR AD-DIN, SHAH OF PERSIA . . 30 
From a photograph. 

THE PERSIAN INN AT JULFA, WITH THE CUSTOM HOUSE ON THE 

LEFT 32 

From a photograph by the author. 

A PART OF THE 'ROAD' BETWEEN JULFA AND MARAND . . 32 
From a photograph by the author. 

MOUNTAIN VILLAGE BURIED IN SNOW BETWEEN MARAND AND 

SOFIAN 35 

From a photograph by the author. 

A RUINED CARAVANSARAI ON THE WAY 37 

From a photograph by the author. 

TABRIZ 39 

From a photograph. 

THE ARK, OR CITADEL, OF TABRIZ 40 

From a photograph. 

THE BLUE MOSQUE, TABRIZ 42 

From a photograph. 

INTERIOR OF BAZAAR AT TABRIZ 44 

From Weeks, From the Black Sea, p. 51. 

A COURTYARD ADJOINING THE BAZAAR, TABRIZ .... 46 
From a photograph. 



LIST OF ILLUSTRATIONS X V 



PAOK 



AN ACH/EMENIAN SEAL : THE KlNG SLAYING A MONSTER . . 51 

From a drawing by Mr. Percy Bodenstab. 

THE CROWN PRINCE'S GARDEN (Bagh-i Shamal, Northern Garden) . 51 

Courtesy of * House and Garden.' 

SAFAR AWAITING DIRECTIONS 55 

From a photograph. 

A PERSIAN PAGE 58 

From a photograph. 

IDEALIZED PORTRAIT FROM A SCULPTURE SUPPOSED TO REPRESENT 

ZOROASTER 62 

From Karaka, History of the Parsis, 2. 146. 

AN ANCIENT MAGIAN 69 

From Hyde, Historia Religionis Veterum Persarum, p. 369. 

A MANUSCRIPT OF THE AVESTA WITH PAHLAVI TRANSLATION . 69 
From the Ms. Jp. 1. 

THE TYPICAL VILLAGE DOG OF AZARBAIJAN 76 

From a photograph by the author. 

MY GUIDE TO THE SASANIAN SCULPTURE 76 

From a photograph by the author. 

SASANIAN BAS-RELIEF, NEAR THE PLAIN OF SALMAS ... 81 
From a photograph. 

THE GREAT PLAIN OF URUMIAH . 83 

From a photograph. 

PLOWING WITH BUFFALOES NEAR URUMIAH 83 

From a photograph. 

A GLIMPSE OF URUMIAH . . . 85 

From a photograph. 

NORTHERN END OF LAKE URUMIAH 85 

From a photograph by the author. 

A BUTCHER AND A COBBLER, URUMIAH 87 

From a print. 

WOMEN WASHING CLOTHES, URUMIAH 87 

From a photograph. 

EXCAVATIONS IN DEGALAH ASH-HILL, NEAR URUMIAH ... 90 
From a photograph by the late Rev. B. W. Labaree. 



XVI LIST OF ILLUSTRATIONS 

PAGE 

SPECIMEN OF ANCIENT POTTERY FROM URUMIAH .... 90 
In the author's collection. 

A PlT EXCAVATED IN THE DEGALAH ASH-MOUND .... 92 

From a photograph by the author. 

POTTERY FROM THE TERMANI ASH-HILL, URUMIAH .... 92 
From a photograph by the author. 

A TYPICAL MUD HUT AT TERMANI, NEAR URUMIAH ... 94 
From a photograph by the author. 

AN OLD MILL NEAR URUMIAH 94 

From a photograph by Mrs. C. S. Blackburn. 

BAS-RELIEF CYLINDER FOUND NEAR URUMIAH . . . 96 

In the Metropolitan Museum of Art, New York. 

OPENING AN ANCIENT GRAVE IN THE SIDE OF THE GEOG TAPAH 

ASH-HILL 96 

From a photograph by Mr. Paul Shimmon. 

MART MARIAM CHURCH, URUMIAH 101 

From a photograph by Mrs. C. S. Blackburn. 

A TYPICAL GROUP OF MULLAHS AND SEI'DS 101 

From a photograph. 

AN APPROACHING CAMEL TRAIN 108 

From a photograph by the author. 

PACKING BAGGAGE BEFORE SUNRISE 108 

From Weeks, From the Black Sea, p. 63. 

LOWERING LUGGAGE FROM THE HOUSETOP AT DAWN . . .112 
From Weeks, From the Black Sea, p. 98. 

A PERSIAN TELEGRAM 116 

Reproduced from the original. 

A CARAVAN OF NOMADS 117 

From a photograph. 

CHACHILI HILL, NEAR URUMIAH 117 

From a photograph by the author. 

SARALAN, A VILLAGE OF MUD HOUSES 119 

From a photograph by the author. 

KHOR KHORAH, NEAR THE FOREST OF REEDS 119 

From a photograph by the author. 



LIST OF ILLUSTRATIONS xvii 

PAGE 

A CARAVAN LEADER 122 

From a photograph by the author. 

HOSHABAH, MY GUARD FROM URUMIAH TO MlANDOAB . . . 122 

From a photograph. 

A MOUNTED GUARD 124 

After a photograph by the author. 

A TYPICAL PERSIAN VILLAGE 124 

Courtesy of ' House and Garden.' 

RAWLINSON'S PLAN OF TAKHT-I SULEIMAN 128 

From Rawlinson, Five Oriental Monarchies, 3. 27. 

THE RUINED FIRE-TEMPLE AT TAKHT-I SULEIMAN, BURIED IN THE 

SNOW 133 

From a photograph by the author. 

TAKHT-I SULEIMAN FROM THE EAST 133 

After a sketch by the author. 

AFRASIAB TAKEN CAPTIVE BY HOM 140 

From the Columbia University Manuscript of the Shah Namah. 

BATTLE BETWEEN THE IRANIANS AND TURANIANS .... 140 
From the Columbia University Manuscript of the Shah Narnah. 

DISTANT VIEW OF HAMADAN AND MOUNT ALVAND .... 145 

From a photograph by Mirza Sahak, of Hamadan. 

THE MUSALLAH, HAMADAN 145 

From a photograph. 

ROUGH PLAN OF HAMADAN 146 

Drawn by Aram Zohrabian. 

PERSIAN CATS AT THE MISSION HOUSE, HAMADAN .... 151 

From a photograph. 

THE MUSALLAH HILL, HAMADAN 154 

From a photograph by Mirza Sahak. 

THE LION OF HAMADAN 160 

From a photograph by Rev. C. H. Stileman. 

THE LION'S FACE 160 

From a photograph by Mirza Sahak. 

BRIDGE OVER THE RIVER AT HAMADAN MOUNT ALVAND IN THE 

BACKGROUND 165 

From Flandin and Coste, Voyage en Perse, Moderne, pi. 68. 



xvm LIST OF ILLUSTRATIONS 



PAGE 



TOMB OF AVICENNA (Ibn Sina) . . . . . . . .167 

From a photograph by the author. 

TOMB OF ESTHER AND MORDECAI 167 

From a photograph by Mirza Sahak. 

THE GANJ NAMAH INSCRIPTIONS IN WINTER 170 

From a photograph by the author. 

THE WORD ' GOD' (BAGA) IN ANCIENT PERSIAN CUNEIFORM LETTERS 170 
From a drawing by the author. 

THE GANJ NAMAH CUNEIFORM INSCRIPTIONS OF DARIUS AND XERXES 

NEAR HAMADAN 172 

From a photograph by Mirza Sahak. 

THE ROCK OF BEHISTAN FROM THE SOUTH 177 

From Rawlinson, Five Oriental Monarchies, 3. 31. 

VIEW OF THE BEHISTAN ROCK FROM THE UPPER END . . . 177 
Sketched from a photograph by the author. 

THE KERMAN INSCRIPTION OF KING DARIUS. 1. THE PERSIAN TEXT 181 
From a photograph. 

THE KERMAN INSCRIPTION OF KING DARIUS. 2. THE ELAMITIC TEXT 183 
From a photograph. 

THE KERMAN INSCRIPTION OF KING DARIUS. 3. THE BABYLONIAN 

TEXT 184 

From a photograph. 

REAR VIEW OF THE ROCK OF BEHISTAN 186 

From a photograph by Antoin Sevruguin, of Teheran. 

THE UNFINISHED ROCK-SCREEN AT BEHISTAN 188 

From a photograph by Antoin Sevruguin. 

UP ON THE LEDGE 188 

From a photograph by the author. 

THE BEHISTAN SCULPTURES AND INSCRIPTIONS . . . .192 
From Rawlinson, Journal of the Royal Asiatic Society, 10. pi. 2. 

KULI, THE HEAD GUIDE IN THE ASCENT U'J 

From a photograph by the author. 

PHOTOGRAPH OF A SECTION OF THE INSCRIPTION OF DARIUS, BE- 
HISTAN 1. 61-71 196 

Taken by the author. 



LIST OF ILLUSTRATIONS xix 

PAGE 

FLANDIN'S SKETCH OF THE GOTARZES SCULPTURE .... 211 
From Flandin and Coste, Voyage en Perse, Ancienne, 1. pi. 19. 

THE PARTHIAN SCULPTURE OF GOTARZES, MUTILATED BY A Moi>- 

ERN INSCRIPTION 211 

From a photograph by A. Sevruguin. 

THE VAULTED STONE GROTTOS OF TAK-I BOSTAN .... 213 
From a photograph. 

THE SCULPTURED PANEL AT TAK-I BOSTAN 215 

From de Morgan, Mission Scient'ifique en Perse, 2. pi. 34; 4. pi. 35. 

THE NIMBUS-CROWNED FIGURE 216 

From a photograph by the author. 

THE SCULPTURED FIGURE OF THE FALLEN FOE AT TAK-I BOSTAN 218 
From a photograph by the author. 

THE LARGER AND THE SMALLER ARCH 220 

From Flandin and Coste, Voyage en Perse, Ancienne, 1. pi. 3. 

TAK-I BOSTAN, OR THE ARCHED GARDEN 220 

From a photograph by the author. 

VIEW OF THE LARGER ARCH .' 222 

From a photograph. 

BAS-RELIEFS OF KHOSRU II 222 

From a photograph. 

KERMANSHAH : THE PARADE-GROUND AND THE GOVERNOR'S PALACE 231 
From a photograph. 

A PERSIAN SHEPHERD (near Sahnah) 234 

From a photograph by the author. 

A TYPICAL CARAVANSARAI 234 

From a photograph. 

THE RUINED TEMPLE AT KANGAVAR 243 

From Flandin and Coste, Voyage en Perse, Ancienne, 1. pi. 21. 

RUINS OF THE TEMPLE AT KANGAVAR 243 

From a photograph by the author. 

PLOWING WITH Cows 246 

From a photograph by the author. 

PERSIAN SHOVELS 246 

From a photograph by the author. 



XX LIST OF ILLUSTRATIONS 

PAGE 

THE ANCIENT FIRE-TEMPLE NEAR ISFAHAN 254 

From a photograph by the author. 

PART OF THE TEMPLE-PRECINCT ADJOINING THE SHRINE OF FIRE 254 
From a photograph by the author. 

FLANDIN'S SKETCH OF THE HILL OF THE ISFAHAN FIRE-TEMPLE . 263 
From Coste, Monuments Modernes de la Perse, pi. 53. 

ISFAHAN AND THE BRIDGE OF ALI VERDI KHAN .... 263 
From a photograph. 

MASJID-I SHAH, THE KING'S MOSQUE . 266 

From a photograph by Mr. A. O. Wood. 

PALACE OF THE ZIL AS-SULTAN, PRINCE GOVERNOR OF ISFAHAN . 266 
From a photograph by Mr. A. O. Wood. 

MADRASAH OF SHAH HUSEIN, FROM THE CHAHAR BAGH . . 268 
From a photograph by Mr. A. O. Wood. 

WITHIN THE COURT OF THE MADRASAH 268 

From a photograph by Mr. A. O. Wood. 

THE MEIDAN OF ISFAHAN 270 

From a photograph by Tooni Johannes, of Isfahan. 

PAVILION OF MIRRORS 270 

From Dieulafoy, L'Art Antique de la Perse, 3. pi. 6. 

THE TOMB OF CYRUS THE GREAT 279 

From a photograph by the author. 

THE PLATFORM OVERLOOKING PASARGAD.E 279 

From Dieulafoy, L'Art Antique de la Perse, 1. pi. 3. 

A FIRE-TEMPLE OR A TOMB ? 280 

From a photograph by the author. 

SCULPTURE OF CYRUS 280 

From a photograph by the author. 

I, CYRUS, THE KING, THE ACH^EMENIAN (Inscribed column of Cyrus) 282 
From a photograph by the author. 

THE TOMB OF CYRUS THE GREAT 291 

From a photograph by Antoin Sevruguin. 

TOMBS OF THE ACH^EMENIAN KlNGS AT NAKSH-I RUSTAM . . 295 

From a photograph by Antoin Sevruguin. 

ENTRANCE TO THE TOMB OF DARIUS 298 

From a photograph. 



LIST OF ILLUSTRATIONS xxi 

PACK 

INTERIOR or THE TOMB OF DARIUS 298 

From a photograph by Antoin Sevruguin. 

FIRST SASANIAN SCULPTURE AT NAKSH-I RUSTAM .... 300 

From a photograph by Antoin Sevruguin. 

FOURTH SASANIAN SCULPTURE AT NAKSH-I RUSTAM . . . 300 
From a photograph by Antoin Sevruguin. 

SIXTH AND SEVENTH SASANIAN SCULPTURES AT NAKSH-I RUSTAM 302 
From a photograph by Antoin Sevruguin. 

STONE EDIFICE NEAR THE TOMBS AT NAKSH-I RUSTAM . . . 302 
From a photograph by the author. 

ROCK-HEWN FIRE-ALTARS AT NAKSH-I RUSTAM .... 305 
From a photograph by Antoin Sevruguin. 

FIRST TOMB AT NAKSH-I RUSTAM 305 

From a photograph by Antoin Sevruguin. 

RUINS OF STAKHRA . . 307 

From a photograph by Antoin Sevruguin. 

TAKHT-I TAUS, OR RUSTAM'S THRONE . . . . . . 307 

From a photograph by Antoin Sevruguin. 

SASANIAN SCULPTURE IN THE REAR OF THE RECESS AT NAKSH-I 

RAJAB 309 

From a photograph by Antoin Sevruguin. 

SASANIAN SCULPTURE AT THE LEFT OF THE RECESS AT NAKSH-I 

RAJAB 309 

From a photograph by Antoin Sevruguin. 

THE GRAND STAIRCASE AT PERSEPOLIS 311 

From a photograph by Antoin Sevruguin. 

CURZON'S PLAN OF PERSEPOLIS 312 

From Curzon, Persia, 2. 150. 

PERSEPOLIS . 314 

From a photograph. 

SUBJECT NATIONS BRINGING TRIBUTE TO XERXES .... 314 
From a photograph. 

THE PLATFORM OF PERSEPOLIS 316 

From Dieulafoy, L' 'Art Antique de la Perse, 2. pi. 4. 

THE TOMBS BACK OF THE PLATFORM OF PERSEPOLIS . . . 318 
From photographs by Mr. A. O. Wood. 



LIST OF ILLUSTRATIONS 

PAGE 

THE PALACE OF DARIUS 320 

From a photograph by Mr. A. O. Wood. 

A PORTAL AND THE HALL OF A HUNDRED COLUMNS . . . 320 
From a photograph by Antoin Sevruguin. 

KING DARIUS ON HIS THRONE 323 

From a photograph. 

BRIDGE LEADING INTO SHIRAZ 323 

From a photograph by Rev. C. H. Stileman. 

THE NEW MOSQUE (MASJID-I No) AT SHIRAZ 325 

From a photograph. 

OVERLOOKING SHIRAZ 325 

From a photograph. 

VIEW OF SHIRAZ FROM THE ALLAHU AKBAR GATE .... 327 
From a photograph. 

CULTIVATED FIELDS AROUND SHIRAZ 330 

From a photograph. 

GARDENS AROUND SHIRAZ 330 

From a photograph. 

THE TOMB OF HAFIZ 332 

From a photograph by Antoin Sevruguin. 

THE TOMB OF SAADI 334 

From a photograph by Antoin Sevruguin. 

FROM DEH-BID TO KHAN-I KHORAH 341 

From a photograph. 

OPIUM-SMOKERS 341 

Frorfi a photograph. 

LETTER OF CONVOY FROM THE GOVERNOR OF SHIRAZ . . . 344 
Photographed from the original. 

VIEW OF YEZD 348 

From a photograph by llev. C. H. Stileman. 

A STREET IN YEZD, SHOWING A WIND-TOWER 348 

From a photograph by Rev. C. H. Stileman. 

SCENE IN YEZD 355 

From a photograph by Rev. C. H. Stileman. 

THE RESERVOIR IN THE MEIDAN AT YEZD 355 

From a photograph by Rev. C. H. Stileman. 



LIST OF ILLUSTRATIONS xxiii 

PAGE 

THE ZOROASTRIAN ANJUMAN AT YEZD 360 

From a photograph by the author. 

A WIND-TOWER AT YEZD 366 

From a photograph by Rev. C. H. Stileman. 

Two ZOROASTRIAN PRIESTS AT YEZD 366 

After a photograph by the author. 

SPRAYS OF THE BARSOM PLANT 371 

From a drawing by Mr. Percy Bodenstab- 

YEZD TYPES 374 

From a photograph. 

SCHOOLBOYS AT YEZD, MOSTLY ZOROASTRIANS . . . . . 374 

From a photograph. 

A ZOROASTRIAN CONVERT TO ISLAM 374 

From a photograph. 

Two PICTURES OF KHODABAKHSH BAHRAM RAIS . . . . 382 

From photographs. 

PARSI WOMEN AND GIRLS 382 

From a photograph. 

ZOROASTRIANS OF KERMAN 387 

From Stileman, The Kingdom of the Shah, p. 71. 

A BRANCH OF THE SACRED HOM PLANT 387 

From a photograph. 

A GABAR FAMILY 398 

From a photograph. 

A ZOROASTRIAN MERCHANT 398 

From a photograph. 

SHOEING A DONKEY 403 

Courtesy of the ' Burr Mclntosh Magazine.' 

THE ZOROASTRIAN DAKHMAH AT REI NEAR TEHERAN . . . 403 
From a photograph by the author. 

THE LITTLE SHAGIRD 407 

From a photograph by the author. 

MY DONKEY-CARAVAN IN THE DESERT NEAR KHALATABAD . . 407 
From a photograph by the author. 

DOMED ROOFS OF KASHAN 410 

From a photograph. 



XXIV LIST OF ILLUSTRATIONS 

PAGE 

KASHAN . 410 

From a photograph. 

KUM 412 

From a photograph. 

PERSIAN GLEANERS 412 

From Stileman, The Kingdom of the Shah, p. 11. 

A PERSIAN VENDER OP SHERBET 414 

From a photograph. 

SHRINE OF FATIMA AT KUM 414 

From a photograph. 

A VIEW OF TEHERAN 417 

Courtesy of the 'Burr Mclntosh Magazine.' 

A PERSIAN GARDEN 417 

From a photograph. 

THE MEIDAN-I TOP KHANAH AT TEHERAN 419 

From a photograph. 

THE SOUTHWEST GATE OF THE MEIDAN AT TEHERAN . . . 421 
From a print. 

THE AMERICAN LEGATION AT TEHERAN 421 

From a photograph. 

A GRADUATING CLASS OF THE AMERICAN SCHOOL FOR BOYS AT 

TEHERAN 423 

From a photograph. 

A PERSIAN FAMILY 423 

From a photograph. 

THE PARSI SCHOOL AT TEHERAN 424 

From a photograph. 

ZOROASTRIAN SCHOOLBOYS AT TfiHKRAN 424 

From a photograph. 

ARBAB JAMSHID AND HIS STAFF 426 

From a photograph. 

REI, THE ANCIENT RAGHA 428 

From a photograph. 

MOSQUE OF SHAH ABDUL AZIM, NEAR REI 430 

From a photograph. 



LIST OF ILLUSTRATIONS XXV 

PAGE 

CRUMBLING WALLS OF REI 430 

From a photograph by the author. 

MOUNT DAMAVAND 432 

From a photograph. 

KER PORTER'S PLAN OF REI 435 

From Ker Porter, Travels, 1. 360. 

TOWER ON A HILL NEAR REI 435 

From a photograph. 

TOWER AMONG THE RUINS OF Rfil, AFTER RESTORATION . . 437 

From a photograph. 

TOWER AMONG THE RUINS OF REI, BEFORE RESTORATION . . 437 

From a photograph. 

SASANIAN SCULPTURE NOW DESTROYED 439 

From the drawing of Ouseley, Travels, 3. pi. 65. 

MODERN BAS-RELIEF REPLACING THE SCULPTURE SHOWN ABOVE . 439 
From Flandin and Coste, Voyage en Perse, Moderne, pi. 30. 

SCULPTURED PANEL OF FATH ALI SHAH, ABOVE THE 'SPRING OF 

ALI' 440 

From a photograph by the author. 

THE ZOROASTRIAN TOWER OF SILENCE AT REI .... 440 

From a photograph. 

THE KAZVIN GATE, TEHERAN 442 

From a photograph. 

THE MEIDAN-I MASHK, DRILL SQUARE, AT TEHERAN . . . 442 
From a photograph. 

THE MAHMAN-KHANAH AT KAZVIN 444 

From a photograph. 

THE RIVER KARAJ 444 

From a photograph by the author. 

PEICHANAR, BETWEEN KAZVIN AND RESHT 446 

From a photograph. 

ENZALI ' 446 

From a photograph. 



LIST OF WORKS OF REFERENCE 

This list includes only the works most often referred to. Detailed information 
concerning other books and papers is given in the footnotes. 

Adams, Isaac. Persia by a Persian. (Grand Rapids, Michigan), 1900. 
Barbaro, Josafa. Travels to Tana and Persia by Josafa Barbaro and 

Ambrogio Contarini, tr. William Thomas. London, 1873. (Hakluyt 

Society publications, vol. 49/) 
Barbier de Meynard. See Mas'udi and Yakut. 
Bartholomae, Christian, Altiranisches Worterbuch. Strassburg, 1905. 

[Air. Wb.] 
Bell, John. Travels in Asia. In Pinkerton, General Collection of Voyages 

and Travels, 7. 273-516, London, Igll. 

Benjamin, S. G. W. Persia and the Persians. Boston, (1886). 
Bharucha, Sheriarji Dadabhai. A Brief Sketch of the Zoroastrian Religion 

and Customs. Bombay, 1893. 
Bishop, Mrs. (Isabella L. Bird). Joirrneys in Persia and Kurdistan. 2 vols. 

New York, 1891. 

Browne, Edward G. A Year Amongst the Persians. London, 1893. 
Browne, Edward G. A Literary History of Persia from the Earliest Times 

until Firdawsi. New York, 1902. 
Brugsch, Heinrich. Im Lande der Sonne : Wanderungen in Persien. 2d ed. 

Berlin, 1886. 
Chardin, Sir John. Voyages en Perse et autres lieux de 1' Orient. 4 vols. 

Amsterdam, 1735. 
Clavijo, Ruy Gonzalez de. Narrative of the Embassy to the Court of Timour 

at Samarcand A.D. 1403-6, tr. C. R. Markham. London, 1859. 

(Hakluyt Society publications, vol. 26.) 
Contarini, Ambrogio. Travels to Tana and Persia by Josafa Barbaro and 

Ambrogio Contarini, tr. William Thomas. London, 1873. (Hakluyt 

Society publications, vol. 49.) 
Ctesias. The Fragments of the Persika of Ktesias, ed. John Gilmore. 

London, 1888. 

Curzon, George N. Persia and the Persian Question. 2 vols. London, 1892. 
Darmesteter, James. Le Zend-Avesta, traduction nouvelle. 3 vols. Paris, 

1892-3. ( Annales du Muse'e Guimet, vols. 21, 22, 24.) [Le ZA.] 

xxvi 



LIST OF WORKS OF REFERENCE xxvii 

De Goeje, M. J. Bibliotheca Geographorum Arabicorum. 8 vols. Leiden, 
1870-94. 

della Valle, Pietro. Viaggi di Pietro della Valle il Pellegrino. 2 vols. 
Brighton, 1843. 

della Valle, Pietro. Extracts from the Travels of Pietro della Valle in 
Persia. In Pinkerton, General Collection of Voyages and Travels, 9. 
1-137, London, 1811. 

Dieulafoy, Marcel A. L'Art Antique de la Perse. 5 vols. Paris, 1884-5. 
Ethe", Hermann. Neupersische Litteratur. In Grundriss der Iranischen 

Philologie, 2. 212-368, Strassburg, 1896-1904. 
Firdausi. Firdusii Liber Regum qui inscribitur Shah Name, ed. J. A. 

Vullers (et S. Landauer). 3 vols. Leiden, 1877-84. 
Firdausi. Le Livre des Rois, traduit et commente par Jules Mohl. 7 vols. 

Paris, 1876-8. 
Firdausi. H Libro dei Re, poema epico, recato dal Persiano in versi Ita- 

liani da Italo Pizzi. 8 vols. Turin, 1886-8. 
Flandin, E. N., and Coste, X. P. Voyage en Perse. 8 vols. Paris, (1843-54). 

(Relation du Voyage, 2 vols.; Perse Ancienne, texte, 1 vol., planches, 

4 vols. ; Perse Moderne, 1 vol.) 

Gordon, Sir Thomas Edward. Persia Revisited (1895). New York, 1896. 
Gottheil, R. J. H. References to Zoro'aster in Syriac and Arabic Literature. 

In Classical Studies in Honour of Henry Drisler, pp. 24-51, New York, 

1894. 

Haug, Martin. Essays 04 the Sacred Language, Writings, and Religion of 
the Parsis. 3d ed., edited and enlarged by E. W. West. London, 
1884. 

Horn, Paul. Geschichte Irans in Islamitischer Zeit. In Grundriss der 
Iranischen Philologie, 2. 551-604, Strassburg, 1896-1904. 

Ibn Haukal. The Oriental Geography of Ebn Haukal, tr. Sir William 
Ouseley. London, 1800. 

Jackson, A. V. Williams. Zoroaster, the Prophet of Ancient Iran. New 
York, 1899. 

Justi, Ferdinand. Geschichte Irans von den Altesten Zeiten bis zum Aus- 
gang der Sasaniden. In Gruudriss der Iranischen Philologie, 2. 395- 
550, Strassburg, 1896-1904. 

Justi, Ferdinand. Empire of the Persians. In History of All Nations, 
vol. 2, Philadelphia and New York, 1905. 

Justi, Ferdinand. Iranisches Namenbuch. Marburg, 1895. 

Karaka, Dosabhai Framji. History of the Parsis. 2 vols. London, 1884. 

Karnamak-i Artakhshlr-i Papakan. The original Pahlavi Text, edited (and 
translated) by Darab Dastur Peshotan Sanjana. Bombay, 1896. 



xxviii LIST OF WORKS OF REFERENCE 

Ker Porter, Sir Robert. Travels in Georgia, Persia, Armenia, Ancient 

Babylonia, etc. 2 vols. London, 1821. 

Knanishu, Joseph. About Persia and its People. Rock Island, Illinois, 1899. 
Koran. The Koran, tr. George Sale. London, no date. (Chandos Classics.) 
Landor, A. Henry Savage. Across Coveted Lands. 2 vols. New York, 1903. 
Lynch, H. F. B. Armenia, Travels and Studies. 2 vols. New York, 1901. 
Malcolm, Napier. Five Years in a Persian Town. New York, 1905. ' 
Marquart, J. Eransahr nach der Geographic des Ps. Moses Xorenac'i. 

Berlin, 1901. (Abh. d. Kgl. Ges. d. Wiss. zu Gottingen, Phil.-Hist. 

Klasse, new series, vol. 3, no. 2.) 
Mas'udi, al-. Ma9oudi, Les Prairies d'Or, ed. et tr. C. Barbier de Meynard. 

9 vols. Paris, 1861-77. 
Morgan, J. de. Mission Scientifique en Perse. Vol. 2, Etudes geogra- 

phiques, Paris, 1895. Vol. 4, Recherches archeologiques, Paris, 1896. 
Noldeke, Theodor. See Tabari. 
Odorico da Pordenone. Les Voyages en Asie du bienheureux frere Odoric de 

Pordenone, public par Henri Cordier. Paris, 1891. (Recueil de 

Voyages, vol. 10.) 

Ouseley, Sir William. Travels in Various Countries of the East, more par- 
ticularly Persia. 3 vols. London, 1819-23. 

Perkins, Justin. A Residence of Eight Years in Persia. Andover, Mass., 
1843. 

Perrot, G., and Chipiez, C. Histoire de 1'Art dans 1'Antiquite. Vol. 5, Perse, 
etc., Paris, 1890. 

Pietro della Valle. See della Valle. 

Polo, Marco. The Book of Ser Marco Polo the Venetian concerning the 
Kingdoms and Marvels of the East, tr. and ed. Sir Henry Yule. 3d ed., 
revised by Henri Cordier. 2 vols. London, 1903. 

Rawlinson, George. The Five Great Monarchies of the Ancient Eastern 

World. 4 vols. London, 1862-7. 

Rawlinson, George. The Sixth Great Oriental Monarchy. London, 1873. 
Rawlinson, George. The Seventh Great Oriental Monarchy. London, 1876. 
Rosenberg, Fr6de"ric. Le Livre de Zoroastre (Zaratusht Nama) de Zar- 

tusht-i Bahram ben Pajdu. St. Petersburg, 1904. 
Sadik Isfahanl. The Geographical Works of Sadik Isfahani, tr. by J. C. 

from original Persian Mas. in the collection of Sir William Ouseley, 

the editor. London, 1832. 
Schwarz, Paul. Iran im Mittelalter nach den Arabischen Geographen, 1. 

Leipzig, 1896. 

Shah Namah. See Firdausl. 



LIST OF WORKS OF REFERENCE xxix 

Shatroiha-i Airan. In Pahlavi Texts, 1, edited by Jamaspji Dastur 
Minocheherji Jamasp Asana, pp. 18-24, Bombay, 1897. 

Shatroiha-i Airan. In Aiyadgar-i-Zariran, Shatroiha-i-Airan, etc., translated 

with notes by Jivanji Jamshedji Modi, pp. 50-180, Bombay, 1899. 
Spiegel, Friedrich. Eranische Alterthumskunde. 3 vols. Leipzig, 1871-8. 
Spiegel, Friedrich. Die Altpersischen Keilinschriften. 2d ed. Leipzig, 

1881. 

Stolze, F., and Andreas, F. C. Persepolis. 2 vols. Berlin, 1882. 
Sykes, Percy M. Ten Thousand Miles in Persia, or Eight Years in Iran. 

New York, 1902. 
Tabari, al-. Geschichte der Perser und Araber zur Zeit der Sasaniden, aus 

der Arabischen Chronik des Tabari, von Theodor Nb'ldeke. Leiden, 

1879. 
Tavernier, Jean Baptiste. Six Travels through Turkey and Persia to the 

Indies, tr. J. Philips. London, 1684. 
Texier, C. F. M. Description de PArmenie, la Perse, et la Mesopotamie. 

2 vols. Paris, 1842-52. 
Tha'alibi, al- . Histoire des Rois des Perses. Texte Arabe public' et traduit 

par H. Zotenberg. Paris, 1900. 
Tomaschek, Wilhelm. Zur Historischen Topographic von Persien. In Sb. 

d. Phil.-Hist. Classe d. kais. Akad. d. Wiss. zu Wien, 102 (1883), 

pp. 145-231. 
Weeks, Edwin Lord. From the Black Sea through Persia and India. New 

York, 1896. 
Weissbach, F. H., and Bang, W. Die Altpersischen Keilinschriften. 

1. Lieferung. Leipzig, 1893. 

Weissbach, F. H. Die Achamenideninschriften Zweiter Art. Leipzig, 1890. 
West, E. W. Pahlavi Literature. In Grundriss der Iranischen Philologie, 

2. 75-129, Strassburg, 1896-1904. 

West, E. W. Pahlavi Texts. 5 vols. Oxford, 1880-97. (Sacked Books 

of the East, vols. 5, 18, 24, 37, 47.) 
Wilson, Charles. Handbook for Travellers in Asia Minor, Transcaucasia, 

Persia, etc. London, 1895. 

Wilson, S. G. Persian Life and Customs. New York, 1895. 
Yakut. Geographisches Worterbuch, ed. Wiistenfeld. Leipzig, 1866. 
Yakut. Dictionnaire geographique, historique, et litteraire de la Perse, 

extrait du Mddjem el-Bouldan de Yaqout, par C. Barbier de Meynard. 

Paris, 1861. 
Zotenberg, H. See Tha'alibi. 



ABBREVIATIONS 

Abh. = Abhandlung. 

A.H. ' = (Anno Hegirae), Mohammedan era. 

AJP. = American Journal of Philology. 

Artax. Pers. = inscriptions of Artaxerxes at Persepolis. 

Av. = Avestan. 

Bd. = Bundahishn. 

Bh. = Behistan inscription of Darius. 

BYt. = Pahlavi Bahman Yasht. 

c. = (circa), about. 

d. = died. 

Dar. Alv. = inscription of Darius on Mt. Alvand (Elvend),near Hamadan. 

Dar. Pers. = inscriptions of Darius at Persepolis. 

ed. = edition of, edited by. 

Gk. = Greek. 

ibid. = (ibidem), in the same work. 

id. = (idem), the same author. 

IF. = Indogermanische Forschungen. 

JAOS. = Journal of the American Oriental Society. 

JRAS. = Journal of the Royal Asiatic Society. 

JRGS. = Journal of the Royal Geographical Society. 

KZ. = Kuhn's Zeitschrift fur Vergleichende Sprachforschung. 

I.e. = (loco citato), at the place previously cited. 

MKh. = Dlna-I Mamog-I Khirad. 

Mod. Pers. = Modern Persian. 

Nir. = Nirangistan. 

NR. = inscriptions of Darius at Naksh-i Rustam. 

Ny. = Nyaish. 

OP. = Old Persian. 

op. cit. = (opus citatum), the work previously cited. 

Pers. = Persian. 

Phi. = Pahlavi. 

pi. = plate. 



ABBREVIATIONS XXxi 

Sb. = Sitzungsberichte. 

SEE. = Sacred Books of the East, 

seq. = (sequentia) , and the following. 

Sir. = Sirozah. 

s.v. = (sub verl)o) y under the word, 

r. = translation of, translated by. 

Vd. = Vendldad. 

v.l. = (varia lectio), variant reading. 

Vsp. = Visperad. 

WZKM. = Wiener Zeitschrift fur die Kunde des Morgenlandes. 

Xerx. Pers. = inscriptions of Xerxes at Persepolis. 

Ys. = Yasna. 

Yt. = Yasht. 

ZDMG. = Zeitschrift der Deutschen Morgenlandischen Gesellschaft. 

Zsp. = Zatsparam. 

Zt. = Zeitschrift. 

Zt. f. Assyr. = Zeitschrift fur Assyriologie. 



OF THF 

UNIVERSITY 

OF 




PERSIA PAST AND PRESENT 

CHAPTER I 

EN ROUTE FOR THE LAND OF THE LION AND THE SUN 

'I am bound 
To Persia, and want guilders for my voyage.' 

SHAKSPERE, A Comedy of Errors, 4. 1. 3. 

IT was at the end of January, 1903, that I received leave 
of absence from Columbia University for half a year to enable 
me to visit the Orient again. The previous voyage had been 
to India and Ceylon, two years before ; my present goal was 
Persia and Central Asia. The purpose of my journey was 
antiquarian study and scholarly research, especially with 
regard to Zoroaster and the ancient faith of the Magi, for 
I had come early under the spell of those Wise Men from 
the East and had long felt the charm drawing me toward the 
Province of the Sun ; but I hoped also to contribute something 
to our knowledge of Persia's present, as well as past, and to 
a better understanding of the relations existing between them. 

My plan was to traverse as much of the territory known 
to Zoroaster as I could, including Transcaspia and Turkistan, 
and to visit the places most celebrated in the history of Persia. 
The route which I marked in advance on the map, and was 
able to accomplish, carried me from the Caucasus on the north 
nearly to the Persian Gulf on the south, thence to Yezd in 
the central desert and back northward to Teheran and the 
Caspian Sea. Crossing this, I continued the journey into the 
heart of Asia, to Merv, Bokhara, and Samarkand. 



2 EN ROUTE FOR THE LAND OF THE LION AND THE SUN 

It is a far cry from New York to the Caspian more than 
seven thousand miles but this distance had first to be covered 
before the real journey began. The first four thousand miles of 
the trip by steamer and rail were uneventful. At Berlin I stayed 
a few days to see my teacher, Professor Karl F. Geldner, whose 
writings first inspired me with an interest in Persia, and to 
make some visits among old friends, as well as calls upon 
officials, before starting on my forty-eight-hour journey to St. 
Petersburg, where I found my friend and former pupil Mr. 
Montgomery Schuyler, Jr., then a secretary at the American 
Embassy. I had brought with me official letters of recom- 
mendation from Washington to St. Petersburg, as well as to 
Teheran and Tabriz, which proved an open sesame to the 
Ministers of State, and I shall not forget the kindness of 
the various official representatives, Russian and American, at 
the capital of the Tsar. Besides this I met with courtesies 
from scholar friends both there and at Moscow ; they had not 
only been in Persia, but were well acquainted with Transcaspia 
and Turkistan and could give valuable suggestions for the 
journey. 

At Moscow, rich in historic memories, I had an excellent 
opportunity to inspect the interesting Verestchagin collection 
of paintings in the Tretiakoff Gallery. This collection is 
particularly rich in paintings which illustrate Central Asian 
life, and they gave me a glimpse in advance of the scenes 
which I should see at Merv, Bokhara, and Samarkand. After 
a day's stay I took the weekly train de luxe bound southward 
toward the Caucasus, the Caspian, and the highway to Persia. 

For three full days the train rolled on through the steppe- 
country, level, and uninteresting when covered with snow. 
The stops were infrequent but long, and I welcomed each time 
the third of the set of bell signals which the Russian railways 
employ at their stations, because it was a sign for the journey 
to be resumed toward the goal I had in view. 

At Vladikavkas, on the morning of the fourth day it was 



THROUGH RUSSIA TO THE CASPIAN 3 

Thursday, I remember I caught my first sight of the giant 
barrier of the Caucasus towering against a cloudy sky and 
frowning down on the white plain beneath. Its beetling 
cliffs were bare of snow in places, and here and there a deep 
gorge or ravine looked like the scar of some Titan wound upon 
its sullen face. The lonely scene grew in impressive grandeur 
as the day wore on. The old myth of Prometheus rose before 
my imagination. Far in the distance I could picture the 
desolate vulture-peak where the demigod lay chained in fetters 
because he had stolen fire from heaven as a boon for men. 
Well might the suffering benefactor of mankind have longed 
for the sun to rise and 'dispel the hoar frost at dawn,' or, 
when scorched by the heat and torn by the ravening bird, 
have yearned for * starry-kirtled night to hide day's sheen.' 1 
I could hear faint echoes of the dialogue with lo and mutter- 
ings of the Titan's curse against the wrath of Zeus. Little did 
I dream when I read Prometheus in college days that I should 
ever see the place where ^Eschylus had laid his tragic scene. 

The streams rushing from the mountains and the flocks of 
sheep huddled together in the open places of the snow re- 
called to my mind the story of Colchis and the Golden Fleece. 
I learned en route that tradition tells how the shepherds of 
by-gone days were wont to find grains of gold clinging to the 
new-shorn fleece when they lifted it from the stream where it 
was washed, because the mountain torrent had left a golden 
deposit amid the woolly strands. The legend of the rich reward 
seems not to be quite forgotten. 

For a moment, Greek mythology, classic reminiscences, and 
thoughts of college days made me forget that the land of 
my quest was Iran, not Hellas, and that I was seemingly 
deserting the Orient for Greece. I had to recall myself once 
more to the East. 

All day the railway skirted the great plain beneath the 
Caucasus, which was never more than twenty miles distant. 
1 JEschylus, Prometheus Bound, 20 seq. 



4 EN ROUTE FOR THE LAND OF THE LION AND THE SUN 

The scenery at this time of the year was barren and dreary. 
Hardly a trace of vegetation was visible except where the wind 
had blown a space bare in the snow and revealed a possible 
promise of verdure when spring should come. Flocks of sheep 
and goats were to be seen wherever a bit of fodder could be 
found, and scant herds of rugged cattle lounged disconsolately 
about. 

I was interested in watching the changes in the types of the 
people as the journey progressed. Some of the natives repre- 
sented to perfection the type of the Scythian shepherd in 
antiquity ; they wore huge sheepskin coats and had their feet 
wrapped in coarse bagging which was lashed about the legs 
with thongs ; their heads were covered with a cap of heavy 
fur which was almost indistinguishable from the shock of hair 
and heavy beard. A few looked a little more modern because 
of the long rifle with which they stood guard over their flocks. 
Most of them had the shambling gait of the East and the Ori- 
ental fashion of squatting, which was particularly noticeable 
around the railway stations. All of them had dark complex- 
ions which looked weather-beaten and coarse. The Iranian 
type of features grew more and more pronounced as the Cas- 
pian was approached, and I could recognize distinct likenesses 
to the Pathans and Waziris, those Afghan tribesmen of Iranian 
blood, whom I had seen in the Khaiber Pass two years before. 
It is clear that Iran begins ethnologically with the Caucasus 
and the Caspian, the historic borderland between Europe and 
Asia, although the Russian frontier line to-day has encroached 
a hundred or two miles over the old Persian border. 

Darkness had fallen when the train arrived at Petrorvsk, but 
through the gloom I could catch sight of the white waves of 
the Caspian lashed into foam by the wintry winds. Before 
daylight the next morning, Friday, March 6, I had reached 
Baku. Instead of proceeding thence across the Caspian to 
Teheran, which would have been more convenient in many 
ways, I decided to continue the journey to Tiflis in order to 




THE OLD CITADEL OF TIFLIS 




FROM BAKU TO TIFLIS 5 

enter Persia through Azarbaijan, the region which gave birth 
to Zoroaster, as I believe. 

The journey from Baku to Tiflis by rail occupies about four- 
teen hours. A great part of the route traverses the southern 
side of the Caucasus range along whose northern base we 
coasted the day before. The scenery is less magnificent, but 
the plain is rich, and the mountain range on the south running 
parallel with its sister chain on the north, from the Caspian to 
the Black Sea, looks higher because of its proximity to the 
railway. The cattle visible on the plain were of the same 
general character as the day before, but troop after troop of 
camels showed that the East had been reached. Some of these 
had rough blankets thrown over their backs to protect them 
from the rigors of the weather, as the climate was a contrast to 
that of the desert where they had been bred. The Persian 
buffalo could be seen here and there, and all were busily en- 
gaged in nibbling the few traces of prospective grass that 
could be found in the plain. 

There is little of interest to note on the way from Baku to 
Tiflis, but about halfway between the two cities the railroad 
passes the town of Elizabetpol, which was originally the Per- 
sian town of Ganjah, its name having been changed when the 
Russians took possession of the place in 1804. Its mosque 
near the bazaar is said to have been erected by the Persian 
monarch Shah Abbas in the seventeenth century, and numer- 
ous ancient remains in the neighborhood show the antiquity of 
the city. Its chief claim upon our interest perhaps is the fact 
that Ganjah was the home of the Persian poet Nizami, who died 
there about the year A.D. 1203. This romantic minstrel sang 
of the love of Khosru Parviz for the Armenian princess Shirin 
and the tragic passion of her artist lover Farhad, besides tell- 
ing the legendary adventures of Alexander in epic strain and 
recounting old tales with imaginative grace. 1 

That whole afternoon of my journey the sky had for me a 
1 See p. 226, below. 



6 EN ROUTE FOR THE LAND OF THE LION AND THE SUN 

poetic coloring ; the sunshine was tempered by clouds which 
imparted a tinge of melancholy ; there was a faint suggestion 
of spring in the air ; but dusk and its accompanying chill came 
early, and evening had already closed in before Tiflis was 
reached. The night turned out clear and cold. The lamps 
of the terraced town shone bright like a myriad lights in the 
blue, and gave a fairy-like effect to the scene. Much of this 
impression was dispelled by day, but this was not due to lack 
of good accommodations at the Hotel de Londres, which is 
one of the last places for bidding adieu to Western comforts 
before entering Persia. 



OF THE 

UNIVERSITY 

OF 

C*i IFOR!^ 




THE MICHAELIS BRIDGE OVER THE KURA AT TIFLIS 




THE CAUCASIAN MUSEUM, TIFLIS 



CHAPTER II 
TIFLIS, THE CAPITAL OF TRANSCAUCASIA 

* Parthians, and Medes, and Elamites, and the dwellers in Mesopotamia, and 
in Judaea, and Cappadocia, in Poutus, and Asia.' 

NEW TESTAMENT, Acts 2. 9. 

TIFLIS is a combination of Orient and Occident. It is one 
of those cities in which Western civilization has been welded 
on to Eastern custom, but the signs of the joining will never 
disappear. Its languages are as many as those at the gift of 
tongues at Pentecost, and its types are as multifarious as a 
union of ancient and modern life can bring together. Along 
its crowded thoroughfares the sheepskin-clad dweller of the 
Caucasus rubs elbows with the Armenian, Georgian, Persian, 
Kurd, Turk, and Tartar, or moves side by side with the Euro- 
pean dressed in broadcloth. The ever shifting groups and 
constantly changing colors rival a kaleidoscope in variety. 
The winding alleys of the native quarters, the mazes of the 
bazaars, and the crowded passages between the booths are 
quite Oriental, but the European sections of the town, with 
broad streets, long avenues, and large squares, are Occidental 
and show the evidences of Russian advance. 

Being the capital of Transcaspia, Tiflis is the head of the 
civil and military authority, as well as the seat of government. 
Its growth commercially has been remarkable in recent years ; 
its busy heart throbs with the double pulse of East and West, 
and its claim to one hundred and sixty thousand inhabitants 
seems not at all exaggerated. 

Historically Tiflis is a place of interest also, as it was the 
capital of ancient Georgia. It is said to have owed much of 

7 



8 TIFLIS, THE CAPITAL OF TRANSCAUCASIA 

its original renown to the Georgian emperor Vakhtang Gurgas- 
lan, in the fifth century of our era, who was attracted to the 
place by the health-giving properties of its sulphur baths. The 
qualities of these hot springs are especially mentioned by the 
Arab traveller Ibn Haukal, in the tenth century, who states 
that ' the water is warm without fire,' and adds, ' Tiflis is 
a pleasant place and abounds in provisions ; it has two walls 
of clay and produces much fruit, and agriculture is practised 
in its territories.' 1 In the latter part of the fourteenth cen- 
tury it suffered, in common with most cities of Asia, the mis- 
fortune of being plundered by Timur Lang, or Tamerlane. Its 
general development as a metropolis, though menaced at times 
by the Turks, went on under Persian rule until 1801, when the 
Russians took possession of Tiflis, and it has remained ever 
since under the sway of Russian authority, although riots and 
uprisings, due to the unruly character of the mixed population 
of the Caucasus, have occurred from time to time. 

The town is situated partly upon a hill, the site of the 
ancient citadel, and approaches another elevation to the north 
and east, so that the city has a somewhat terraced effect. 
Through the middle of it, running from northwest to south- 
east and roughly dividing it in half, flows the river Kura, the 
Cyrus of the ancients. On either side of this rapid stream, 
and forming the southern part of the town, are the native 
quarters of Tiflis, the Georgian section (Avlabar) being on 
the east, or left, bank, the bazaars on the west, or right, bank. 
Adjoining the bazaars as we look northward, and still on the 
same side of the river, is the Russian quarter, with its fine 
edifices, broad avenues, and imposing Alexander Garden as a 
centre. Crossing the river from this point by one of its 
several bridges, we enter the German district, which extends 
northward from the Georgian quarter and owes its name to 
a colony of sturdy dissenters from Wiirtemberg, who were 
among a number that left the Fatherland early in the last 
1 Ibn Haukal, tr. Ouseley, p. 160. 



THE CAUCASIAN MUSEUM 9 

century because of certain religious differences of opinion and 
made Tiflis their adopted home. 

In the Russian quarter are stately buildings devoted to 
administration, the post, and banking, together with churches, 
theatres, clubs, shops, hotels, residences official and private, 
parks, and gardens, all of which show the introduction of 
Western ideas. One of the most interesting edifices is the 
Caucasian Museum with its rich collection of illustrative mate- 
rial relating to the region lying between the Caspian and the 
Black Sea. Here the student will find a storehouse of anti- 
quarian wealth amassed with care and judgment by Dr. 
Gustav Radde, who devoted years of his life to the cause. 
When I arrived at Tiflis, I learned that this enthusiastic and 
scholarly collector was seriously ill with what eventually 
proved to be his last sickness. Despite his feeble condition he 
insisted upon my coming to call at his bedside and sent a 
special guide to conduct me around the museum. The kindly 
greeting which he gave me and the gentle farewell that 
followed I shall always remember. 

The museum well repays a careful visit. As a special col- 
lection to illustrate the natural history, flora and fauna, eth- 
nology, and archaeology of the Caucasus region a region 
particularly interesting as being the bridge between Asia and 
Europe it is unmatched. Two exhibits of aquatic life and 
land animals around the Caspian Sea particularly interested 
me because of their being referred to in the Avesta, or Zoroas- 
trian Bible. One of these was a fine specimen of the giant 
sturgeon, known as the accipenser huso, a fish fifteen feet or 
more long, and mentioned in the Avesta, as I believe, under 
the name of Jcara masya, or fozr-fish. 1 The other was a group 
of wild boars, admirably mounted by the taxidermist so as to 
display all the fierceness and combativeness of the vardza, or 
wild boar, described in the Zoroastrian texts. 2 

To us of the West the chief attraction of Tiflis lies not in its 
1 See Vd. 19. 42 ; Yt. 14. 29 ; Pahlavi Vsp. 1. 1 ; Bd. 18. 3 ; etc. 2 See Yt. 14. 15 seq. 



10 TIFLIS, THE CAPITAL OF TRANSCAUCASIA 

European features, but in its Oriental side and the remains 
which it shows of an older civilization. A survival of Oriental 
mediaevalism is seen in the fortress which crowns the height 
overlooking the city and commands the town with its old-time 
battlements. Still older are the remains of a tower and ruined 
aqueduct which overlook the Botanical Garden to the south of 
Tiflis. The bridges that cross the river Kura (Cyrus) and 
connect the two halves of the city are partly old and partly 
new. The most interesting, perhaps, because most crowded, is 
the bridge of the Tartar Meidan, which leads to one of the 
sections of the native bazaars. These bazaars are not so Oriental 
as the Persian bazaars, but their crowded booths, the variety of 
wares displayed, and the bargains they offer in Daghistan rugs 
and Caucasian armor, afford an attractive place of visit for 
those who have not travelled in the East before. 

During my stay of three or four days at Tiflis I gathered 
some additional information regarding the Yezidis, or Devil- 
Worshippers, a people to whom my attention had previously 
been drawn in connection with my studies about Zoroaster and 
the religion of ancient Persia. A Swedish missionary, the Rev. 
E. John Larson and his wife, who have done much evangelistic 
work among the Yezidis of Tiflis and the vicinity and have 
thus become familiar with their manners, customs, and beliefs, 
were my informants. 

The Yezidis are chiefly found in the Caucasus, Armenia, 
and Kurdistan, although they are scattered over a considerably 
wider territory, their headquarters being in the province of 
Mosul, Mesopotamia. Owing to the persecutions which they 
have suffered throughout their history, their number is not large; 
nevertheless they are said to number twelve thousand in the 
region of the Caucasus alone, and there are at least several 
hundred Yezidis living in the immediate vicinity of Tiflis. 
They do not speak of themselves ordinarily as Yezidis, but em- 
ploy the names of their respective tribes or adopt by preference 
the term Dasni, a tribal designation in the neighborhood of 




RUINED TOWER AND AQUEDUCT OVERLOOKING THE BOTANICAL GARDEN 




THE TARTAR MEIDAN AT TIFLIS 



THE YEZIDIS, OR DEVIL-WORSHIPPERS 11 

Mosul, close to the site of ancient Nineveh, which was one of 
the original homes of the religion. Various explanations have 
been proposed for the name Yezidi, among them one which 
associates it with Yazddn, the Persian word for God, as the 
Yezidis undeniably believe in a god, although they do not ordi- 
narily speak of him. A second suggestion is to connect it with 
the town of Yezd. A third seeks to derive the name from 
Yezid, the detested Mussulman Kaliph who slew Husein, the 
grandson of Mohammed, for Yezid is fabled to have been a cham- 
pion of their faith. But none of these suggestions seem very 
satisfactory. 1 

According to the belief of the Yezidis, God, the creator of 
heaven and earth, first made from his own essence six other 
divinities, the sun, the moon, and the principal stars, and these 
joined with him in creating the angels. The devil, who was 
God's own creation, rebelled against his lord and was cast into 
hell. He afterward repented of his sin, did penance for seven 
thousand years, and shed tears of contrition which fill seven 
vessels that will be used at the Day of Judgment to quench 
the fires of the seven hells. God in his mercy pardoned the 
recreant, restored him to heavenly rank, made him one with 
himself, and forbade the angels to look with scorn upon their 
reinstated brother. Inasmuch as God's grace thus forgave 
and exalted even Satan himself, man should not look with 
hatred upon this so-called representative of evil. On this 
account the Yezidis never allow the name of Satan to pass their 
lips, avoiding even a syllable that suggests the word, and 
shrinking with horror from any mention of the devil by others. 
They venerate his sacred majesty under the name of Malik 
Td'us, 4 King Peacock,' a title which they apply to the holy 

!For further details on this and letin Soc. NeuchateL Geog. 12. 275 

the general subject of the Yezidis, see seq.) ; Adams, Persia by a Persian, 

my article 'Yezidi' in the New In- pp. 497-509, Grand Rapids, Mich., 1900 ; 

ternational Encyclopedia, 17. 939 ; Layard, Nineveh and its Remains, part 

my note in JAOS. 25. 178-181 ; Spiro, 1, ch. 9, pp. 270-325, London, 1854 ; id. 

Les Yezidi, Neuchatel, 1900 (in Bui- Nineveh and Babylon, London, 1853. 



12 TIFL1S, THE CAPITAL OF TRANSCAUCASIA 

standard (sanjak) or symbol of their religion, which is a peacock, 
conventionalized in their art so as almost to resemble a cock. 
Malik Taus revealed himself in the form of a handsome youth 
with a peacock's tail when he appeared in a vision before Sheikh 
Aadi, the prophet of the faith. 

I have sometimes thought that this reverence shown for the 
power of evil may be similar in character to the propitiatory 
sacrifice offered in ancient times to the divinity beneath the 
earth by Amestris, the wife of Xerxes, according to Herodotus 
and other authorities. 1 It seems possible also that the daeva- 
yasna, or 'Devil-Worshippers,' anathematized in the Avesta 
may have entertained kindred ideas about venerating the 
realm of darkness, and that the Yezidis and their strange 
beliefs preserve traces of the devil-worship in Mazandaran 
which Zoroaster so bitterly denounced. 

It is clear that the Yezidi faith shows some ancient Iranian 
traits, such as certain marked dualistic features, a reverence 
for the elements, fire, water, and earth, and a belief in a father 
primeval, who lived before Adam and did not fall in sin. This 
latter belief appears to be a reminiscence of the Avestan Gaya- 
maretan, who lived before Mashya and Mashyoi, the Iranian 
Adam and Eve. The Yezidis likewise refrain from spitting 
upon the ground an observance as old as the Magi 2 nor 
will they pour boiling water upon the earth for fear of scald- 
ing the face of the little devils. I have often been told that 
if a circle be drawn on the ground around a Yezidi he will 
stand for hours in the middle of it without venturing to step 
over the charmed line, which reminds one of the karshvars 
drawn in Avestan rites of exorcism according to the Vendidad. 
It is thought, moreover, that the Yezidi religion shows distant 
survivals of the old Assyro-Babylonian worship of the sun, 
moon, and stars, for the faith appears to have retained the 
sun-god Shamash under the form of Sheikh Shems and the 

1 Herodotus, Hist. 7. 114 ; cf. 3. 35. paedia, 8. 1. 42. See also Adams, 

2 Ibid. 1. 99, 138 ; Xenopbon, Cyro- Persia, pp. 497, 499. 




YEZIDIS, OR SO-CALLED DEVIL-WORSHIPPERS, IN TIFLIS 



BELIEFS OF THE DEVIL-WORSHIPPERS 13 

moon-god Sin as Sheikh Sinn, an emanation of God him- 
self. 1 

In many respects the Yezidi doctrines have been influenced 
by Manichseism, and its doctrines of purity, by Nestorian 
Christianity and especially by Mohammedanism. With each 
of these religions the Yezidis have come into contact. They 
recognize Mohammed as a prophet of equal rank with Abraham 
and the patriarchs, and they believe that Christ was an angel 
in human form. One curious statement that I heard is that 
the Yezidis sacrifice one sheep every year to Christ and thirty 
to the devil. The rite of baptism is practised among them, 
and circumcision is general, but not universal. A part of their 
marriage ceremony consists in the bridegroom and bride divid- 
ing a piece of bread between them, but the Yezidis also allow 
Mohammedan priests formally to officiate at their weddings 
and even at their funerals. They recognize, however, a regu- 
lar system of priesthood of their own, headed by a Myr, or 
high priest, together with various ecclesiastics and clerical 
functionaries, and including an order of mendicant devotees, 
male and female, fakir and faklriah. The Yezidis have no sac- 
rifices or temples in the true sense of the term, but they have 
a number of religious ceremonies and observe several fasts and 
festivals, most important among the latter being the autumnal 
worship of the effigy of Malik Taus, an occasion which is 
accompanied by offerings and prayers. Among their peculiar 
superstitions is a curious abhorrence for the color blue. 

In their daily life the Yezidis are not forbidden the use of 
wine, as the Mohammedans are, but they do not indulge in 
it to excess. Nor is polygamy prohibited, although it is ap- 
parently not much practised on account of the poverty of the 
people. The belief in a future life, and a system of rewards 
and punishments hereafter, forms part of their faith. Among 
these punishments is the condemnation to assume in another 

existence the form of some animal, for the doctrine of trans- 
1 Spiro, Les Yezidi, pp. 20, 26. 



14 TIFLIS, THE CAPITAL OF TRANSCAUCASIA 

migration is a recognized one. 1 The Yezidis have a book of 
divine revelation, which they call Al-Yalvah, and they name 
as its great expounder their sainted head Sheikh Aadi, who 
lived about A.D. 1200. 2 

As to their social status I was informed that the Yezidis 
around Tiflis and Erivan occupy an inferior position, and I 
understand that the same is true elsewhere. Their occupations 
at Tiflis are largely menial, as they are employed chiefly in 
drudgery work and as scavengers. For that reason they go 
clad in the meanest rags. The stories which were told me 
about them reminded me somewhat of the 'sweeper class' in 
India ; but recent Russian municipal ordinances at Tiflis have 
partly transferred the duties of the Yezidis to other hands, so 
that their occupations have varied considerably within the past 
year or two. In general they are accustomed to live outside 
the town and come into the city for work during the day. The 
wife carries on the household duties and does the agricultural 
work connected with the soil. The wants of the Yezidi and 
his family are meagre, and they appear to lead contented lives. 
Around Tiflis, moreover, there are said to be a considerable 
number of Yezidis who, despite their impoverished appearance, 
possess considerable money. 

Most of the last day in Tiflis was devoted to making final 
preparations for my journey and to completing the necessary 
outfit, for Persia is still without railroads and hotels, except at 
Teheran, and travelling through the country is synonymous 
with roughing it. 

For caravan transport the 4 kit ' should be as light and com- 
pact as possible, but a folding cot, serving alike as bed, desk, 
and dining-table, is indispensable. A European saddle and 
bridle, boots or riding-leggings, will also be found to be a sine 



1 For general details see my note, 2 So my informant ; but Adams, 

JAOS.25. 181 ; Adams, Persia, pp.499, Persia by a Persian, p. 501, says 

605-506; SpiTo,Les Yezidi, pp. 14 (286), 'about the middle of the tenth cen- 

16 (288), 29 (301). tury.' 




TYPES OF YEZIDIS, OR DEVIL-WORSHIPPERS, AT TIFLIS 




TIFLIS CITADEL (ANOTHER VIEW) 



MAKING UP A KIT FOR THE JOURNEY 15 

qua non. As for clothing, it is well to remember that the cold 
in the mountains is intense in winter and the heat of the 
deserts is equally extreme in summer. In either case dark 
smoked glasses will be found to be a great protection against 
the snow, sun, or dust. Besides the necessary cooking utensils 
and travelling articles, I took also an extra supply of straps, 
padlocks, two cameras, some books, including a map, several 
knick-knacks for gifts, and sundry other things which one is 
likely to need on a long journey. For packing this outfit my 
experience showed that two stout leather suit-cases and a 
portmanteau were sufficient, if supplemented by some native 
pouches. It is important above all in caravaning to have 
several small bundles, rather than one or two large ones, 
for the parcels can thus be more easily distributed so as to 
balance equally on the packhorse. 

At the hotel I engaged the services of a Georgian guide, old 
Rustom, to conduct me to the frontier ; his name pleased me 
because of its historic associations with Rustam, the Persian 
hero. As a guide he was well informed and reliable, and 
well equipped linguistically, since he spoke no less than six 
languages. French was our medium of communication. 

The luggage, including camp-bed and saddle, was dexter- 
ously packed, the final arrangements were all completed, and 
before many hours I was on the train, started for a trip of 
fifteen more hours to Erivan, where the railroad now ends 
beneath the shadow of Mount Ararat. 



CHAPTER III 

ERIVAN, MOUNT ARARAT, AND THE ROAD TO THE 
PERSIAN BORDER 

4 The ark rested upon the mountains of Ararat.' 

OLD TESTAMENT, Genesis 8. 4. 

TOWARD sunset on the following day, as the train ap- 
proached the ancient town of Erivan, I caught my first view 
of Mount Ararat, crested with clouds and wrapped in snow 
over which the sun shed a soft roseate hue. Rising solitary 
from the plain, not backed by ranges of lesser hills, Ararat 
possesses a lonely grandeur which makes it a fitting place for 
the ark to have rested upon at the solemn death-hour of an 
older race and the birth of a new generation. 

This stupendous mountain, which comprises two contiguous 
bases, lifts its huge mass nearly seventeen thousand feet above 
sea-level, or fourteen thousand above the surrounding plain. 
The summit of the larger peak, known as Great Ararat, is 
16,916 feet high, and is crowned with eternal snow ; the crest 
of Little Ararat, seven miles distant, is 12,840 feet, and it is 
hardly less impressive. Both peaks owe their origin to forces 
of a volcanic nature, and a great chasm, thousands of feet deep, 
torn in the northwestern side of Great Ararat, lays bare the 
dead giant's heart. 1 

The Avesta mentions Ararat, it is thought, under the name 
of Mazishvant (Yt. 19. 2), a term which recalls Masis, the 
designation given it by the Armenians, who believe it to be 
the mountain on which the ark rested, having taken over the 
tradition of the deluge with their acceptance of Christianity. 

1 For an excellent description of to its top, see Lynch, Armenia, 1. 
Ararat and an account of an ascent 142-199, London, 1901. 

16 




MOUNT ARARAT, NEAR ERIVAN 




RUSTOM, MY GEORGIAN GUIDE TO THE FRONTIER 



MOUNT ARARAT 17 

The Persian name for it is Koh-i Nuh, l Noah's Mountain,' 1 
while the Tartars merely call it Aghri Dagh, 'Steep Moun- 
tain,' the name given to the range running eastward. The 
old superstition current among the natives, to the effect that 
its summit cannot be reached by man, has long been dispelled, 
since no less than sixteen different ascents by Europeans have 
been recorded within the past hundred years. 

Legends about Noah naturally cluster around Ararat and its 
vicinity. The place where the patriarch planted the vine and 
partook to excess of the juice of the grape was formerly shown 
near the village of Akhuri (Akori), or Arguri a hamlet 
whose name, by popular etymology in Armenian, is supposed 
to mean 4 he (Noah) planted the vine' (ark ur)? whence the 
modern form Arguri. In like manner the spot where he built 
an altar and offered burnt sacrifices to Jehovah used to be 
pointed out, as well as a stunted willow sprung from a plank 
of the ark. 3 These relics were swept away by a terrific earth- 
quake which took place in the chasm near Akori, July 2, 1840, 
overwhelming the village and destroying its inhabitants. 
From the descriptions by those who survived the cataclysm it 
must have resembled the terrors of the Day of Judgment. 

For two nights I slept a frozen sleep at Erivan beneath the 
heights of Ararat. The so-called 'hotel' at which I lodged 
was one merely in name and sign, and I almost perished with 
the cold. I look back with a shiver at the experience in those 
uncomfortable quarters ; but later, when I became acquainted 
with exposure in Persian mud hovels, I thought that the 
hostelry may not actually have been so bad as it seemed. 

Erivan is the capital of Russian Armenia and a place of 
nearly thirty thousand inhabitants. Its history is obscure, but 
local tradition is naturally not loath to carry the origin of the 

1 The ark is alluded to in the Koran, Ortsnamen, in Indogermanische For- 
ch. 29. schungen, 16. 395, Strassburg, 1904. 

2 On this mistaken folk-etymology, 8 See Wilson, Persian Life, p. 46 ; 
see Hubschmann, Die altarmenischen Lynch, Armenia, 1. 182. 



18 ERIVAN AND THE ROAD TO THE PERSIAN BORDER 

town back to the days of Noah. The name Erivan, according 
to some authorities, stands for Erevan and is derived from the 
Armenian erevan, ' appearance,' since 4 dry land first made its 
appearance here after Noah's flood.' According to others, it 
owes its origin to the legendary king or heros eponymus Ero- 
vand, or to the Armenian leader Erovant or Ervand, who was 
overthrown by the Persians in the first century of our era. 1 
The one etymology is as unlikely as the other. 2 According 
to the Armenian historian, John Katholikos, Erivan was a 
place of considerable size in the seventh century, although we 
know little if anything about it till the sixteenth, when the 
possession of Erivan became a bone of contention between the 
Persians and the Ottoman Turks. 3 The question was finally 
settled in 1827, when the Russians took possession of the city. 
The bazaars accordingly show abundant evidence of the Rus- 
sian occupation and advance in trade, and I was able at Erivan 
to add to my kit a number of European necessaries not ordi- 
narily found in Asiatic marts. 

In its general characteristics Erivan is Oriental, not Occi- 
dental, but it does not show the many signs of antiquity which 
might be expected of its age. This is largely due to the 
frequent wars between the Turks and the Persians wars 
rendered more cruel by reason of the bitter hatred that exists 
between the rival Mohammedan sects of Sunnis and Shiahs 
to which the two nations respectively belong. These conflicts 
have helped to destroy numerous monuments which Erivan 
must have boasted in the past. There are still some mosques 
and minarets worthy of attention, but the chief memorial of 
the Persian period is the palace of the governors, or sarddrs, 

1 See Lynch, Armenia, 1. 209 seq. A.D. 925 (Lynch, Armenia, 1. 210, 

For the name Erovand, see Justi, n. 2). The fact that the Arab traveller 

Iranisches Namenbuch, p. 89. Ibn Haukal does not mention Erivan 

3 Hubschmann, IF. 16. 426. in the tenth century, nor Yakut in the 

8 John Katholikos wrote in the twelfth, can hardly be used as an 

eleventh century, compiling the an- argument, as John Katholikos flour- 

nals of his country down to the year ished between these two writers. 




ECHMIADZIN IN THE SNOW 




BAZAAR OF ERIVAN 



THE MONASTERY OF ECHMIADZIN NEAR ERIVAN 19 

adjoining a mosque whose dome is decorated with colored tiles 
and inlaid arabesques. The palace itself forms part of a for- 
tified enclosure overlooking the Zangi River, which sweeps 
beneath its mud walls. Although falling into decay, the 
audience-hall of the palace still shows signs of former splendor, 
with fretted ceilings inlaid in Persian style with tiny mirrors, 
and with walls adorned by historic paintings. Among the 
more modern portraits is that of Fath Ali Shah, the great- 
great-grandfather of the present Shah ; but the paintings 
which interested me most were those of Sohrab and Rustam, 
whose fatal conflict Matthew Arnold has retold from Firdausi, 
and of Feramurz, Rustam's second son. At the end of the 
great chamber was a huge latticed window, whose panels were 
inlaid with small patines of colored glass, through which the 
sun shone with a rich iridescence. The view through this 
dazzling frame as the afternoon faded was fine. Far in the 
distance, but ever close at hand, loomed snow-clad Ararat ; 
beneath the palace walls and the buttresses of the fort swept 
the river Zangi, held in check only by its precipitous banks ; 
a caravan wended its way slowly across the bridge ; and 
through the deserted hall of princes shone a gleam of sunlight 
reflecting the splendor of by-gone days. 

The greater part of my second day at Erivan was spent in 
a visit to the Armenian church and monastery of Echmiadzin, 
about thirteen miles distant, near the village of Vagharshapat. 
This celebrated cloister is the seat of the Katholikos, or patri- 
arch, of the Armenian church. It is said to have been founded 
by Gregory the Illuminator, who converted the Armenians to 
Christianity early in the fourth century, and relics of the saint 
are preserved among its treasures. The library contains a 
valuable collection of Armenian manuscripts, among which 
I noted some particularly fine copies of the gospels, 1 some 

1 The most recent notes on these 1904 ; id. Quelques JZvangeliaires Ar- 
are by Meillet, Journal Asiatique, meniens Accentues, extr. from Des 
165 (10th ser. 14), pp. 487-507, Paris, Memoires Orientaux, Paris, 1905. 



20 ERIVAN AND THE ROAD TO THE PERSIAN BORDER 

ancient boundary stones inscribed with cuneiform characters 
(not Persian), a rich assortment of Parthian, Roman, and 
Sasanian coins, and many other objects of interest to the 
archaeological student. The priest who conducted me through 
the various buildings had studied Oriental languages under 
my teacher Geldner in Berlin, so that we had a bond of 
sympathy at once, and excellent opportunities were afforded 
me to see the collections. 

The next afternoon I arranged to leave Erivan for Julfa on 
the Persian border. It was now necessary to procure a post- 
chaise, as the railroad had been laid no farther than Erivan, 
but I had the greatest difficulty in obtaining any sort of 
conveyance. A party of Russian engineers had first claim 
on all horses, as they were on their way to Persia to build 
for the government macadamized roads, -over which the rail- 
way lines can be laid down later. At last a vehicle was 
secured ; it was a heavy phaeton, with a large leather hood 
over the back, and it was drawn by four horses. There was 
much ado in getting the baggage lashed securely on behind, 
and after many delays and petty annoyances the journey was 
finally begun. 

Progress was difficult and slow. Ice, snow, and slimy mud 
filled the roads and rendered speed impossible. The inner court 
of the first caravansarai at which we halted looked like a lake, 
although the room on each side of the entrance was habitable. 
I took the one on the right, as the left was already occupied by 
rough natives who wore heavy sheepskin caps and were armed 
with daggers and long guns. If there had been any Arme- 
nians among them, I presume my Georgian guide Rustom would 
have called them hogs, for he muttered cochon every time we 
passed a group of Armenians on the road. I was too tired to 
take particular notice of my surroundings and was glad at 
the haste with which Rustom arranged my camp-bed. In an 
instant I was asleep. 

Shortly after midnight I awoke with a start. A dark figure 




ECHMIADZIN IN THE SNOW 




BAZAAR OF ERIVAN 



FROM EEIVAN TO JULFA 21 

with long ebon beard and black cowl concealing the face stood 
at the foot of my cot. In the uncertain light the form looked 
like that of a giant. I grasped instinctively for my revolver, 
but a salutation came from the muffled lips, and I saw that 
the newcomer had no malicious intent. He was accompanied 
by an elderly woman, robed partly in black, and she nodded 
a silent greeting. Both used Russian, but spoke very little. 
We interchanged civilities and cigarettes, but I did not inquire 
about the purpose of their journey, nor they about mine. As 
we were to start before daylight, I made no attempt to sleep 
again, and in a couple of hours I had resumed my seat in the 
post-chaise, followed by the vehicle of my fellow-travellers. 
The darkness was thick, the snow blinding, and the road was 
continually blocked by trains of camels, caravan succeeding 
caravan in endless procession. I counted two hundred of 
these dromedaries in less than a quarter of an hour. They 
were chained in strings of eight, ten, or a dozen, and their 
heavy bells, donging monotonously as each one passed, made 
the count easy. 1 

Toward dawn I could catch a glimpse of the wagon of our 
companions following through the snow. Across the front of 
their vehicle was a long, narrow box, and I then learned that it 
was an empty coffin in which the woman was to bring back the 
body of one of her sons, who had died ten days before in a re- 
mote district. I now knew the reason for her travelling through 
the storm. The return journey must have been even more sad 
for her. 

The sunless day was passed driving mile after mile in the 
teeth of biting sleet and hail. My face was frost-bitten, my 
hands were chapped, and altogether I suffered severely from 
the cold, relieved only when we halted at a post-house for 
a change of horses. I did not, however, mind the discom- 
forts so much as might be expected, since something new was 
constantly occurring in this desolate region shrouded in snow, 
1 The Persian camel bell dongs rather than dings. 



22 ERIVAN AND THE ROAD TO THE PERSIAN BORDER 

and I enjoyed the novelty of the experiences. Once a wolf 
skulked across the plain near our path, but it was too dark 
for a shot, and he disappeared into the gloom. 

We travelled on through the night until we reached Nakhi- 
chavan, or Nakhjavan, where there was a chance for a slight 
rest. This place was known to the Greek geographer Ptolemy, 
in the second century A.D., as Naxouana 1 ; but it lays claim to 
far greater antiquity, since tradition and popular etymology 
make it ' Noah's first station ' (Armenian nakh-ijevari) after he 
came from the ark, and his reputed tomb is shown as a sacred 
shrine. 2 Ibn Haukal, among others, mentions Nakhjavan in 
the tenth century of our era, and Yakut, two hundred years 
later, tells something of its history, 3 for events of considerable 
moment have taken place there, and we know that, situated near 
the river Aras, it has been several times the scene of bloody 
encounters between warring armies. 4 Ridiculously enough, my 
recollections of Nakhichavan are chiefly associated with the 
excellent bread I found at the rest-house, in contrast to the 
Persian bread on which I had afterward to subsist. Finally, 
in the forenoon of Saturday, March 14, I reached Julfa, on the 
Aras, the ancient Araxes. In the classic writers this river was 
proverbial for its swift current, whose rushing descent from the 
Armenian mountains carried away the bridges in the winter 
time, so that Vergil calls it ' the stream intolerant of any span ' 
pontem indignatus Araxes. 5 The river is now the boundary 
between Persia and Russia, although historically the confines 
of Iran have always extended far beyond the Aras. 

At Julfa an incident occurred in connection with the Rus- 
sian export customs for which I was not prepared. The cus- 
toms officers when examining my baggage kept asking only one 

1 Ptolemy, Geog. 5. 13 (941). 8 Ibn Haukal, p. 165 ; Yakut, tr. 

2 See Wilson, Persian Life, p. 47 ; Barbier de Meynard, pp. 561, 665, 
Perkins, Eight Years in Persia, p. 134, n. 1. 

Andover, 1843 ; and, on the fanciful 4 Cf. Lynch, Armenia, 1. 345. 

etymology, consult Hubschmann, IF. 6 Vergil, ^Eneid, 8. 728. 

16. 465. 



CROSSING THE ARAS 23 

question, but I did not understand Russian, nor did they 
know either French, German, or English. The inspection 
came to a standstill, therefore, until old Rustom, who had not 
previously been allowed to enter the customs lines, was sum- 
moned. In answer to the inspector's query he promptly as- 
serted that I was not carrying arms, to which I responded as 
promptly that I had a revolver at my belt. My straightforward 
reply cost me the pistol, as a new tariff law forbade the export- 
ing of firearms to Persia ; but my honesty was repaid by my 
not having to see the weapon confiscated, and permission was 
gran ted* me to send it back to Tiflis by Rustom. The person 
to whom I consigned it afterward presented the weapon to an 
American missionary whose revolver had been stolen on a 
journey. I parted with the pistol with regret, and afterward 
when alone in remote places in Persia I several times missed 
this steel companion with heart of lead. 

The customs examination being finished, I said good-by to 
Rustom, since he could accompany me no farther, being a 
Russian subject and therefore not allowed to leave the Tsar's 
kingdom without permission. I had to cross the Aras alone. 
Standing on the shore and awaiting some means of transport, I 
felt with Troilus 4 like a strange soul upon the Stygian bank, 
staying for waftage.' Charon's boat, which was to ferry me 
over this Persian Styx, arrived at last. It came in the form of 
a rude scow made of coarse planks knocked together in the 
roughest possible fashion. Fiends clamoring for obols held the 
oars, dropping them eagerly to grasp bakhshish or clenching 
the fist to demand more. Fortunately the crossing was 
quickly effected, and when we reached the Iranian ' Lethe's 
wharf ' I quite forgot the dark past, remembering only the joy 
of awakening in the longed-for paradise of the Province of the 
Sun. 



CHAPTER IV 

PERSIA, THE LAND AND ITS HISTORY, AND OUR INTEREST 
IN THE COUNTRY 

* What have we to do 
With Kaikobad the Great, or Kaikhosrii? ' 

FITZGERALD, liubaiyat of Omar Khayyam, 10. 

4 MY father's kingdom,' said Cyrus the Younger, < extends so 
far to the south that men cannot live there because of the 
heat, and northward to where they cannot exist because of the 
cold.' 1 This proud boast may be taken almost literally, for 
Persia is a land of extremes, from the frigid winters of the 
high altitudes of Azarbaijan to the torrid summers of the 
Persian Gulf ; but the climate in the interior as a whole is 
temperate when we consider the latitude Shiraz, for example, 
being farther south than Lahore in upper India. This fact 
is sometimes lost sight of. 

Geographically the country of Persia is a great tableland - 
the plateau of Iran which extends beyond its eastern bor- 
ders into Afghanistan and covers altogether an area nearly 
I one-fifth as large as the United States. Mountains with rocky 
passes guard the approach at almost every point of the frontier 
and run their barriers into the interior to hold back the great 
deserts which threaten to invade from the east. A part of the 
vast tableland is well watered, but there are no rivers in Persia 
that are worthy of the name, and most of them lose their streams 
in the soil before becoming tributary to other bodies. In many 
parts of the kingdom, therefore, owing to the dearth of water, 
it is necessary to resort to irrigation in order to convert other- 

1 Xenophon, Anabasis, 1. 7. 6. 
24 




THE DIEULAFOY FRIEZE OF ARCHERS FROM SUSA 



ONE OF THE GREAT NATIONS OF ANTIQUITY 25 

wise arid districts into arable lands or to prevent them from 
sinking back into barren wastes. Irrigation was synonymous 
with righteousness in the old Zoroastrian religion, and agricul- 
ture a religious duty. The soil of the country responds rap- 
idly to tillage, and there are districts in Persia which are 
accounted among the most fertile in the world. 

Over this wide and varied expanse of territory there is 
spread a population which is estimated at more than ten million 
souls. The number is not large for the area occupied, and 
Persia is not counted as a densely settled country. Ethnologi- 
cally the people are of Aryan stock, but they show an admix- 
ture of foreign blood introduced by conquest or due to contact 
with border nations. This latter is especially true in the case 
of the strong infusion of Turkish and Tartar blood in the north- 
west and northeast. In general the inhabitants of Farsistan, 
the original Persis, have remained freest from foreign elements 
and have preserved more nearly the Persian type of Darius, 
who boasts in his inscription that he is c a Persian, the son of 
a Persian, an Aryan and of Aryan blood.' 1 Purest of all, per- 
haps, though few in number, are the Zoroastrians, who have 
maintained the old Iranian religion and have never intermar- 
ried with alien races. 

Historically Persia is one of the great nations of antiquity. 
Of all the Eastern countries which came into contact with 
Greece and Rome, Persia alone has preserved her indepen- 
dence. Her monarchs have been rulers for three thousand 
years, and the Shah on the Peacock Throne to-day may boast 
his claim as inheritor of King Jamshid's legendary rule as 
well as the sceptre of the Median Deioces and the crown of 
Cyrus the Great. 

Bactria, Media, and Persia were the three historic kingdoms 

of Iran. Bactria, whose dynasties are partly legendary, was 

subjugated by Media after the latter, under Deioces, had 

thrown off the Assyrian yoke about B.C. 708. Phraortes, his 

1 Inscr. Nakhsh-i Rustam, a 13-15. 



26 PERSIA, THE LAND AND ITS HISTORY 

son and successor (B.C. 647-625), and Cyaxares, his grandson 
(B.C. 625-585) , were wise and powerful rulers, extending the sway 
of Media as far as Egypt ; but Astyages, the son of Cyaxares, 
proved a feeble monarch, and Media forfeited her supremacy. 
The province of Persia, led by Cyrus, revolted against her ; 
Astyages was defeated in battle, and Cyrus became king of the 
united Medo-Persian empire (B.C. 558-530), and founder of the 
Achsemenian dynasty. Then followed Cambyses, his mad son, 
whose misrule lasted for eight years, ending in death by his 
own hand (B.C. 522) on the way back from Egypt to recover 
his crown, which had been seized by a Magian priest, Gaumata, 
called Smerdis the Usurper, from his impersonating the king's 
dead brother. The imposture was discovered, and the false 
Smerdis was slain by Darius Hystaspes, who now ascended the 
throne (B.C. 522). This able monarch reorganized the empire 
on broad and far-reaching lines and ruled with great ability 
for more than thirty years (B.C. 522-486). Signs of weakness 
had already shown themselves, however, in the unsuccessful 
attempt of Darius to invade Greece, but these marks of de- 
cadence became more and more manifest in the reigns of Xerxes 
and Artaxerxes, until the tottering throne of the Achsemenidse 
fell when Darius III (Codomannus) was conquered by Alexan- 
der the Great and afterward perished (B.C. 323). The Grecian 
arms thus proved triumphant over the Persians within the 
borders of Iran, as they already had at Marathon, Salamis, and 
Platsea. 

The invasion and partial subjugation of Persia by Alexander 
resulted in the establishment of the Grseco-Bactrian govern- 
ment of the Seleucidae, which lasted for seventy years and was 
followed by the Parthian dynasty, which ruled the fortunes of 
Persia for five centuries (B.C. 250-A.D. 226). They in turn 
yielded to the triumphant ascendancy of the Iranian house of 
Sasan, who restored the Zoroastrian faith as state religion and 
dreamed of forming a great national power. Their rule lasted 
over four centuries (A.D. 226-651), but their hope of establish- 




DARIUS III DEFEATED BY ALEXANDER IN THE BATTLE OF ISSUS 

(Pompeian Mosaic in the Naples Museum) 




YAZDAGARD III SLAIN IN A MILL AT MERV 
(From the Columbia University Manuscript of the Shah Namah) 



r*mAi*5 

O^ THE 

UNIVERSITY 

rr 



FROM CYRUS TO THE PRESENT SHAH 27 

ing a world -empire was shattered by the Arab invasion, which 
resulted in the conquest of Persia and the overthrow of the 
Zoroastrian dynasty of the Sasanidse, A.D. 651. With the 
death of Yazdagard III, who was treacherously slain in that 
year after being defeated in battle, the Sasanid line came to 
an end, the Zoroastrian faith, which had been the state re- 
ligion for more than a thousand years, was deposed, and Islam 
took its place as the national religion of Iran. 

The centuries which followed were often marked by mis- 
rule, invasion, and even foreign rule. A succession of longer 
or shorter lived dynasties, like the Ommiads (A.D. 661- 
749), Abbasids (749-847), Ghaznavids (961-1186), Seljuks 
(about 1030-1200), the Mongols under Jenghiz Khan (1162- 
1227) and under his grandson Hulagu (d. 1265), who main- 
tained his court at Maraghah, and the Tartars under Timur 
Lang (d. 1405) and his successors, fill the pages of Persia's 
history until the fifteenth century. At the end of the six- 
teenth, Persia saw the rise of a great sovereign, Shah Abbas 
(1585-1628), who wielded the sceptre alike with regal power 
and magnificence, and at his court representatives of Euro- 
pean potentates were received and entertained with pomp. 
His successors unfortunately proved inferior in ability, and 
the Afghans invaded Persia in the eighteenth century and 
contributed to the general disorganization that lasted until 
about the year 1789. The Kajar dynasty was then established 
by Agha Mohammed Shah, the eunuch monarch, who was 
succeeded by his nephew Fath Ali Shah (1798), and he in 
turn by Mohammed Shah (1835), and this line has held the 
Persian throne ever since. 

The scope of this book does not permit my touching upon 
the past or present relations between Persia and the Occident, 
or making any forecasts as to the future. I shall also forego 
saying anything about the social institutions of modern Persia 
as compared with ancient Iran. In religion, however, Persia 
has played so important a part, a part not wholly laid aside, 



28 PERSIA, THE LAND AND ITS HISTORY 

that some idea of her religious history must be given in order 
to make clear many points in the chapters which follow. 

Zoroastrianism was the ancient faith of Iran and is impor- 
tant because of the likenesses which it presents to Judaism and 
Christianity. 1 A phase of this religion known as Mithraism 
penetrated into the Roman world during the early Christian 
ages and spread so rapidly in many parts of Europe that altars 
were set up and cave-temples built to celebrate the mysteries 
of the Persian divinity Mithra and to glorify this personi- 
fication of light, the sun, and truth. Furthermore, the sys- 
tem of Manichaeism, which sprang up on Persian soil, was 
powerful enough to compete for a time with Neo-Platonism 
and Christianity for the religious and intellectual supremacy 
of the Roman Empire. Mohammedanism is the religion of 
Persia to-day, as she accepted Islam at the time of the Arab 
conquest, but Persia belongs to the Shiite sect of the faith and 
acknowledges Ali, Mohammed's first cousin and son-in-law, 
as the Prophet's successor in opposition to the Sunnite branch 
of Islam. She is in fact the chief representative of Shiism and 
has been largely instrumental in the growth of this factional 
movement which divides the Mohammedan world with a bloody 
schism. In Persia, moreover, within the last seventy years a 
new religious movement, eclectic in its character and known 
as Babism, has sprung up and assumed such proportions as to 
menace the universal supremacy of Mohammedanism in Iran 
and even to attract attention and some followers in the 
Occident. 

In art and architecture Persia is renowned for the grandeur 
of some of her ancient monuments and for the beauty and 
decorative design of much of her later work. In both these 
fields she is believed to have borrowed in early times largely 
from Assyria and Babylon and slightly from Egypt, and later 
also from Greece, Rome, and Byzantium, as well as somewhat 
from China. Nevertheless she has dealt with the importations 
i See pp. 57-69, below. 




THE PORTAL OF XERXES AT PERSEPOLIS 



OF TH. 

UNIVERSITY 

OF 




RELIGION, ARCHITECTURE, AND LANGUAGE 29 

freely, added much, and made the production so character- 
istically her own as often to bring forth a new creation ; and 
if she has accepted gifts in artistic lines from China, it was 
only in part return for generous loans previously made to 
Chinese art by herself. 

In the domain of linguistics there are a number of points 
which are interesting to consider in connection with Persia. 
For the older languages I need only refer to the contributions 
which were made to comparative philology, as well as religion 
and history, when the Zoroastrian scriptures were discovered 
and the cuneiform inscriptions deciphered. The study of the// 
Pahlavi, or Middle Persian, texts, inscriptions, coins, and gems 
has yielded valuable results for general history as well as for 
linguistic science. The modern language of Persia has an 
interest even for the student of English who is not an 
Iranian specialist, because the loss of inflections and the 
admixture of Arabic words in Modern Persian, due to the 
Mohammedan conquest, may be paralleled with similar phe- 
nomena in our own tongue with its levelled case-endings, 
analytic structure, and its vast infusion of words brought in 
by the Norman invasion. In the matter of linguistic purity 
and the avoidance of foreign words in a national epic, Firdausi's 
Shah Namah, Book of Kings (A.D. 1000), affords an excellent 
parallel to Layamon's poetic chronicle, the Brut (A.D. 1200). 
The Iranian poet is as free from the contamination of Arabic 
words, which later became fashionable, as the British bard 
from elements of Norman-French origin. 

Our ordinary vocabulary of to-day owes something to 
Persia. 1 So common a word as van, a heavy vehicle, is an 
abbreviation of caravan (which has been etymologized in the 
folk-speech as ' carry-van '), and is as Persian as Shah, tiara, 

1 1 am indebted for suggestions to Breslau, 1900. See also Skeat, Ety- 

the sketch by my friend, Professor mological Dictionary, p. 759, Oxford, 

Horn, Was verdanken wir Persien, 1882 ; and my address in Congress of 

in Nord und Sild, Heft 282, p. 379, Arts and Science, St. Louis, 1904. 



30 PERSIA, THE LAND AND ITS HISTORY 

bakhshish, and magic (from Magi). The Persian term bazaar 
is current in English, and shawls, sashes, awnings, turquoises, 
and taffeta are standard articles in our linguistic stock in trade 
as the goods themselves in our markets. Products so common 
in America as the orange, lemon, melon, and peach (the last 
word being a disguised form of the Latin malum Persicum, 
which has come to us through the French) are Iranian in name 
as well as in origin. | The vegetable spinach is Persian, and the 
word asparagus also traces its lineage apparently through the 
Greek ao-Trdpayos ultimately to Avestan sparegha, 4 shoot, stalk.' 
I must add, however, that this vegetable has gained much in 
delicacy by being transplanted to the West, if I may judge by 
the asparagus which now grows in Persia. The list of our 
linguistic indebtedness might be increased by including a score 
of words like julep (familiar in 'mint julep'), which is really 
an arabicized form of the Persian guldb, 4 rose-water ' ; hazard, 
applied to taking the one chance in a ' thousand ' (Pers. 
hazdr) ; and last but not least, Paradise, which has come to us 
from Persian through the Greek, while gul and bulbul, the 
Persian 'nightingale and rose,' are familiar to all readers of 
Eastern poetry. 

The title of Persian literature to a place among the great 
literatures of the world is a recognized one, and it is in this 
domain perhaps that Persia makes the greatest claim upon our 
interest. In age the Avesta and the Old Persian Inscriptions 
carry us back at least to the sixth century before Christ and 
possibly earlier ; the Pahlavi literature belongs to the Sasanian 
period from the third to the sixth century after Christ ; and 
the Modern Persian began within the last thousand years. It 
sprang up a century or two after the Arab conquest as a re- 
naissance movement with the revival of the old national feeling ; 
and this period is certainly the most interesting of all. Some 
knowledge of Firdausi, Saadi, and Hafiz belongs to true cul- 
ture, and Omar Khayyam has become an English classic 
through FitzGerald's version. The less-known names of the 




THE PORTAL OF XERXES AT PERSEPOLIS 



PERSIAN LITERATURE 31 

romantic poetic Nizami, the dervish Jalal ad-Din Rumi, and 
the mystic Jami (d. 1492), the last classic poet of Persia, 
should be mentioned as deserving to be known to lovers of 
literature. 

Little space remains for writing about the influence of Persia 
upon our own poetry. Persia was hardly known to England 
before the sixteenth century, yet Chaucer alludes to Persian 
blue, ' pers,' in the Prologue. Among the Elizabethans, Pres- 
ton dramatized the story of 4 Cambises,' Marlowe has Persian 
names and Persian scenes in his Tamburlaine, and Shakspere 
alludes to 4 Persian attire ' in King Lear, to 4 a Persian prince ' 
in Merchant of Venice, and to a voyage to Persia in his 
Comedy of Errors. Milton summarizes the early history of 
Persia in the third book of his Paradise Regained, besides 
referring to ' Ecbatan,' ' Hispahan,' 4 Tauris,' and 4 Casbeen ' in 
Paradise Lost. Shelley appears to have a faint reminiscence 
of the pillared halls at Persepolis in his Alastor, and Byron in 
the G-iaour and Landor in the Grebir hark back to the old 
Zoroastrian faith of Iran. Matthew Arnold and Edmund 
Gosse, as poetical writers, came under Firdausi's spell, and a 
dozen other instances might be mentioned where Persia has 
influenced English poets, one of the best known being Tom 
Moore, whose Lalla Rookh is full of the melody, perfume, color, 
beauty, tenderness, and tremulous ecstasy which imagination 
associates with the East. 

In the realm of English prose the two volumes of Persian 
Tales by Ambrose Philips, after a French version, were widely 
read in the latter part of the eighteenth century, and the 
familiar Arabian Nights are really largely Persian. The inimi- 
table Persian novel Hajji Baba of Isfahan, by Morier, is so 
thoroughly Oriental that Persians who read English mistake 
it for a serious composition and take umbrage at some of its 
amusing accounts. One of our American contemporaries, 
moreover, the novelist Marion Crawford, chose Zoroaster as a 
character around which to weave a romantic story. To these 



32 PERSIA, THE LAND AND ITS HISTORY 

examples I might add dozens of others if I chose to go outside 
of English and speak of the influence of Persia upon French, 
German, and other European literatures. I shall restrict my- 
self, however, and return to my main theme, resuming tht 
journey through the country whose history and position in the 
world I have briefly sketched. 




THE PERSIAN INN AT JULFA, WITH THE CUSTOM HOUSE ON THE LEFT 




A PART OF THE ' ROAD ' BETWEEN JULFA AND MARAND 



OF THE 

UNIVERSITY 

OF 



CHAPTER V 
THROUGH THE SNOW FROM THE ARAS TO TABRIZ 

' They proceeded thence all the next day through the snow. ' 

XENOPHON, Anabasis, 4. 5. 7. 

IMMEDIATELY after crossing the Aras at Julfa I had to 
proceed to the custom house. There I was received by the 
Director of the Persian Customs, a Belgian gentleman, who 
was in charge of the frontier at this point. After scanning 
my letters of introduction and my official papers, he made an 
inquiry only as to whether I carried arms and ammunition. 
I told about the episode across the river with my revolver. 
When the formalities were over, he extended to me a cordial 
invitation to be his guest at dinner that evening, an invitation 
which I gladly promised to accept as soon as I could dispose 
of my luggage at the Persian rest-house across the way. 

This lodging-place was a 'house founded literally upon the 
sand, for it was built near the low bank of the Aras ; it was 
long, but not deep, had two stories and fairly large rooms, a 
double veranda across the front, and a flag-pole on top the 
latter a mark of Western influence. About the entrance were 
strewn bales of cotton, which a caravan had just unloaded, and 
in the rear was the camel train. The dromedaries were being 
quartered for the night in the open. They were forced to 
kneel down in a circle around a bundle of fodder, which 
helped to keep them in order. The shouts, kicks, blows, and 
punches of the drivers, which accompanied this proceeding, called 
forth a score of inarticulate growls, protests, and objections on 
the part of the camels. It was fortunate perhaps that I did 
not understand either camel language or camel-driver jargon. 
*> 33 



34 THROUGH THE SNOW FROM THE ARAS TO TABRIZ 

At the telegraph office adjoining the rest-house I received 
a message from the head of the American Christian Mission 
at Tabriz saying he had despatched an Armenian servant to 
meet me, and sent a wagon drawn by four horses, with a 
Turkish driver named Meshad Seyid Ullah. I welcomed this 
assurance of a conveyance to take me to Tabriz, found my 
attendants had arrived, and then enjoyed a delightful evening 
with my host, who gave me much information regarding the 
route over which I was to travel. I rested well in my Persian 
quarters except at intervals when the camels set up a cry of 
protest against some wrong, real or imaginary. 

It was ten o'clock next morning before I succeeded in get- 
ting everything ready to start on what proved to be a two 
days' journey through the snow, and altogether the worst expe- 
rience I had yet encountered ; but when travelling in Persia we 
become accustomed to discomforts and inconveniences which 
otherwise would seem unbearable. Two quotations from 
Hamlet kept recurring to my mind: one was, 4 the hand of 
little employment hath the daintier sense ; ' the other, 4 thus 
bad begins, but worse remains behind ' and worse did remain 
behind. 

For part of the first day the route was through the exposed 
bed of a river filled with boulders of stone and blocks of ice. 
Now we were sinking in the water, next plunging into a snow- 
bank, and again extricating the wagon from a deep gulch. 
The mud on the side hills was nearly up to the hubs, so there 
was no chance for progress with a vehicle there ; nevertheless 
I was glad to climb up on the heights for a while, and try walk- 
ing, in order to lighten the load for the struggling horses below. 

At distant intervals along the trail there were mud cabins 
which served as tea-houses (chdi Jchdnah). These gave a wel- 
come excuse for a halt and refreshment. The tea was good, 
but dirt was plentiful, yet I soon began to be accustomed to 
that, for the descent to Avernus is easy. The delays in getting 
started again were exasperating, and I had to keep incessantly 



OF THE 

UNIVERSITY 

OF 





MOUNTAIN VILLAGE BURIED IN SNOW BETWEEN MARAND AND SOFIAN 



MISHAPS ON THE WAY 35 

urging, scolding, begging, and bribing the driver to make haste 
in order to reach Maraud that night. The device of the bribe 
proved the more effective, and resulted in a series of lashes, 
plied savagely upon the tired horses and accompanied by a 
succession of encouraging shouts, whistles, grunts, cries, 
squeals, yells, and chirrups, infinite in variety, but of endless 
weariness, and alternating with the humming of a tune which 
might have been the Turkish equivalent of that of which the 
old cow died. 

We managed to keep fairly well in the caravan trail (I can- 
not call it a road), but once in the darkness we lost it, and a 
violent collision with a telegraph pole was the result. Fortu- 
nately only the harness was broken, not our bones. After 
making repairs we proceeded tolerably until the village of 
Marand was reached ; there on the bank of a stream the wagon 
suddenly upset, and I was sent sprawling into the mud, amid 
bags, boxes, and bundles. The only thing to do was to take 
the matter good-naturedly and laugh ; this cheered the situa- 
tion immediately, and the villagers came out in a friendly 
manner from their simple homes, helped me to replace my 
scattered belongings, and guided us to a place of lodging. 

The upper room where I spent the night was fairly comfort- 
able, thanks to a blazing fire, but the heat had the disadvan- 
tage of bringing out from the cracks and crevices scores of huge 
vermin, descendants, perhaps, of the noxious khrafstras of the 
Avesta. I slept soundly, nevertheless, for a journey of eleven 
hours is conducive to weariness, although the distance covered, 
despite all my efforts, was only forty-five miles. 

As a place, Marand is no longer of any consequence, although 
it was once an important town. Yakut says that even in his 
time, seven centuries ago, it was partly abandoned and falling 
into ruins because of the ravages of the Turkish tribes who 
swept down upon it, carrying off the inhabitants and leaving 
desolation in their wake. 1 It is clear from his account that 

i Yakut, p. 524. 



36 THROUGH THE SNOW FROM THE ARAS TO TABRIZ 

religiously Maraud must have been a stronghold of Islam at his 
time, and no longer Zoroastrian, as he states that it was the 
birthplace of a number of eminent Mohammedan teachers. At 
the time of my visit I did not know that at Maraud there are 
the remains of an ash-hillock which is believed to go back to the 
days of Zoroastrian fire-worship, and, like the mounds at Uru- 
miah, to owe its origin to a vast accretion of ashes from a 
fire-temple. If I had known of this fact at the time, I should 
have examined the mound. The antiquity of the town, as it 
was once the capital of the Sasanian canton of Vaspurakan, 
would favor the likelihood of a reward for undertaking exca- 
vations in the vicinity for Zoroastrian researches ; but as far as 
biblical matters are concerned, there is nothing except the fan- 
ciful etymology of the name as Mair-and, 4 the Mother is there,' 
to support the tradition that Noah's wife is buried at Marand. 1 
The next morning the weather was dull and dreary, and it 
was nine o'clock before I could start. A few minutes later we 
were crossing a ford of the stream on whose bank the wagon 
had upset the night before, and whose water we had drunk for 
breakfast with the assurance that it was 'most excellent' 
(db-i Jchaill khub). I now saw a dead cat floating on its sur- 
face and the villagers washing their dirty clothes in the stream. 
A short drive through slush and mud, after crossing the ford, 
brought us to the foothills, and all that day the route lay 
up steep mountains and down into deep valleys, although the 
altitude of the latter was rarely less than four thousand feet 
above sea-level, as the tableland is high at this point. The 
mountain scenery looked like a sea of gigantic billows raised 
by some Titan storm that had torn up its surface. And snow 
was everywhere. The depth of the snow made progress very 
slow, and once our vehicle became hopelessly stuck in a huge 
drift, and I had to pay handsomely for extra horses to pull 
it out. The country was sparsely settled, and many of the 

1 On the name Marand see the ar- ischen Ortsnamen, in IF. 16. 347, 
tide by Hiibschmann, Die altarmen- 451. 



MARAND AND SOFIAN, HISTORIC PLACES 37 

hamlets were buried in the snow, one or two on the mountain- 
side (like the one in my picture) looking as if their occupants 
had hibernated all winter, so completely was their communica- 
tion cut off. Most desolate among all the sights, however, 
was a ruined caravansarai, which, like a hundred others in 
Persia, was attributed to Shah Abbas the Great as builder. 
The magnificence of its founder and the former splendor of 
its many kingly occupants, in contrast with its present ruinous 
condition, struck me as an illustration of Omar Khayyam's 
familiar quatrain : 

' Think, in this batter'd Caravanserai 
Whose Portals are alternate Night and Day, 

How Sultan after Sultan with his Pomp 
Abode his destined Hour, and went his way.' 

It was late in the afternoon when we finally drew near the 
village of Sofian, or Zofian, which had been the scene of a bloody 
battle between the Turks and the Persians in the latter part of 
the sixteenth century. 1 As there is a post-house at Sofian, I 
proceeded to arrange for hiring horses, since those of my 
driver were completely fagged out. The master of the post, 
with whom I had to bargain, had a nose shaped like a carrot, 
and evidently Turanian blood in his veins ; the Persian linea- 
ments seemed to be almost wanting in his face. In his man- 
ners, however, he was kindly ; in his movements, Orientally 
slow ; but with much dignity he conducted me to his own room, 
which was warm and comfortable and furnished with some 
good rugs and divans. Two Persian merchants were lolling 
on cushions, drinking tea, and seemed to have so much time 
at their disposal that they were willing to spend more of it 
in asking questions of the newly arrived farangl, than I had 
time to spend in answering. 

A start was made at last, and as I left Sofian I got a good 
view of the northeastern shore of Lake Urumiah and of Mount 
Sahand. Both of these places were for me historic landmarks, 
1 See Ker Porter, Travels, 1. 219. 



38 THROUGH THE SNOW FROM THE ARAS TO TABRIZ 

the former as a lake which Zoroaster knew well, and the latter 
possibly, I believe, as Mount Asnavand of the Avesta, on which 
it is said that Zoroaster beheld a vision of heaven and conversed 
with Haurvatat, the guardian angel that presides over waters. 1 
To me the sight of the mountain and the lake was a joyful 
one, because I felt as if I had met friends whom I had longed 
to see ; this made the journey for the remainder of the day 
seem shorter. 

It was between eight and nine in the evening when I finally 
reached Tabriz and found a welcome at the mission house. 
I became a 4 Fire- Worshipper ' in earnest, as my hosts laugh- 
ingly said, when I greeted the blazing logs whose cheery flame 
brought back the blood to my face, which had been cracked 
in deep gashes by the cold. I had been for two full days on 
the road through the snow, having taken all that time to 
accomplish a journey of eighty-five miles. It was a pleasant 
prospect now to be able to look forward to a rest for several 
days, and that in one of the largest of the cities of Persia. 

1 See the suggestions in my Zoro- identifications may be suggested, cf. 
aster, pp. 48, 100, 207, although other p. 141, below. 



CHAPTER VI 
TABRIZ, THE RESIDENCE OF THE CROWN PRINCE 

' In his retreat to Tauris or Casbeen. ' 

MILTON, Paradise Lost, 10. 435. 

TABRIZ, the residence of the heir apparent to the Persian 
throne, and the commercial centre of Azarbaijan, is a city 
whose age and birthplace are not known, but it may count a 
thousand years as but a fraction of its life. The Persian tra- 
dition which ascribes its founding to Zobeidah, the wife of 
Harun al-Rashid (better known as the Caliph Haroun Alra- 
schid of the Arabian Nights, A.D. 800), sets its date too late, as 
in the case of Kashan and other cities which are said to have 
been built by this heroine. It is true that a fountain at Tabriz 
is called after her name, but the city can be shown to have 
existed under the Sasanians, four centuries before her time. 1 

Tabriz has been identified with the ancient Gaza, Ganzaca, 
by some scholars, but this identification is not accurate, 2 nor are 
we positive that it was formerly called Shahistan,' King's-town,' 
by the Persians and had its name changed to Tabriz ('this 
revenge,' ta-vrezJi) by the Armenian king Khosru I, who sacked 
the city, A.D. 346, in revenge for the death of his brother, and 
then called the place i this revenge,' a name which the city has 



tradition of Zobeidah as to Rev. S. G. Wilson, Persian Life 

f6under is given by Mustaufi, Nau- and Customs, pp. 323-325, New York, 

zhat al-Kulub (A.D. 1340), and he 1895. To Mr. Wilson, who was my 

gives the Mohammedan year of the host during a stay of five days in 

founding as A.H. 175 = A.D. 790; see Tabriz, I am indebted for much in- 

Barbier de Meynard, Diet. geog. de formation regarding the city. 

la Perse, p. 132, n. 2. For a sketch 2 For the more likely association of 

of the history of Tabriz I would refer Gaza with Shiz, see p. 131, below. 



40 TABRIZ, THE RESIDENCE OF THE CROWN PRINCE 

borne ever since as a memorial of the event. 1 The Persians 
again recovered possession of Tabriz from its Armenian con- 
querors, but owing to its exposed position on the frontier the 
town has frequently been subjected to foreign invasion and occu- 
pation, by Arab, Seljuk, and Mongol, one of the fiercest of the 
stormings being that by Timur Lang (Tamerlane), who sacked 
it with his Tartar hosts in the latter part of the fourteenth 
century. Terrific earthquakes also have shattered it again 
and again, A.D. 858, 1041, 1721, and 1780, killing thousands 
of people and destroying its main buildings. Nevertheless 
the city has maintained its position as a great Persian metrop- 
olis, inherited from its rank as a capital under the Mongols, and 
has enjoyed prosperity as a centre of trade and commerce, so 
that Tabriz remains to-day what the Arab traveller Yakut 
called it when he visited it in 1203 (A.H. 610), 'the principal 
city of Azarbaijan, flourishing and well populated.' 2 The 
number of its inhabitants is not less than one hundred and 
seventy thousand. 

A view of the city is disappointing if we expect to find lofty 
buildings and that variety of color which we associate with the 
Orient. Instead of this, there is a monotonous expanse of flat- 
roofed, single-storied houses, broken only by the domed arches 
of the bazaars and the high wall of the ancient citadel. Clay 
and mud plaster, for the most part, are used in the construction 
of buildings, and these give a dull appearance to the unimpos- 
ing architecture. The houses, with windowless outer walls, 
turn their backs on the street and show their faces only to the 
exclusive brick courtyard in the interior. The entrance is made 
through an unpainted wooden door, studded with heavy nails, 
like the portal of a Norman keep, and having a small grating 
above to admit light and air. In the courtyard we may find 
a small garden, and, if so, a tank for preserving that precious 
commodity, water ; but the general appearance of the interior, 
like its unattractive entrance, is not such as to lead one to 
1 Wilson, Persian Life, p. 323. 2 Yakut, p. 132. 



THE CITY AND ITS GOVERNMENT 41 

suppose how handsomely the house may be decorated on the 
inside with rugs, old tapestry, pieces of Persian armor, and 
faience. 

As we walk about the town we have to find our way through 
a labyrinth of streets, narrow passages, and side alleys, some of 
the latter being less than six feet wide, and ultimately we reach 
the outskirts of the city. Walls surround Tabriz, as they 
have from the earliest times, and their circuit has gradually 
increased with the compass of the town. Gardens border these 
walls, with vineyards and orchards lying beyond, and Tabriz 
has been famous for its fruits and vegetables for over a thou- 
sand years. 1 But there was little to suggest this abundance. 
The suburbs, when I saw them, were buried in snow, and so 
also were low hills adjacent to the plain on the north and 
northeast, which looked dwarfed in comparison with the heights 
of Mount Sahand that rise to an altitude of nearly twelve 
thousand feet on the south and are clad in ermine most of the 
year round. 

As a municipality, Tabriz has more pretence to government 
than any other Persian city except Teheran, although there 
could be no comparison, of course, with a well regulated Euro- 
pean city in the matter of efficiency. Twenty-four different 
wards are recognized, each managed by a magistrate (Jcad- 
khuda), who is responsible to the burgomaster (bagldr-bagl), 
and he in turn to the governor of the province (hakim), and 
thus ultimately answerable to the Shah. 2 The streets are 
generally unpaved, except in a few places where cobble-stones 
are laid, and when I was at Tabriz in March, little attempt was 
made to remove the snow and slush, and I understand that the 
dust and dirt in summer are equally intolerable. As the 
streets are not regularly lighted, persons who go out after 
dark carry huge cylindrical lanterns, resembling our Chinese 

1 See the praise of its apricots by the bier de Meynard, Diet. geog. de la 
Arab geographers Yakut (A.D. 1200) Perse, p. 132. 
and Mustaufi (A.D. 1340) given in Bar- 2 Cf. Wilson, Persian Life, p. 66. 



42 TABRIZ, THE RESIDENCE OF THE CROWN PRINCE 

lanterns, but made of thin muslin. The size of these luminous 
transparencies is in proportion to the dignity and position of 
the person escorted (for in Persia one generally goes out with 
a servant as an escort), and a grandee may be recognized by 
a giant lantern, three feet high and twenty inches in diameter, 
carried before him. 

The water-supply of the city interested me, because this prob- 
lem is often more difficult to solve in Persia than elsewhere. 
Most of the water in Tabriz is carried in underground channels 
from the outlying districts and distributed through the town 
by means of cemented conduits and clay pipes. During one 
of the days of my stay the garden of the mission was irrigated, 
so I had an opportunity of seeing how the water was dis- 
tributed. The plug of the pipe leading from the street into 
the yard, which in many cases is a mass of clay or waste rags 
to serve as a stopper, is removed, and the water allowed to 
stream through the channels of the courtyard and into a reser- 
voir (amldr) in the cellar of the house. The gardener super- 
intends the distribution of the stream, which is allowed to run 
for several hours, depending upon the contract, and then the 
waterman again shuts off the supply and opens it in turn for 
the next house. In the management of the water system, 
however, little attention is paid to matters of hygiene, and the 
water becomes much polluted by surface drainage, so that it is 
easy to see how an epidemic like cholera can spread. 1 

The two architectural monuments in Tabriz which have a 
special claim to interest are partly in ruins. The most conspic- 
uous of these is the Ark, or citadel, which may be seen from 
almost every part of the town. This massive structure prob- 
ably occupies the same position as the old building which 
Yakut described seven hundred years ago as ' the Palace of the 
Amir, built of red brick artistically set, and very solidly con- 
structed.' 2 The people call the citadel the Arch of Ali Shah 
(Tak-i All Shah), after the name of Taj ad-Din Ali Shah, who 
1 See also Wilson, Persian Life, p. 70. 2 Yakut, p. 133. 



THE CITADEL AND THE BLUE MOSQUE 43 

was the grand vizir of the Mongol ruler Ghazan Khan, at the 
beginning of the fourteenth century, and built the mosque 
which once formed a part of the structure. 1 The battlements 
of the Ark rise more than a hundred feet in height, and the 
walls are fully twenty feet in thickness. The forbidding 
appearance of the solemn pile agrees well with the story that 
criminals were formerly executed by being thrown from its 
summit. Local accounts add a narrative of a curious escape of 
a woman who was condemned to this horrible death ; her skirts 
and balloon-like pantalets acted as a parachute to break the 
fall, so that she received no injury. 

As I surveyed the towering wall and observed its ancient 
style of architecture, I could not help thinking that it prob- 
ably did not differ much from that on the top of which a tragic 
scene was enacted in the days of the tyrant Cambyses, son of 
Cyrus, as told by Herodotus. According to the famous his- 
torian, the king's grand vizir Prexaspes determined that the 
truth about the usurpation of the False Smerdis should be 
known, though his own life be forfeited in the cause. He 
therefore ascended the tower of the palace and began to harangue 
the people who were gathered below to listen, telling them of 
the glorious reign of their former king, Cyrus, the atrocities 
committed by Cambyses, and describing how treachery had 
brought Smerdis and the Magians into power ; then, before he 
could be seized, he flung himself headlong from the summit to 
destruction. 2 

Far more interesting from the architectural standpoint is the 
second monument of Tabriz, the well-known Masjid-i Kabud, or 
Blue Mosque. This fine specimen of Mohammedan art, dating 
from the middle of the fifteenth century, is now falling into 
utter decay, but its crumbling walls and arches still show grace- 
ful lines and are encrusted with tiles of a rich blue color, set 

1 See the statement of Mustaufi, Wilson, Persian Life, p. 64, and Cur- 
cited by Barbier de Meynard, Diet. zon, Persia, 1. 522. 
geog. p. 132, n. 1, and compare also 2 Herodotus, History, 3. 75. 



44 

off by exquisite faience of yellow, salmon, white, and black, 
interwoven with patterns and arabesque scrolls. It is fortu- 
nate that the French archaeologist and artist Texier preserved 
some of its perishing beauty in his handsome reproductions 
published nearly three quarters of a century ago, and that the 
younger German scholar Sarre, in his fine photographs and 
colored engravings, has also contributed his share toward sav- 
ing more of the Blue Mosque, for it will probably have fallen 
into utter decay before another hundred years are past. 1 

The other buildings in the city require only brief mention. 
There are said to be no less than three hundred and eighteen 
mosques in Tabriz, but none of them can bear comparison with 
the Blue Mosque. The religious merit of the city is enhanced 
by the fact that it can boast of having the tombs of eight 
Imamzadahs, sainted followers of Mohammed and of his son-in- 
law Ali ; while some 4 companions of the Prophet ' are said 
also to be interred on Mount Sahand. 2 These facts bear out 
the proud title 4 Cupola of Islam,' which Tabriz bore even six 
centuries ago, 3 and the city is so bigotedly Mohammedan that 
Jews are said not to be ordinarily tolerated in the town. 

The oldest portion of the city is known as the Kalah, or 
Fortress, although its walls have practically disappeared and 
the moat has been largely filled in and built upon. The 
central part of this old-time fortified section is occupied by 
the bazaars. These marts of trade are among the finest in Per- 
sia, if not in all the East, and are a source of endless interest 
to the traveller. In construction they are of the characteristic 
Oriental type, consisting of acres of vaulted arches built of 
brick and masonry, roofed over and divided by long narrow 
passageways, with shops and alcoves on each side. At occa- 
sional intervals large portals lead out into square courts en- 

1 See Texier, Description de TAr- nard, Diet. geog. p. 133, n. 1. ' Com- 
menie, la Perse, etc., Paris, 1842- panions ' (ashdb) is a technical term 
1845 ; Sarre, Denkmdler Persischer in Islam. 

Baukunst, Berlin, 1901. 8 Mustaufi, op. cit. p. 132, n. 1. 

2 Mustaufi, cited by Barbier de Mey- 




INTERIOR OF BAZAAR AT TABRIZ 



IN THE BAZAARS 45 

closed to serve as places for the caravans to unload and lodge. 
The light in the bazaar is generally dim except where holes 
pierce the domes at regular intervals and let in shafts of sun- 
shine, which serve also to reveal the dirt below. The passage- 
ways of the bazaars always seem crowded ; camels, donkeys, 
and pack-mules add to the confusion of the disordered mass 
of buyers and sellers, and the incessant cry khabarddr! kha- 
barddrf 'take care! take care!' ejaculated by the drivers, 
becomes exasperating, especially when you have to crowd 
against the wall to let some grandee pass. This local notable 
may be mounted on horseback and preceded by a lackey called 
a fardsh, who clears the way with a mace ; but sometimes the 
noteworthy is seated in a carriage and preceded by outriders. 

The bazaars open shortly after sunrise and do not close until 
sunset, at which time the shops are shut with wooden shutters 
and the gates of the bazaar barred and locked. The booths in 
which the goods are displayed measure scarcely more than 
ten or twelve feet square, and often much less. The customers 
do not enter the shop, but walk along the narrow passageways 
and bargain with the salesman, who squats" lazily on the 
brick ledge in front of his store or sometimes rises slowly 
to bring from the rear of the booth an article which the pro- 
spective purchaser wishes to examine. If the price cannot 
finally be agreed upon after long bartering, the face of the 
merchant assumes a look of stolid indifference or Oriental dis- 
dain, and the customer passes along to the next shop. Be- 
yond a rough grouping of industries there is little order or 
arrangement in the distribution of the booths. Here may be 
a fruit-stall with a rich supply of melons, which are kept on 
sale even in winter ; there an Armenian silversmith doing fine 
filigree-work by hand; on this side a cap-maker busy with 
finishing a lambskin hat or a black Persian fez ; yonder a 
baker flapping huge sheets of dough against the sides of an 
earthen oven (tandur). The oven itself is simply a hollow 
scooped in the earth and lined on the sides with pebbles, which 



46 TABRIZ, THE RESIDENCE OF THE CROWN PRINCE 

absorb the heat and bake the giant flap- jack, but impart to 
the bread a peculiar pitted appearance and often a gritty taste. 
We can well understand why this bread is called 'pebble- 
bread ' (ndn-i sangak). During my stay at Tabriz, bonbons and 
tinsel decorations were much in evidence in the bazaars, as 
the season of No-Ruz, the Persian New Year, was approaching. 

In making purchases in the bazaars the monetary unit is 
the 4 kran' (krdn), or rather the two-kran bit, the latter being 
equivalent, roughly speaking, to twenty cents in American 
money. The kran itself is made up of twenty shdhls, each 
worth half a cent, and ten krans make up the ' toman ' (toman), 
something less than a dollar. As an actual coin, however, the 
toman no longer exists except in rare gold pieces, although 
the Imperial Bank of Persia has issued paper tomans, which 
are handsomely engraved notes, but are little current outside 
of Tabriz and Teheran, being exchanged at a considerable loss 
in other towns. For this reason the traveller has to go weighted 
down with bags of silver, when he starts on a journey, and 
they make a heavy addition to his load. 

As Tabriz is 'the commercial centre of northwestern Persia 
and its trade with Europe is constantly growing, I may appro- 
priately add a few words regarding the commercial relations of 
Persia and the United States. 1 The ' Treaty of Friendship and 
Commerce ' between the United States and Persia was concluded 
December 13, 1856, and .came into force in the following year. 
The phrasing of its opening paragraph is interesting, as it gives 
the royal titles of the Shah. I reproduce it verbatim. 

4 IN THE NAME OF GOD, THE CLEMENT AND MERCIFUL 

* The President of the United States of America and His Majesty, 
as exalted as the planet Saturn; the Sovereign to whom the Sun 
serves as a standard ; whose splendor and magnificence are equal to 
that of the skies ; the Sublime Sovereign and the Monarch whose 

J For various printed reports on kindness of Mr. David C. Beatty, of 
Persian trade I am indebted to the Yonkers, N.Y. 



OUR TRADE WITH PERSIA 47 

armies are as numerous as the stars ; whose greatness calls to mind that 
of Jemshid ; whose magnificence equals that of Darius ; the Heir of 
the Crown and Throne of the Kaianians ; the Sublime Emperor of all 
Persia : being equally and sincerely desirous of establishing relations 
of friendship between the two governments which they wish to 
strengthen by a treaty of commerce and friendship and useful to 
the citizens and subjects of the two high contracting parties have 
for this purpose named for their plenipotentiaries . . .' 

Then the names of the respective appointees are given and 
these are followed by a series of eight articles regarding the 
friendly and diplomatic relations of the two countries and 
treating of matters of trade and commerce, as well as the obli- 
gations to be fulfilled by both parties and the privileges to be 
enjoyed. The various items of the treaty contain the 'most 
favored nation clause ' throughout, so that the United States is 
entitled to the same rights and privileges in commercial matters 
as any other nation. 

Trade between our country and Persia is yet in its infancy, 
as is shown by the Consular Reports ; but there are several 
points to which attention may be called as significant. Russia 
has the bulk (about fifty per cent) of Persia's export and im- 
port trade ; Great Britain comes next with about twenty-five 
per cent ; the remainder goes to ' other countries,' under which 
general heading the United States is also included. The lately 
appointed Persian Minister to Washington emphasizes the pos- 
sibility of an extensive increase in the trade between his coun- 
try and our own ; and Mr. John Tyler, our Vice-Consul-General 
at Teheran, shows in his recent reports that there is at least a 
prospective opening for American manufactures, especially for 
agricultural machinery, and a growing demand for American 
merchandise. 

' American lamps, clocks, matches, and locks have a steadily in- 
creasing sale in the Teheran bazaars, especially locks, which excel 
in mechanical complexity, combined with lightness and convenience 
of handling (important considerations), anything hitherto put on 
sale. American hand pumps and cooking and warming stoves find 



48 TABRIZ, THE RESIDENCE OF THE CROWN PRINCE 

appreciative purchasers and should, with proper management and 
competitive enterprise, soon monopolize the market.' 

From experience I can understand how so indispensable an 
article as the padlock is in Persia might find a ready market. 
The wholesale introduction of clocks, I believe, is one of the 
greatest needs in timeless Persia, but I fear that their general 
use will depend largely upon the introduction of railroads, 
which would help to spread more widely the idea that time 
and money are synonyms. I might add that the admiration 
shown by the natives for the leather articles in my travelling- 
outfit leads me to think that American straps, clasps, buckles, 
riding-leggings, and top-boots would find a good sale, for the 
Persians themselves are capable workers in leather and know 
how to appreciate a good calfskin product. The time, there- 
fore, may not be distant when we shall see a larger sale both 
of American merchandise and of 4 Yankee notions ' in the 
bazaars of Persia and a complimentary import in return of 
precious stones, like the topaz, pearls from the Persian Gulf, 
silks, shawls, and embroideries, besides the well-known con- 
signments of carpets and rugs. 

Not far from the bazaars is a large public square to which a 
particular interest attaches, not because of the armory and 
the gunsmiths' shops, the arsenal, prison, royal stables, and 
buildings belonging to the Crown Prince, but because it was 
the scene of the execution of the Bab, a Persian reformer, on 
July 9, 1850. This religious enthusiast and moral teacher, 
whose real name was Mirza Ali Mohammed, was born in Shiraz 
about the year 1820. He was trained at first to commer- 
cial life, but a pilgrimage to Kerbela and Najaf, and after- 
ward to Mecca, awakened in his heart the religious enthusiasm 
which made him devote his life henceforth to developing the 
tenets which he held. Upon his return to his native city, about 
1844, he assumed the title of Bab, or ' Gate ' leading to the 
spiritual life. His religious views were somewhat eclectic ; 
his doctrines leaned toward a mystic pantheism, with elements 



THE BAB EXECUTED AT TABRIZ 49 

of gnosticism, and were of a highly moral order, and so liberal 
as to include steps toward the emancipation of woman. 

In the eyes of the strict Mohammedan, however, the tenets 
upheld by the Bab were rank heresy. Nevertheless, they 
spread rapidly and awakened such intense sympathy among 
those who were dissatisfied with the regime maintained by the 
Persian mullahs, on the one hand, and raised such bitter op- 
position, on the other, among those who were pronouncedly 
conservative, that they led finally to bloody conflicts which 
resulted in the imprisonment of the Bab. He was ultimately 
taken to Tabriz and there condemned to be shot. The place 
of execution was this very square of the arsenal and gun- 
smiths which I am describing. Cords were passed under his 
arms, and he was suspended from the wall above a small shop 
which was pointed out to me. By his side was suspended also 
a devoted disciple, a young merchant of Tabriz, and orders 
were given to the soldiers to fire their volley. When the 
smoke cleared away, the body of the young follower of the Bab 
was discovered, riddled with bullets ; but by some strange hap 
the Bab had escaped. The shots had simply cut the cords 
that held him, so that he fell to the ground unhurt and took 
refuge in the shop below. He was probably dazed ; for had 
he retained his presence of mind, he might at once have turned 
the incident into a miracle before the astonished multitude. 
He was seized, however, dragged forth from the shop and 
again suspended, and shot to death by a different company of 
soldiers, since the first absolutely refused to fire another volley. 
The bodies of the two religious martyrs were then cruelly 
dragged through the streets and thrown to the dogs and 
birds, but they were afterward taken up and buried by 
sympathetic Babis, as the movement had gained a large 
number of adherents. It still has many followers, de- 
spite the persecution to which the sect has been subjected. 1 

1 See Browne, A Year Amongst cially the same author's translation of 
the Persians, pp. 58-64, and espe- the Tarlkh-i-Jadld, or New History of 



50 

Babism, in fact, is not confined to Persia, but has adherents 
in Mesopotamia, Syria, Egypt, India, and even in America, 
where some of its believers have tried to disseminate their 
doctrines. 1 

On the opposite side of the same public square, in the prison, 
another religious martyr was executed some years ago. This 
was a Mohammedan priest who had abjured Islam and adopted 
Christianity. He was cast into prison, confined in an upper 
room which looks out upon the square, and, after being nearly 
starved to death, was finally strangled by a bowstring, refusing 
to the last to renounce his belief in Christ. 

One afternoon of my stay in Tabriz was devoted to a visit to 
the gardens and summer palace of the Vali Ahd, or Crown 
Prince, who makes this city his chosen place of residence, as 
his predecessors have done for the past hundred years. This 
summer abode, with its fine garden, lies on the southern side 
of the city, although it is called 'Northern Garden' (BagJi-i 
Shamal), having taken that name from an older residence on the 
north side which it replaced. The snow lay so deep when I 
saw it that I could gain no real impression of what the park 
might be in spring and summer, but the driveways and avenues 
of trees were attractively laid out, the arched gateway of brick 
was effective as an entrance, and the palace itself more worthy 
of the name than some of the so-called palaces in Persia, which 
are not always kept up well. 

In paying visits in the city I learned something of the 
native etiquette, which has a great deal of charm as well as 
formality. The Persians are distinctly a social people and 
their manners in company are extremely polite. Their vo- 

Mirzd 'All Muhammad the Bab, by 1 There is a society of Babists in 

Mirzd Huseyn of Hamaddn, pp. 299- Chicago who call themselves Behaists, 

312, especially pp. 303-306, Cam- after Beha Ullah, who claimed to be 

bridge, 1893 ; compare also Browne, the successor of the Bab and a mani- 

The Episode of the Bab, 2. 43- festation of the glory of God. See 

45, 182, 190, 321-322, Cambridge, Open Court, 18. 356 seq., 398 seq., 

1891. Chicago, 1904. 




AN ACILEMENIAN SEAL: THE KlNG SLAYING A MONSTER 

(Exact Size) 




THE CROWN PRINCE'S GARDEN 
(Bagh-i Shamal, Northern Garden) 



PERSIAN VISITS AND SOCIAL ETIQUETTE 51 

cabulary of etiquette is rich in courteous phrases and com- 
plimentary terms, and the saldms, or benedictions of peace, 
which form part of the greeting to the visitor, serve as a 
charming introduction to conversation. 

When a visit is to be paid to a person of rank, it is customary 
for the visitor to send word in advance to inquire of the digni- 
tary what hour would be convenient for receiving the call. 
The response comes back couched in some courteous phrase and 
names the time, 4 two hours before sunset,' or perhaps earlier, as 
the case may be. On being ushered into the reception chamber, 
we find ourselves in a large room, richly carpeted with soft rugs 
and lined with divans, but otherwise little furnished 'except with 
a few chairs for Europeans when they are received. The host 
enters a moment later and comes forward to greet his guest. 
As an Oriental he appears in his stockinged feet, for shoes are 
forbidden indoors, and wears his black lambskin cap (JculdTi), 
as it would be bad form to have the head uncovered. His polite 
saldm aleikum, 4 Peace be unto you,' is responded to in kind, with 
a mutual inquiry about the 'august health' of each, after which 
the talk proceeds easily and unaffectedly. 

In a few minutes one of the troop of servants enters, bringing 
the kalian, or water-pipe, as an added mark of hospitable 
attention. This pipe stands about two feet high and is some- 
what elaborate in its structure. The base is a large glass 
vessel of a graceful shape, holding a quart of water. From 
this vase there rises the tube of the pipe, which is about fifteen 
inches long, made of dark wood, sometimes elaborately carved, 
and capped by a China bowl, which is usually decorated with a 
picture of the Shah and a fringe of silver chains hanging from 
its rim. The tobacco is placed in this bowl, after the leaves 
have been moistened and squeezed out, and a square piece of 
charcoal is used to light it and left burning on the top while 
the pipe is in use. The stem itself is about eighteen inches 
long, and is inserted into the water-vase at a convenient angle. 
It is made of the same dark wood as the tube and is capped 



52 TABRIZ, THE RESIDENCE OF THE CROWN PRINCE 

with a mouthpiece of silver. In using the pipe the smoke is 
not drawn by whiffs into the cheeks, as in the case of a cigar, but 
is inhaled in long draughts directly into the lungs, the strength 
of the nicotine being diminished somewhat in passing through 
the water, which is occasionally also scented. After three or 
four long puffs it is en regie to pass the pipe to the next person 
at the right and so on throughout the company. A tray of 
sweetmeats and some tea served in tiny glasses form an addi- 
tional course and complete the hospitality, arid then the guest 
asks permission to 'remove the cause of trouble to the host,' 
and take his leave, which is finally granted after a variety of 
protests. Formality is resumed at the final leave-taking, and 
many compliments are passed in saying good-by. 

The Persians at their banquets, I am told, show the same 
grace in entertaining, and the conversation is easy, bright, and 
witty. There is a great variety of courses, if we count the 
sweets, dried fruit, and other delicacies, which are partaken 
during the evening before the solid dishes are served. This 
takes place before the company is to break up, so that the latter 
are hardly more than a supplement to the lighter delicacies 
which precede them. The custom with the ancient Persians 
was the same. Herodotus says that ' the Persians indulge in 
very few solid dishes, but they eat many desserts, which are not 
served up on the table all together at the same time.' He adds, 
'the Persians are very much addicted to wine.' 1 Since the time 
of Mohammed this abuse has been forbidden, but the injunctions 
of the Koran in this respect are not so strictly complied with 
as they might be. I was informed also that an occasional 
feature of lavish entertainments is an exhibition of dancing 
boys, somewhat similar to the nautch girls of India. These 
boys are said to be handsome youths, but spoiled and effeminate, 
like those at Bokhara and Samarkand, and it may have been 

1 Herodotus, History, 1. 133, cf. Persians, pp. 108-111; Wilson, Per- 
Rawlinson, Herodotus, 1. 219, n. 6. sian Life, p. 243 seq. 
See also Browne, A Year Amongst the 



SAVALAN, A HOLY MOUNTAIN IN AZARBAIJAN 53 

against such minions as these that some of the shafts of Zoro- 
aster's invective in the A vesta were launched. 1 

Although situated in a region which was historically con- 
nected with Zoroaster's name, Tabriz yielded little for my 
researches in that respect : first, because Mohammedanism has 
obliterated the traces of Zoroastrianism ; second, because 
the winter season prevented my making investigations in 
the mountains, which possibly might have yielded some results. 
I was particularly anxious, for example, to ascend Mount 
Sahand, the mountain which possibly may be identical with 
Asnavand of the Avesta, but the heavy snows cut off all 
approach. Still more inaccessible at this time of year was 
Mount Savalan, near Ardabil, three or four days' journey dis- 
tant from Tabriz. This is the mountain which I think is to be 
identified with the ' Mount of the two Holy Communicants ' 
in the Avesta, where Zoroaster communed with Ormazd. A 
number of the Oriental writers, such as Ibn Haukal (tenth 
century), Kazvini (1263), Mirkhond (1474), and others, ex- 
pressly record the tradition that Zoroaster received a revela- 
tion from Ormazd on the heights of the Iranian Sinai and that 
he wrote the Avesta there. 2 Among these authorities is the 
author of the Suvar Akldlm Sab' ah, or c Outline of Countries,' 
who, writing in Persian, about A.D. 1400, attributes the tremen- 
dous snows around Ardabil, near which Mount Savalan rises, to 
a curse uttered against the people by Zoroaster because they 
rejected his faith. The fulfilment of this anathema seemed 
to me a veritable fact, for I had to abandon all hope of reach- 
ing Savalan, owing to the snow-bound roads, and to content 
myself with a distant view of this sacred mountain from the 
Caspian Sea when I returned to Baku in June. One of the 

1 Avesta, Ys. 51. 12 ; Vd. 8. 26, 27, fur Kunde des Morgenlandes, 12. 230- 
32, etc. ; see also Herodotus, History, 234, Wieii, 1898, and Brunnhofer, Vom 
1. 135. Pontus bis zum Indus, p. 182, Leipzig, 

2 See my Zoroaster, pp. 34, 195 ; 1893 ; see also Ibn Haukal, tr. Ouse- 
consult also Stackelberg, Persische ley, p. 173. 

Sagengeschichte, in Wiener Zeitschrift 



54 TABRIZ, THE RESIDENCE OF THE CROWN PRINCE 

native lords of the district about Tabriz, Anton Khan, an 
Armenian, gave me some description of the crater of Mount 
Savalan and of the hot springs which Kazvini mentions, but 
he said there is no tradition that he knew of, regarding the 
Fire- Worshippers in this region. 1 Another lord, Sadir Khan, 
a Persian, informed me about an ash-hillock of a fire-temple at 
Marand which I had missed on my way to Tabriz. 

My inquiries regarding inscriptions or sculptures in the 
vicinity of Tabriz did not result in eliciting any information, 2 
but I found that coins and gems are occasionally unearthed in 
the neighborhood. I purchased several specimens of coins 
dating from the Parthian and Sasanian periods, and a seal 
which is of considerable interest because of its age, as it is 
certainly to be attributed to the Achsemenian era. The seal 
is oval in shape, flat on the carved face and rounded at the 
back, and it measures one inch by three fourths of an inch 
(twenty -five centimeters by twenty centimeters). The stone 
is a blue chalcedony or sapphirine, which came into use during 
the early Persian period. It is carved with the figure of a 
king or warrior, slaying a monster with his dagger, somewhat 
after the manner of the sculptures at Persepolis. The work- 
ing out of the design, in my judgment, shows too much origi- 
nality to be a mere later imitation of this motive, and there is 
no evidence to show that the seal is a forgery. I am sup- 
ported in my view that it belongs to Achsemenian times by 
other scholars who have seen it, among them Dr. William 
Hayes Ward, of New York, an authority on seals and cylinders. 

The last day of my stay at Tabriz, which I should gladly 
have prolonged in order to enjoy the hospitality extended from 
many sides, was spent in making visits among friends in the 



l lt was still famed as a seat of Tabriz a fragment of an Ancient 

Magism in the tenth century of our Persian cuneiform inscription, but 

era, according to Ibn Haukal, tr. Ouse- it had been brought from Susa by M. 

ley, p. 173. de Morgan and, I believe, already 

8 1 saw at the French Consulate at published. 




SAFAR AWAITING DIRECTIONS 



AN ACH^EMENIAN SEAL 55 

European section, which forms a part of the Armenian quarter 
of the town. A final tour of the bazaars had also to be made 
to secure a stock of provisions to carry ' on the road,' and an 
extra supply of warm clothing had to be purchased, for the 
cold was still intense. The evening was passed at a dinner 
party given by my host and hostess, with a final good-by to 
American and European friends. 

At the eleventh hour, when the dinner was over (and it was 
literally eleven o'clock), I learned that the Armenian servant 
whom I had engaged, and to whom I had paid part of his 
month's wages in advance, had 'decided not to go on the 
journey.' This placed me in a great predicament: the horses 
had been hired and preparations made for an early start in the 
morning, so that postponement even for a day meant a serious 
change in my plans. A ray of light, however, shone through 
the darkness. I bethought myself of a young Persian, named 
Safar Adilbeg, a convert from Mohammedanism, whom I had 
noticed working about the mission grounds. I had been at- 
tracted by his honest face and demeanor, and after a hurried 
conference with my host and mission friends, to ask if the 
young man might be allowed to go, I received hearty ap- 
proval on their part, and Safar was aroused from bed to 
be questioned on the subject. He accepted the proposal 
at once, hesitating only lest his lack of experience should 
disqualify him. I felt sure, however, that he would fill the 
post well, for I was convinced of his merit, and we struck 
a bargain on the spot. My confidence was rightly placed, 
and although I used occasionally to wonder whether some 
serious blemish in his character might not develop, it never 
did, and I sometimes amusingly thought that wings would 
sprout and that his name would have to be changed to 
Raphael, after the story in Tobit. His true worth grew 
more and more in my esteem as time went on, and I am 
glad to be able to add that he has since then happily 
realized the ambition of his life in studying medicine at 



56 TABRIZ, THE RESIDENCE OF THE CROWN PRINCE 

Teheran, in order to become a physician and practise among 
his people. 

Before proceeding with the narrative of my journey after 
leaving Tabriz it may be advisable in the next chapter to give 
some account of Zoroaster and the A vesta, as these names have 
often been mentioned already and will frequently be referred 
to hereafter. 



CHAPTER VII 
ZOROASTER AND THE AVESTA 

'At whose birth and growth all creatures of the holy creation cried, " Hail ! " ' 

AVESTA, Yasht 13. 93. 

' TELL me how it comes to pass,' says one of the fathers of 
the early Christian church, ' that the majority of people know 
nothing more about Zoroaster than the name.' And yet there 
is a tradition that the wise men who came from the East to 
worship at the manger cradle in Bethlehem were led to under- 
take their pious journey ' by reason of a prophecy of Zoro- 
aster.' The name, moreover, of this forerunner of the Magi 
has been used in literature of later times as a synonym for 
wisdom. To Byron, Zoroaster was a 4 sage ' ; to Shelley he 
appeared as 4 the Magus ' or as 4 Earth's dead child ' ; and the 
German writer Nietzsche chose to veil his recently published 
thoughts under the title ' Thus Spake Zarathushtra.' It is 
the more interesting to know something about the life and 
character of this Persian lawgiver and philosopher of old, this 
religious teacher of ancient Iran, because much has been added 
in the last few years to our knowledge of Zoroaster as a his- 
torical personage a man to whom we may perhaps be justified 
in assigning, indirectly at least, a place in the line of prophets 
that have been since the world began. 1 

In the early dawn of the seventh century before Christ he 
appears as a star on the horizon, remotely a heathen herald of 
the Christian day to come. He comes as an elder contemporary 

1 lam indebted to Mr. J. B. Walker 349-357, New York, 1900. For a 
for permission to reprint with additions detailed life of the teacher see my 
and alterations my article on Zoroaster Zoroaster, the Prophet of Ancient Iran, 
in the Cosmopolitan Magazine, 28. New York, 1899. 

67 



58 ZOROASTER AND THE A VEST A 

of the Grecian sages Thales and Solon. If our calculation 
be right, he must still have been living when the Jews were 
carried up into captivity at Babylon. His birthplace was 
the district to the west or southwest of the Caspian Sea, not far, 
apparently, from the city of Urumiah, which we are approach- 
ing on our journey, and the scene of a part of his early mis- 
sionary preaching and teaching was the territory south of this 
very sea. The crest of Mount Alborz gilded with God's eternal 
sunrise for Alborz is a holy mountain in the Avesta may 
have suggested the theme for more than one inspired discourse. 
The wellsprings of blazing oil and phenomena of igneous origin 
familiar in the volcanic regions of Iran may have seemed to 
him a symbol of the source of the Fire Divine. 

Dressed in white flowing robes we may picture him preach- 
ing before his people. The priestly vestments of the Parsi 
dastur to-day, and the Gheber mantle and belt of the fire- 
worshipper in Lalla Rookh are lineal descendants, no doubt, 
of the ancient Median garb which he wore. Herodotus tells us 
that the Persians took their style of dress from the Medes 
because they thought it handsomer than their own. The form 
of worship and manner of chanting the ritual which the great 
historian describes is largely kept up in the Zoroastrian religion 
to-day. In speaking, Zoroaster used a language akin to the 
ancient Sanskrit, but more abounding in long final vowels, as he 
lifted up his voice in exhortation of the masses or sang the 
praises of the god Ahura Mazda. 

In his youth, so far as we can gather, he must have been 
reared in a state of society that showed marked and paradoxical 
extremes. From the ancient records we may judge that the cul- 
ture, such as it was in those early times, was offset by extraor- 
dinary crudeness and barbarism ; and Media in some respects 
has changed but little since then. Zoroaster was well ac- 
quainted, we may believe, with the civilization of the ancient 
cities, but better acquainted with the gross ignorance and base 
superstition of unlettered country life. These two widely 




A PERSIAN PAGE 



THE" 

UNIVERSITY 

* OF 
AI 



THE PROPHET AND HIS TIME 59 

separated stages may be seen in the Avesta itself. The 
wattled hut of the peasant or the temporary habitation of 
the marauding nomad form a contrast to the 'palace of a 
thousand columns ' which is incidentally referred to in the 
sacred text. The rude hovels are still perpetuated in the mud 
dwellings of the poor ; and the ruins of Susa and of Persepolis, 
with their pillared courts, best seen in the restorations of 
Dieulafoy or of Perrot and Chipiez, bear witness to how noble 
the grander architecture of the Persians must have been. The 
ancient city, of Ecbatana, known as Achmetha in the Bible, 
possessed a pile of buildings of no mean order, if we may judge 
from the Father of History, who tells of its seven concentric 
walls, painted each in a different color and crowned by a 
citadel whose battlements shone with silver and gold. It is 
true that Zoroaster never mentions this city, but we are 
justified in regarding it as a type of the cities which he 
knew. 

In contrast to this possible approach to luxury we must place 
the other view ; for whatever may have been the higher civili- 
zation, or whatever there may have been of incipient culture, 
in Zoroaster's day, we have in his own words evidence enough 
of a prevailing density of superstition and of a mist of religious 
unbelief that hung like a pall over the benighted people whose 
eyes and ears he came to open and whose hearts and minds he 
came to illumine and enlighten. Messiahlike he appears, and 
the land of Iran rings with his clarion note of reform. He is 
born as one out of the fulness of time. He arises to revolu- 
tionize the religious thought of Iran, to stir the soul of Media 
and Bactria, and to form for the coming nation of Persia a 
creed that is to boast a Cyrus, ' the shepherd of the Lord,' and 
a Darius who shall give command for rebuilding the Temple of 
Jerusalem. The details of Zoroaster's life may be in a measure 
legendary, but behind them all we can see the figure of a great 
historic personage, whose actual existence we have no longer 
any reason to doubt. 



60 ZOROASTER AND THE A VESTA 

The year B.C. 660 was perhaps the date of Zoroaster's birth, 
although much uncertainty has prevailed on the subject and 
some scholars argue that he flourished a century or two earlier. 
The Avesta and the Middle Persian books known as the 
Pahlavi writings describe the prophecies that foretold his com- 
ing and the signs and wonders that heralded his entrance into the 
world. Some idea of the youth and personality of this future 
master may be gathered from these texts, which the reader may 
easily consult in translation in Max Miiller's series of Sacred 
Books of the East. In reading of his life we cannot help 
feeling that we are in the presence of a person whose character 
even in youthful years combines vigorous thought with specu- 
lative imagination. It is instinctive for such spirits to act as 
guides. We may be certain that Zarathushtra early heard the 
call that spoke to his heart and made him feel an individual 
fitness, a peculiar consecration, for his hallowed office. His 
person is sanctified ; his bearing is that of one who is to receive 
a weighty charge. 

The extravagant stories of plots against his life by sorcerers 
and demons, which the old books of Iran enjoy repeating, or of 
his disputes with perverted and crafty ministers of a false 
creed, under whose blinding influence even his own father lived, 
present a dark picture of a foul religion and a depraved priest- 
hood, which he felt himself destined to combat and overthrow. 
4 Tell me,' he later says in one of his Gathas, or Psalms, in 
which he alludes to the false priests and devils, 4 tell me truly, O 
Lord, have such demons ever been good rulers ? ' It is pre- 
cisely these who, to quote his own words used on another 
occasion, 'have united themselves with power for the purpose 
of destroying the life of man by their evil deeds ; but their 
own soul and their own conscience will make them howl 
when they come to the Bridge of Judgment, to be inmates 
forever and ever of the House of Falsehood [i.e. hell].' And 
again he exclaims, ' Tell me truly how we shall banish False- 
hood from ourselves even unto those who, full of unbelief, take 



THE MISSION OF THE PROPHET 61 

no thought which accords with Righteousness, nor have felt 
delight in the communion of Good Thought.' 

On the other hand, a moment after these anathemas or im- 
passioned utterances against the wicked, we find evidence of 
Zoroaster's mild-heartedness and of his loving-kindness toward 
the good. In the liberality of his spirit, if we are to believe 
tradition, he was so broad-minded as to show a willingness to 
pick out and adopt what was noble even in the corrupt exist- 
ing faith. 

But the path for full inspiration must first be prepared, and 
the way to enlightenment must previously be laid open, before 
the revelation comes to a man of Zarathushtra's nature. Tra- 
dition says that Zoroaster retired from the world when he came 
of age and that he lived for some years upon a remote moun- 
tain in the silence of the forest or taking shelter in a lonely 
cave. In this connection I have already referred to Mount 
Savalan (p. 53), and in Mount Sahand there is a cavernous 
vault which is said to be Zoroaster's cave, and a subterranean 
chamber near Maraghah, with a fire-altar, is attributed to his 
worship. 1 It was the solemn stillness of such surroundings 
that lifted his soul into direct communion with God. A divine 
vision is accorded him, apparently on the occasion of some 
religious conference, and at the age of thirty, after leaving 
the Iranian Sinai, he is prepared to teach a new law. ' Right- 
eousness is the best good' ashem vohu vahishtem astl is 
his watchword ; but he finds little fruitful soil for his theme. 
He wanders over the land of Iran and through the territory 
that is now Afghanistan, and even tarries for a time in the 
country of Turan. But it is to deaf ears that he preaches, and 
his inspiration seems almost destined to be in vain. 

The rulers harden their hearts before the newly inspired 
prophet ; the people fail to accept the message of the god 

1 Mr. Arter, of Ziegler & Co., Te- For the cave at Maraghah consult Ker 
heran, told me that there is some such Porter, Travels, 2. 495-497. See also 
story about a cave in Mount Sahand. p. 103, below. 



62 ZOROASTER AND THE AVESTA 

Ahura Mazda. And yet Ahura Mazda, or Ormazd, is the 
4 Lord Wisdom, the Sovereign Knowledge.' It is doubtless 
true, moreover, that many persons were deterred from adopt- 
ing the faith because of the doctrine of next-of-kin marriage, 
which Zarathushtra seems to have upheld because he felt that 
this would serve as a means of preserving in its purity and in- 
tegrity the community of faithful adherents and advancing the 
struggling creed of his church militant. 

For ten years, dervishlike, he is a wanderer. This we know 
also from the tone of dejection which still echoes in some of the 
Zoroastrian Psalms. In his peregrinations he appears to have 
found his way once more to the region of the Caspian Sea. 
The darkness of these sad years is illumined, however, by vi- 
sions which help to make strong his faith and to give form to 
his religious system and creed. Seven times the mysteries of 
heaven are revealed to his transported soul; and a number of 
the places where these visions were beheld may be identified 
with a fair degree of certainty. Most of them are to be 
located in Azarbaijan not far from Lake Urumiah. He con- 
verses with Ormazd and is also privileged to interview the 
Archangels of Good Thought (Vohu Manah), Best Right- 
eousness (Asha Vahishta), Wished-for Kingdom (Khshathra 
Vairya), Holy Harmony (Spenta Armaiti, guardian spirit 
of the earth), Saving Health (Haurvatat), and Immortality 
(Ameretat). Such are the names of the Persian hierarchy of 
Amshaspands, and these allegorical figures or personified 
abstractions stand in waiting about the throne of Ahura 
Mazda with a company of attendant angels. From these 
divine beings Zarathushtra receives commands and injunc- 
tions which he is to convey to mankind. They inculcate the 
doctrine of purity of body as well as of soul ; they enjoin the 
care of useful animals, especially the cow and the dog ; they 
emphasize the necessity of keeping the earth, the fire, and 
the water undefiled; and from several of their ordinances 
we can see that Zoroaster was a civil reformer as well as a 




IDEALIZED PORTRAIT FROM A SCULPTURE SUPPOSED TO REPRESENT ZOROASTER 



<t 

OF THE 

UNIVERSITY 




REVELATION AND TEMPTATION 63 

spiritual guide. Foremost among the commandments is the 
abhorrence of falsehood, the universal obligation to speak the 
truth. This is one of the most fundamental of the ethical 
tenets which form the basis of the entire ancient Persian reli- 
gious system. 

A revelation of the future is also vouchsafed to the soul 
of the Prophet during his sojourn in the celestial council, 
and one of the most precious boons which it is the privilege 
of his rapt spirit to receive in these moments of ecstasy 
is a premonition of the resurrection and of the future life. 
Unlike the Mohammedan visions of ethereal bliss, there is no 
jarring note of pleasures of a physical kind to mar the harmony 
and spirituality of this glimpse into the world beyond. But 
before the ecstatic Messenger is allowed to return to the world 
of material things, one word of warning is given to guard him 
against the guile and deceit of the Spiritual Enemy, Angra 
Mainyu, or Ahriman, as the devil is called. At this moment, 
as he turns from the dazzling splendor of heaven, a glimpse of 
the darkness, filth, stench, and torment of the ' Worst World ' 
is disclosed. There in the murky depths of hell, with mocking 
howls and ribald jeers, huddle together and cower the vile crew 
of the archfiends and whole legions of demons, or 'devs,' as 
they are still named in Persian. 

Nor is this caution any too timely, for at once upon the 
hallowed Seer's return to earth there occurs the temptation 
by Ahriman. Like the wily Mara seeking to beguile the 
newly enlightened Buddha, or the tempter Satan striving 
to betray the Saviour of mankind, the maleficent Ahriman 
endeavors to cause the righteous Zarathushtra 'to renounce 
the good religion of the worshippers of Mazda.' This moment 
is a crisis ; it is one of the turning-points in the history of the 
faith. The foul fiend is repulsed and vanquished, and the 
victorious upholder of righteousness chants a kind of Te 
Deum yathd ahu vairyo as a paean of his triumph. But 
he has to face many discouragements in his work ; only one 



64 ZOROASTER AND THE A VEST A 

convert to his faith is won during the first ten years of his 
preaching. This is his own cousin Maidhyoi-Maonha, the St. 
John of the Religion. In the twelfth year of his mission, how- 
ever, came an achievement which made a crowning glory to 
his career ; this was the conversion of King Vishtaspa, who 
became the Constantine of the faith. 

Exactly who this Vishtaspa, or Gushtasp, was, we cannot 
with certainty say. His name is the same as that of Hystaspes, 
the father of King Darius, but there is no convincing ground 
for identifying him with that personage. Whether he was a 
vassal king in Media itself or a monarch in eastern Iran, Bac- 
tria, or more probably the region corresponding to Afghanistan 
and Persian Seistan to-day, belongs to scholars to discuss. 1 It 
suffices here simply to present this pious ruler, whom the Avesta 
portrays as the nonpareil of kings, and to recall how his strong 
arm made Zoroaster's religion current in the Province of the 
Sun. 

Vishtaspa is converted only after a long struggle, hesitancy, 
and deliberation ; but when once convinced, he exhibits all the 
zealous enthusiasm that is characteristic of a new convert. 
His queen, Hutaosa, whose name at least recalls the name of 
Atossa in Persian history, likewise accepts the faith and joins 
the struggling church. The nobles of the court follow the 
high example. Zoroaster's own family becomes a sharer in the 
royal favor. His third wife for he was married three times 
is a sister of the king's grand vizir. His favorite daughter 
in turn is given in marriage to the other chief councillor of the 
sovereign. A number of relatives of the priest unite in the 
confession of the faith. Converts become many. The spark 
of religious enthusiasm kindled in the palace spreads like a 
mighty flame throughout the land. The people press to hear 
Zarathushtra speak. We can still listen to the verses of his 
Gathas (Psalms) that served as texts for sermons. Here is the 
opening of one of his discourses, in which he tells of the two 

1 See my Zoroaster, pp. 205-225. 



TWO ZOROASTRIAN PSALMS 65 

great opposing principles of Good and of Evil. The verses are 
rhythmical, and in translating we may follow the order of the 

lines : 

' Now shall I preach, and do ye give ear and hear, 
Ye who hither press from near and from afar ! 
Now mark him all, for the Devil has been disclosed 
And nevermore shall he, Vile Teacher, the world destroy, 
Wicked Avower, he, of a sinful faith with his tongue. 

4 Now shall I preach of the world's Two Primal Spirits, 
The Holier One of which did thus address the Evil : 
" Neither do our thoughts, our teachings, nor our minds, 
Wishes, nor words, nor works, no, nor our religions, 
Nor do our souls agree in anything at all." ' l 

On another occasion the same theme is taken up again by 
Zoroaster in a Psalm (Gatha) which may be termed an Iranian 
Sermon on the Mount. In this the priest bids his listeners to 
be mindful of these Two Spirits which divide the universe 
between them. People must not be deceived into making 
a false choice, as the demons have done ; but they must 
follow spiritual guidance, so as to be on the right side when 
the judgment shall come ; for annihilation shall then overtake 
Falsehood (Druj, Satan) and ruin shall attend upon the demons 
and upon all who ally themselves with them. Let every man, 
therefore, seek to make the world prepared. In the eighth 
verse the Prophet breaks out into a fervent expression of hope 
of a regenerate world and a new kingdom : 

* Therefore at the time when the retribution of the sinful shall come to pass 
Good Thought [Vohu Manah] shall dispense Thy Kingdom 

To the joy of those who deliver over Falsehood [Druj] into the hand of 
Righteousness [Ash a]. 

1 And so may we be such as make the world renewed, 
And do ye, Lord Mazda and Righteousness, bear your company, 
That our thoughts may wholly be where wisdom is abiding. 

* For at the [final] Dispensation the blow of annihilation of Falsehood 

[Druj] shall come to pass, 

But those who share in a Good Report shall speedily unite together 
In the happy abode of Good Thought, of Mazda, and of Righteousness. 

i Ys. 45. 1-2. 



66 ZOROASTER AND THE AVESTA 

'If, O ye men, ye mark the doctrines which Mazda gave, 
And [mark] the weal and the woe namely, the long torment of the 

wicked 

And the welfare of the righteous then in accordance with these [doc- 
trines] there will be happiness hereafter.' l 

Light begins to dawn upon the people. If good and evil, 
god and devil, are in constant warfare with each other, what 
is to bring about the ultimate solution of the conflict, what is 
to give the victory to Ormazd and to put an end to the strife ? 
It is man. Man, a free agent, shall solve the problem by elect- 
ing right and choosing goodness, if he follow Zoroaster's lead, 
and as his reward he shall win joys eternal at the resurrec- 
tion, 'when the dead again shall rise up, the quick be made 
immortal, and the world, as desired, made perfect.' 

But, exalted and spiritual as these religious tenets were, 
not everything in the faith occupied so lofty, so perfect, so 
transcendental a plane. It is difficult for a people to maintain 
the high and ideal level of the leader. The knowing and en- 
lightened may accept advanced theological doctrines, but the 
masses require that which is more practical, more tangible. 
Temporal considerations and material things cannot be left 
out of sight by any reformer when he is founding his religion. 
There is evidence of concessions being made in Zoroastrianism 
to previous religious views or time-honored practices. The 
glorification of the sun, moon, and stars as part of God's uni- 
verse could not be omitted from the popular creed. The 
ancient divinity Mithra, .an embodiment of light and of the 
sun, as we know from Tom Moore's Fire- War shippers, is 
canonized to stand beside Ormazd. The elements earth, fire, 
and water are idealized as manifestations of purity. Xerxes 
did pious homage to the vegetable kingdom when he decked 
the plane-tree as a solemn rite on his way to Greece. Matters 
and details of this kind were unquestionably recognized by 
Zoroaster himself as elements to be accepted into his creed, as 

1 Ys. so. 8-11. 



LATE HISTORY OF THE FAITH 67 

he labored and taught, preached and converted, counselled and 
encouraged, during a ministry that lasted more than thirty 
years. His was a long life. Forty years of age or over when 
he first converted Vishtaspa, he lived until he was seventy- 
seven. Death came to him by violence in the holy wars which 
arose as a consequence of crusading for the faith, about the 
year B.C. 583. Such at least is the tradition, or legend, for 
there is a story that he was slain by the fanatical Turanians 
when they invaded Iran. 

But the faith did not perish with the founder. We know 
from history how Zoroaster's creed was able to withstand the 
mighty shock given to it three centuries later when Alexander 
invaded Iran and at the request of the frail but beautiful Thais 
allowed the palace at Persepolis to be burned. It is claimed 
that the Avesta perished in the flames ; but the Magian priest 
still held in memory the sacred texts, clung to the creed, and 
upheld the tottering rites. The faith revived once more and 
regained its pristine glory and flourished at the very time 
when the Roman wars with Persia fill the pages of history 
and the Zoroastrian heresy of Manichseism threatens to shake 
the Christian church. The final blow, however, came to 
Zoroastrianism in the sixth century from Islam. From that 
moment Persia practically adopted Mohammedanism and ceased 
to be Zoroastrian. Only a handful remained faithful to the 
old creed and were destined to endure countless sufferings from 
persecution in their native land. Another band, equally stub- 
born, refused to be converted, and chose exile in India, finding 
a place of refuge in Bombay and its vicinity, and thus becoming 
the ancestors of the present Parsis. Owing to their favorable 
surroundings among the Hindus they have prospered more 
than their Persian brethren, and like the latter they have 
remained genuine followers of the old-time faith of Zoroaster. 

It is these two communities that have preserved for us 
until to-day the remnants of the ancient Zoroastrian scriptures, 
the Avesta and the Pahlavi books. In its present form the 



68 ZOROASTER AND THE AVESTA 

Avesta is only a fragment of the original Zoroastrian scriptures. 
Tradition tells of twenty-one books, or a million of verses, 
composed by Zarathushtra and inscribed in letters of gold 
at the order of King Gushtasp, the patron of the faith. 
The destruction of the two archetype copies, the one at Perse- 
polis, the other at Samarkand, is attributed to Alexander the 
Great. 1 

Only imperfect remnants of these originals have been 
preserved ; in compass they would equal about one tenth of 
our Bible, and as in the latter we may recognize several sub- 
divisions with reference to their contents. Most important is 
the Yasna, lit. ' sacrifice,' a liturgical work, comprising also 
the Grdthds, or Psalms of Zoroaster, the most sacred part of the 
Avesta, and supplemented by a series of minor litanies known 
as the Vlsperad, 4 all the lords,' both works being used in the 
ritual and forming a sort of manual of devotion which corre- 
sponds to our prayer-book. Second in interest are the twenty- 
one Yashts, lit. 'praises,' a collection of metrical hymns in 
praise of the ancient divinities and mythical heroes of the reli- 
gion. Third, and interesting in comparison with the Pentateuch, 
is a priestly code comprising twenty-one chapters and entitled 
Vendlddd, c law against the demons,' a series of sacerdotal rules 
for the purification, and some miscellaneous matter of a legendary 
character. The remaining portion of our present Avesta is 
composed of minor prayers, invocations, and miscellaneous 
fragments. 2 

The loss of some parts of the Avesta is made up for in part 
by versions or summaries in the Pahlavi language of the Middle 

1 See p. 306, below, and my article Books of the East, Oxford, 1880-1887. 
Some Additional Data on Zoroaster, in There are French translations by Dar- 
the volume Orientalische Studien in mesteter, Le Zend-Avesta, Paris, 1892- 
honor of Professor Nbldeke, pp. 1031- 1894, 3 vols., and by de Harlez, Avesta, 
1038, Strassburg, 1906. Paris, 1881, a German one by Spiegel, 

2 The best English translation of the and reference will be made hereafter 
Avesta is by Darmesteter and Mills, to German renderings of selections 
Zend-Avesta, 3 vols., in the Sacred by Geldner and by Earth olomae. 



THE SACRED BOOKS 69 

Persian Empire, or supplemented again by later Persian writ- 
ings on Zoroastrian subjects or by traditions which the priests 
have preserved. Most important among the Pahlavi texts is a 
work entitled Biindahislin, c Original Creation,' a sort of Iranian 
Genesis founded on one of the original books of the Avesta 
which has been lost. 1 

An acquaintance with the ancient Zoroastrian literature and 
its language, and a familiarity with the history of the people 
that have preserved it, made me anxious to make the journey 
around Lake Urumiah, which Zoroaster himself must have 
made, and I used my own volume on the Prophet's life as a 
sort of handbook for the journey in laying out my route 
from Tabriz along the shores of the historic lake. 

1 Most of the Pahlavi books have vols. , in the Sacred Books of the East t 
been translated into English or sum- Oxford, 1880-1897. 
marized by West, Pahlavi Texts, 5 



CHAPTER VIII 
AROUND LAKE URUMIAH 

' Wer den Dichter will verstehen 
Muss in Dichters Lande gehen.' 

GOETHE, Westostlicher Divan. 

IN spite of rumors of deep snow I ventured to undertake 
the journey around Lake Urumiah from Tabriz to the city of 
Urumiah by wagon. I was warned in advance by one of my 
friends that if I tried to drive, I should be sure to wish I had 
ridden, and if I started on horseback, I should be certain to 
regret not having gone by carriage. Events proved the truth 
of his words. The ' roads ' were in a vile condition, and the 
journey, which ordinarily occupies three or four days, took 
me six. 

For the first forty-eight hours I had the companionship of 
two Persians whom we overtook on the road ; they were also 
driving. One of them was a native of the village of Khosrova, 
near Dilman, northwest of the lake, and was on his way home 
from Meshad in eastern Persia. 1 The other was connected 
with the bank at Teheran. The latter was a particularly 
fine-looking fellow, with handsome eyes, clear-cut features, and 
a tall, well-developed frame. He wore on his head a hood, the 
ends of which formed a scarf to wrap about his face as a pro- 
tection against the cold. This made him look like a veritable 

1 Is it possible that Khosrova pre- or does it owe its name to the later 
serves a lingering reminiscence of the Sasanian king Khosra Parviz ? I find 
Avestan king Haosravah, who sacri- that this suggestion has been previously 
ficed ' on the other side of Lake Chae- made by Darmesteter, Le ZA. 2. 632, 
chista' (Lake Urumiah), Yt. 5. 49, n. 92. 

70 



THE SNOWS OF AZARBAIJAN 71 

portrait of Darius Codomannus at the battle of Issus, a repro- 
duction of which has been given above. 

The temperature must have registered nearly zero during the 
first two days of the trip, although I never had the courage to 
consult the thermometer which was stowed away somewhere in 
my baggage. In the daytime I was compelled to wear my sleep- 
ing-jacket over my head to shield my frost-bitten face from the 
congealing wind, and as evening fell I muffled a bathrobe over 
this to add some warmth. I envied any one whose lot it might 
be to make the journey in midsummer instead of in winter, and 
I understood why the Avesta regarded winter as ' the work of 
demons ' and said that it was created by Ahriman as a blight to 
mar the perfection of Airyana Vaejah, the Azarbaijan of to-day, 
which otherwise would have been a paradise. 1 In this land the 
Vendidad says ' there are ten months of winter and two months 
of summer.' 2 A gloss, it is true, changes the text to 'five 
months of winter and seven months of summer,' but judging 
from my own discomfort (for March seemed in the Avestan 
words to be the very 4 heart of winter,' zimahe zaredhaem), I 
felt inclined to agree with the original reading. My discom- 
fort was tempered, however, by the thought that the region 
through which I was travelling had probably once been trav- 
ersed by Zoroaster, and this added a zest to my observations 
en route as the trail meandered forward along the northern 
shore of the lake. 

Lake Urumiah is the largest body of water in Persia, 
although not quite so large as our Great Salt Lake in Utah, 
which is about seventy-five miles long and from thirty to fifty 
broad, the Persian lake being about eighty miles in length 
and averaging twenty-four miles in breadth. Both of these 
bodies of salt water lie about four thousand feet above the level 

1 See Vd. 1. 2, zyymca daevo- trast between summer and winter in 

ddtdm. The heat near the northern Azarbaijan' (Wilson, Persian Life, 

shore of Lake Urumiah is corre- p. 83). 
spondingly great in midsummer : ' no 2 Vd. 1. 3. 

place shows better than this the con- 



72 AROUND LAKE U RUM I AH 

of the sea, and neither has any outlet. The waters of both 
are intensely saline and vary considerably in volume accord- 
ing to the condition of the mountain streams that feed them ; 
but the average depth in each case is considerably less than 
twenty feet. Other resemblances might be pointed out, but 
enough have been indicated to show the parallel between the 
two. 

About the shores of Lake Urumiah there are level plains, 
sometimes covering an area of many square miles, such as the 
great Plain of Urmi on the western border, and there are 
high mountains lying beyond them on all sides of the lake. 
These sometimes thrust their spurs down to the very edge of 
the water, as does the ridge of the Karabagh mountain on 
the northwest (six thousand feet high), and the offshoots of the 
great mountain of Sahand on the east (over eleven thousand 
feet). A few small islands dot the surface of the lake toward 
the south-central part, and from the middle of the eastern 
shore the mountain peninsula of Shahi, or Shah Kuh, juts out. 
This tongue of land was once an island twenty-five miles in 
circumference, but it has become a part of the mainland, 
because the lake has lowered somewhat. 1 Of recent years, 
however, the volume of water has tended again to increase, so 
that considerable fluctuations in the outline of the shores are 
still taking place. To-day there is no navigation on Lake 
Urumiah except what is carried on by means of clumsy scows 
propelled by primitive oars and sails. 

We can trace the history of Lake Urumiah far back into 
antiquity, even to Zoroaster's time and still earlier. The 
region was familiar to the Assyrian kings as the scene of some 
of their active campaigns, and the lake appears in their inscrip- 

1 Yakut, who passed by Lake Uru- island in his day (1810-1830) ; cf. 

miah twice (A.H. 612, 617 = A. D. 1215, Curzon, Persia, 1. 532. Similarly, 

1220), speaks of the mountain island Perkins (1843) calls it an island, 

in the midst of the lake (see Barbier de * which is much of the year a penin- 

Meynard, Diet. geog. p. 86), and Sir sula' (Eight Years in Persia, p. 

J. Macdonald Kinneir reports it as an 170). 



DESCRIPTION OF THE LAKE 



73 



tions as the 'lake of the land Nairi.' 1 It was known in the 
Avesta by the name of Chaechasta, which the Arab geogra- 
phers corrupted through Chiz into Shiz. 2 The Avesta calls it 
4 deep ' (jafra), which may be an appropriate epithet accord- 
ing to the ideas of the ancient Persians, who were unacquainted 
with our great lakes; but its average depth hardly exceeds 
fifteen feet. The characteristic Avestan attribute, however, 
which is applied likewise to the Caspian, is urvdpa, uruydpa, 
4 whose water is salt.' 3 Lake Urumiah is so briny that fish are 
not found in its waters, and the only occupant appears to be a 
small crustacean. The Pahlavi treatise Bundahishn, which 
several times mentions Lake Chechast, expressly states that 
'there is nothing whatever living in it.' 4 Ibn Haukal, in the 
tenth century, makes a similar statement. 5 As for the modern 



1 So Schrader, Die Namen der Meere 
in den assyrischen Inschriften, in Abh. 
d. AJcad. d. Wiss. zu Berlin, 1877, pp. 
184-193. For the relations between 
Lake Urumiah and Lake Van, see 
Streck, Armenien, Kurdistan und West- 
persien, in Zt. f. Assyriologie, 13. 11. 
The fact that the region of the lake 
and city of Urumiah is alluded to in 
the Assyrian inscriptions is accepted 
by Ward, Notes on Oriental Antiqui- 
ties in American Journal of Archce- 
ology, 6. 286, and by others. We might 
be tempted to seek the name of Uru- 
miah, or Urmi, in the Assyrian Urume, 
but see Streck, op. cit. pp. 23-24. 

2 See p. 131. The actual Avestan 
form is Vairi Caecasta (or Caetista), 
Yt. 5. 49 ; Ny. 5. 5 ; Sir. 2. 9. On the 
name Slz (Clz) see my Zoroaster, 
pp. 195, 197, 201-202, 204. 

8 Such seems to be the force of Av. 
urvdpa, uruydpa, as first pointed out 
by Darmesteter, Etudes Iraniennes, 
2. 179. See also Geldner, Vedische 
Studien, 2. 270, Stuttgart, 1897, de- 
spite Bartholomae, Altiranisches Wor- 



terbuch, p. 404, Strassburg, 1905. 
The Pahlavi tradition sees in this epi- 
thet 'warm water,' garmdb, garmid. 
Shall we venture to compare Avestan 
Uru-dpa, Uruy-dpa, 'having salt (or 
warm) water,' with the modern name 
Ur-mi, Ur-mia(Ji), 'Urumiah,' which 
the natives commonly understood as 
' place of water ' (the last element be- 
ing the Semitic word for water) ? On 
Pahlavi tfetast see also Rosenberg, 
Livre de Zoroastre, pp. xxviii, 74. 

4 Bd. 22. 2 ; cf. 17. 7 ; 23. 8 ; and 
Bahman Yasht, 3. 10. 

5 Ibn Haukal, tr. Ouseley, p. 162 : 
' There is a lake in Azarbaijan called 
the Lake of Armiah (Urumiah) ; the 
water is salt or bitter and contains not 
any living creature. All round this 
lake are villages and buildings ; from 
the lake to Maraghah is a distance of 
three farsang ; to Armi (Urmi, Uru- 
miah), two farsang. The length of 
this lake is five days' journey by land ; 
and by water, with a fair wind, a per- 
son may traverse it in the space of one 
night.' 



74 AROUND LAKE URUMIAH 

name of the lake, the natives generally term it Dariah-i Shahi, 
or 4 Royal Sea,' after the mountain peninsula of Shahi, or Shah 
Kuh, mentioned above. The early Greek geographer Strabo 
mentions it under the name of Spauta (written ^jravra in 
the Mss.), which is supposed to be an error for Kapauta, the 
Persian Kabuda, lit. 4 blue, cerulean ' ; 1 but since my return 
to America I heard two natives of Urumiah apply the name 
'.Spaut' to the lake, although I did not hear it so called 
while I was in Azarbaijan. 2 The Arab writer Masudi uses 
the name Kabudhdn, saying the lake is so called after the vil- 
lage of Kabodhan on an island in the lake ; 3 but the attribute 
'cerulean' is more probably due to the color of the water, 
which presents a succession of blues melting into purples, 
mingled with ultramarine and green hues which were all 
the more conspicuous against a background of snowy moun- 
tains and a shore whitened with crystals of salt due to the 
incrustation of saline deposits. The old name Shiz, or Chae- 
chasta, seems absolutely to have disappeared, as I could find no 
trace or reminiscence of it among the people, although I 
inquired again and again during the two weeks or more which 
I spent in the vicinity of the lake. 4 

The heavy floods and inundations through which we had to 
make our way during a part of the journey around Lake Uru- 
miah made the Avestan word voighna, 'inundation, over- 
whelming flood, deluge,' a living reality. 6 The disasters and 
misery which follow in the wake of these winter freshets are 
as evident to-day as they were in ages past. Twice we had to 
descend to help the horses, which had been carried off their feet 

1 So, for example, Marquart, Eran- 86, who cites the authority of Saint- 
Sahr, p. 143. Martin, Mem. sur VArmenie, 1. 56 seq. 

2 They were Nestorians, and the Compare also Bittner, Der Kurdengau 
designation may be Syriac. Ushnuje und die Stadt Urumije, in 

8 Marquart, Erdnsahr, p. 143. Sb. Akad. Wiss. 133, Abh L 3, pp. 1-97, 

4 For additional details regarding Wien, 1895 ; Marquart, Erdnsahr, p. 

Lake Urumiah see Barbier de Mey- 143 ; Curzon, Persia, 1. 532-534. 

nard, Diet. geog. de la Perse, pp. 85- 6 Vd. 1. 3 ; Ys. 57. 14. 



SHEEP AND BIRDS 75 

by the treacherous caving-in. of the bank of a stream. It was 
often a matter of the greatest difficulty to find a safe ford, but 
the natives seemed to have a remarkable instinct for discover- 
ing crossings. There were few bridges, if any, and I could 
understand why it was a pious act, according to the Zoroas- 
trian faith, to build a bridge, and why this practice was enjoined 
also upon the worshippers of Ormazd as one means of penance 
for expiating the sin of having killed an otter a sacred 
animal in the eyes of Zoroaster. 1 

Wherever there was a trace of stubbly grass, large flocks of 
sheep and goats were cropping it. The color of most of these 
was black or brown ; the few white sheep among the number 
were conspicuous in contrast to the others. This made clear 
to me a Zoroastrian simile which had always been a puzzling 
one. The Pahlavi book of the Bundahishn, in describing the 
Day of Judgment, says that at the last day, when the souls are 
gathered together in a great assembly after the resurrection, 
4 a wicked man shall be as conspicuous in that assembly as 
a white sheep among the black.' 2 The white sheep was 
certainly the marked one in the flocks I saw on this journey. 

My observations of some of the birds of Azarbaijan, the 
lark, thrush, and long-tailed magpie, were only incidental, but 
I took more careful note of the crow, raven, eagle, and vulture, 
because those are especially mentioned in the Avesta. The 
crow is seen everywhere. The raven is still more plentiful 
and flies in enormous flocks. On one occasion I counted over 
a thousand of these birds in a field near Lake Urumiah. So 
far as my observations went, the raven appears to be more 
common in northern Iran than it is in the south, but this fact 
may have been due to migration at the time when I made my 
notes, or to some accidental cause. The bird is large in size, 

1 Vd. 14. 16. 1868; Westergaard, Bund. p. 73, Copen- 

2 Bd. 30. 10, anddr an anjuman hagen, .1851; Unvalla, Bund. p. 85, 
darvand aetuno petak clgun gospand Bombay, 1897; and the translation of 
i spet andar an sldk bet. See the text West, Pahlavi Texts, in Sacred Books 
of Justi, Bundehesh, p. 73, Leipzig, of the East, 5. 123, Oxford, 1880. 



76 AROUND LAKE URUMIAH 

with a glossy bill and sleek wings. I wondered whether the 
martial bird vdreghna, vdrenjina in the Avesta, a feather of 
whose wing was used by the warrior as an amulet, might possibly 
be identified with the raven. 1 The raven was the bird of battle 
among the Anglo-Saxons and other Germanic peoples, and 
the symbol upon the crest of Verethraghna, the Iranian Mars, 
on the coins of the Indo-Scythian kings, may be the raven. 2 
The only other bird that would answer well to the description 
of the vdrenjina is the peregrine falcon, which is noted for its 
swiftness of flight and habits of prey. It can hardly be the 
ordinary falcon, as that was a bird of sport in Persia, as else- 
where; nor was it the eagle, for that is called saena in the 
Avesta. 3 The eagle, however, abounds in the mountainous 
regions of Iran, and I could understand why Xenophon should 
have represented Cyrus as taking omens from eagles. 4 More- 
over, as I watched the soaring flight of this king of birds above 
the peaks, I was forcibly reminded of the name of that lofty 
range in eastern Iran, Updiri-saena, 'Above-the-Eagle,' whose 
height, according to the Avesta, surmounted even the eagle's 
flight. 6 The vulture (to which bird as well as to the dog the 
Magians used to throw the bodies of their dead to be torn) is 
found in Azarbaijan, but in numbers it did not seem so plenti- 

1 Yt. 14. 19-22 ; 14. 35-40. I find ed. Darab D. P. Sanjana, pp. 16-17, 
that this view has the support of the Bombay, 1896, although Darab San- 
authority of Darab, as cited by Justi, jana (loc. cit.} calls it an 'eagle' 
Handbuch der Zendsprache, s.v., and (reading luk), and Peshotanji, Nol- 
of Tir Andaz and Darmesteter, Le ZA. deke, and Antia (the latter, Kdrnd- 
2. 566, n. 29 (which Bartholomae, mak, p. 16, Bombay, 1900) interpret 
Air. Wb. pp. 1411, 1412, brands the Pahlavi word in this passage as 
as 'falsch'). Geldner, Drei Yasht, 'ram' (reading vardk}. 
p. 65, n. 1, suggests the hawk, 2 See Stein, Zoroastrian Deities on 
'habicht,' as a possibility. The Bun- Indo- Scythian Coins, in Indian Anti- 
dahishn, 14. 23, calls the raven vardk qitary, 17. 207, London, 1877 = re- 
(the Modern Persian word for crow, print, p. 14, Bombay, 1888. 
kaldgh, is not to be confounded with 3 From saena mdrdya comes the 
this), and this is apparently the bird name of the mythical bird Simurgh. 
of victory which accompanied King * Xenophon, Cyropaedia, 2. 1. 1 ; 
Ardavan according to the Pahlavi 2. 4. 19. 
Kdrndme-i Artakhshir-i Pdpakdn, 6 Yt. 19. 3; Ys. 10. 11. 



PERSIAN DOGS 77 

ful as I expected, perhaps owing to the cold season at the 
time when I was there. 1 

The dogs formed a special object of notice because of the 
esteem in which they were formerly held among the ancient 
Zoroastrians, although now despised by the Mohammedans 
except for hunting and as watch-dogs. So far as my knowl- 
edge goes, the Avesta is the only Oriental work which is highly 
complimentary to the dog. It is true that he was venerated in 
Egypt, but in India the Sanskrit writers usually speak of him 
in derogatory terms, and so do the Hebrews in the Bible. 2 A 
special sanctity, however, is attached to the animal in the Ven- 
didad, and three chapters of this sacerdotal work are devoted 
to the faithful friend of man. 3 Amid the high praises of the 
canine virtues, the text does not hesitate to call attention to 
certain vices, so that the eulogy does not degenerate into mere 
flattery. 4 My observations of dogs were not confined to Azarbai- 
jan, but were continued throughout the journey, south, east, and 
north. 

The typical dog of northern Iran, Transcaspia, and Turkis- 
tan is a large brute, resembling the mastiff in size, tawny in 
color, and roughish in coat, although the hair may be thinner 
and smoother in summer. In appearance he is somewhat wolf- 
ish, and in temper extremely savage. These wolfish char- 
acteristics corroborate the allusions in the Vendidad to the 
nature of the hybrid sprung from a dog and a wolf. 5 Most of 
the village dogs in the outlying districts of Azarbaijan have 
their ears cropped quite close to the head, their masters adopting 

iVd. 6. 45, 46; 7. 30; 3. 20; He- New York, 1905. Among Occidental 

rodotus, Hist. 1. 140 ; 3. 16 ; Cicero, writers Dante and Shakspere do not 

Tusc. Disput. 1. 45. hesitate occasionally to give the dog a 

2 An exception may perhaps be metaphorical kick. 

found in Vedic times, when the dog 8 See Fargards 13, 14, 16 of the 

seems to have enjoyed a better repu- Vendidad, and consult Hovelacque, Le 

tation ; cf. Hopkins, The Dog in the Chien dans V Avesta, Paris, 1876. 

Rig-Veda, in Am. Journ. Philol. 15. * See especially Vd. 13. 44-48. 

154-163, Baltimore, 1894. Compare yd. 13. 41-43. 
ulso Watson, The Dog Book, 1. 15-20, 



78 AROUND LAKE URUMIAH 

this practice to prevent the ears of the creatures from being 
torn in the fierce fights in which they constantly indulge. 

One of the best opportunities which I had for noticing dogs 
was later at the city of Urumiah itself, where I once saw a 
motley collection of a dozen or more dogs gathered about the 
slaughter-house. Here I observed a good specimen of the 
4 white dog with yellow ears,' which the Vendidad enjoins as 
one of the two dogs to be used in the sag-did ceremony of exor- 
cising the spirit of death. 1 The second kind of dog required for 
the Avestan rite, namely a ' yellow dog with four eyes ' - - that 
is, with two spots above the eyes I did not specifically see. 
The spots over the eyes are apparently less common, which 
may account for the value of such dogs in the ancient cere- 
monies, and some European friends gave me the interesting 
information that the German dachshund loses the tan spots 
over his eyes after a generation or two in Persia. 

Besides the tawny or yellowish village dog, the black, white, 
and parti-colored dog is also to be seen, especially in the towns. 
As a rule they are smaller in the cities than in the country and 
partake more of the mongrel type. Curiously enough, the 
village dog, despite his savage courage, especially toward 
strangers, proved to be in absolute dread of being 'shot' by 
the snap of a camera. It took me nearly a week before I 
succeeded in getting a photograph of the typical village dog of 
Azarbaijan, but I finally succeeded at a hamlet between Dilman 
and Guchi, by decoying the animal with egg-shells that I had 
thrown away after a hasty meal. 

The stages of my journey as far as Dilman in the Plain of 
Salmas, northwest of the lake, were slow, averaging not more 
than twenty-five or thirty miles a day, with stops for the night 
at the villages of Dizah-Khalil 2 and Tazvich. Dilman was 
reached in the afternoon of Sunday, March 22. This town 
is one of the largest places in the plain, but its manzil seemed 

1 Vd. 8. 10 ; see my article inJAOS. 2 See p. 91, n. 1, below. 

25. 182-183, and cf . p. 388, below. 



DILMAN AND SUM AT DAGHI 79 

little more than a shelter, nor was I attracted by an invitation to 
visit the tea-house near-by, which I later learned was an opium 
den. The carcass of a dead horse lying in the stream that 
runs through the town seemed characteristic of the place, but 
the people appeared to be kindly disposed to a stranger and 
showed something of a holiday spirit, as they had put on their 
best clothes to celebrate the season of No-Ruz and also Sunday, 
which is always a special ' bazaar-day,' the regular Mohamme- 
dan Sabbath being Friday. 

I left Dilman the next morning at eight, with the weather 
bleak and dreary. In about two hours we came to an inter- 
esting old Armenian cemetery which is situated on a hil- 
lock near a small village. To reach this graveyard I had to 
leave the wagon and wade through snow and water for nearly 
half a mile. An inscription on a large monument near the 
summit ascribes the founding of the burying-ground to the 
Mamikonians, an ancient heroic family of Armenia, and I ob- 
served a number of grave-stones rudely shaped like a ram, a 
common image in old Armenian burial-places. I noticed also 
one long Syriac inscription, but the letters were almost illeg- 
ible. 

After a short stay on the hillock I proceeded on my journey 
to see the Sasanian bas-relief of the horsemen, which I knew 
was carved on the side of a rocky hill called Surat Daghi, 
' Picture Mountain,' somewhere on the road between Dilman 
and Guchi. Shortly before noon we reached a tiny hamlet 
and found that we had passed the hill of the sculptures, hav- 
ing met no one on the deserted plain who knew precisely where 
the carvings were located. Accordingly I took a guide and 
returned on foot through mud and snow for a distance of nearly 
three miles in order to examine the bas-reliefs. 

The sculptures are carved about a hundred feet above the 
plain on the face of a somewhat precipitous rock, and are 
undoubtedly Sasanian in origin, as they present all the char- 
acteristic features of the bas-reliefs at Tak-i Bostan, Naksh-i 



80 AROUND LAKE U RUM I AH 

Rustam, and Naksh-i Rajab. The group is composed of four 
figures, two mounted and two standing. The equestrian figures 
are royal personages, apparently represented in the act of receiv- 
ing crowns from the two unmounted figures, which look like 
vassals and almost resemble grooms. The first of the mounted 
individuals appears to be older than the other and wears a 
mustache and apparently also a beard; the second is younger 
and looks almost smooth-faced, but on closer examination this 
absence of the beard is only apparent, not real, being due to a 
mutilation of the lower part of the face by some iconoclast. 1 
Both figures wear the familiar balloon-shaped head-gear with 
streamers floating out behind and a scarf or veil fluttering from 
below the shoulders. The cloak about the shoulders of each is 
clasped in an easy manner, the garment of the elder personage 
being the more elaborate. Each horseman grasps the reins of 
his steed with the left hand, which rests at the same time upon 
the hilt of a long, straight sword; while the right hand is 
extended to receive some proffered gift, which is hidden behind 
the horse's head in the first case, but looks like a chaplet in the 
case of the second cavalier. The close-fitting coat or tunic, the 
baggy trousers flowing in rich drapery from the knees, and 
more elaborately carved in the case of the elder personage than 
in that of the younger, together with the heavy caparison of 
the horses, which includes a massive chain and ball swinging 
at the left flank, are typical of sculpture of the Sasanian 
dynasty. The pose of the two horsemen is lifelike and 
spirited, although the workmanship is imperfect. 

The men on foot are represented as bareheaded and with 
beards, mustaches, and hair bushy at the sides. The face of 
the left figure is much mutilated, but that of the right is pre- 

1 My note-book has the memoran- Porter's sketch, which represents the 

dum ' smooth-faced ' corrected to * no, figure as having a full beard ( Travels 

hardly ' upon a more careful inspec- in Persia, 2. 597, pi. 82), as well as by 

tion of the stone, and this is confirmed the drawing of Flandin and Coste, 

by the photograph, which shows the Voyage en Perse, Ancienne, 4. pi. 204- 

mutilation of the rock, and by Ker 205, and Texier, Description, 1. pi. 40. 







02 



SASANIAN BAS-RELIEF OF THE HORSEMEN 81 

served with sufficient clearness to show details, including what 
seems to be a collar or band about the neck. Both individuals 
are clad in a simple manner, the upper garment being a tunic- 
like coat, the lower being huge bulging trousers. There is a 
double belt about the waist of each, but no sword is noticeable, 
nor is there any characteristic accoutrement or decoration, but 
from the forearm of the figure on the right there hangs a pen- 
dant that looks like a circlet suspended by a short band. 

The generally accepted identification of the group is that 
the bas-relief represents Ardashir Papakan, the first Sasa- 
nian king, and his son Sapor, receiving the submission 
of the Armenians, an event that occurred about A.D. 230, 
to which period the sculptures approximately belong. 1 

My examination of the sculptures took me some time, and 
on the way back I became separated from my servant and the 
guide, who had gone to search for an article I had lost on the 
way, so that I had some anxiety about finding the road to 
the hamlet alone ; but the servants finally came up, and we all 
returned together to the little settlement of mud hovels, and 
there I had something to eat, and succeeded in getting with my 
camera a snapshot of the Azarbaijan dog to which I have 
already referred. 

The afternoon was well advanced when the journey was re- 
sumed, and we turned our faces southward to cross the Kara- 
bagh ridge, over six thousand feet above the sea. The snow 
lay almost as deep as it did in the Avestan winter of Yima 
(Jamshid), 2 and it grew heavier and heavier as we ascended 
the mountain, until about half-past four the track was finally 

1 So Justi, Empire of the Persians, Travels, 2. 599, who attributed an Ar- 

2.259; Wilson, Persian Life, p. 91. menian origin to the sculpture, and saw 

We have an incidental allusion also in the two horsemen the Roman em- 

to the opposition of Armenia to peror Galerius and the Persian king 

Ardashir's authority in the Kama- Narses, the latter making concessions 

mak-i Artakhshir-i Papakan, 6. 2(ed. to the Armenian prince Tiridates. 

Darab Sanjana, p. 24, Bombay, But this explanation of the sculpture 

1896). A different explanation of seems doubtful, 
the scene is given by Ker Porter, 2 See Vd. 2. 22. 



82 AROUND LAKE URUMIAH 

lost in drifts as high as the horses' backs. We could proceed 
no farther for the nonce, and there seemed a good prospect of 
spending a freezing night on the mountains at an altitude of 
several thousand feet. At last traces of the path were discov- 
ered, and the guide went forward to secure horses and assist- 
ance from a post-house which was said to be some distance 
beyond. The falling snow and the biting cold made the de- 
lay seem long, but the messenger returned within an hour, lead- 
ing three horses ; I was able, therefore, to dismiss the wagon, 
giving directions to the driver to return to the hamlet as best 
he could. I never heard how he reached his destination ; but I 
imagine he arrived safely, inshdllah, ' by the Grace of God,' a 
.Deo Volente phrase which the Persians employ in connection 
with everything they do. 

The horses that came from the hill post looked thin and 
poorly kept, but we placed the load upon the back of the 
strongest ; the second animal was assigned to my servant Safar, 
and I mounted the third myself. Away we started through 
the snow. 

As we crossed the first ridge there was still enough light 
to see, on the edge of a stream below, the fresh carcass of "a 
horse that appeared to have fallen over the edge into the 
gorge, and from which a huge shaggy dog was tearing strips 
of fast-disappearing flesh. Dusk fell, and darkness closed in 
rapidly as we began the wild ride over the pass and through 
the gorges of the Karabagh mountain toward Guchi. 1 The 
native guide inquired ominously for our guns, which we did 
not have, and I thought lovingly of my revolver, which by 
this time was peacefully reposing at Tiflis. The night seemed 
too stormy, however, for robbers, but as I now look back on 
it and think of the murder of my friend Mr. Labaree and his 
servant on the same road from Khoi to Urumiah a year later, 2 
and the pillaging of a party of ten in this region by a band of 

1 Or Kuchi, as the name is more 2 See p. 90, n. 1. 

accurately rendered. 




THE GREAT PLAIN OF URUMIAH 




PLOWING WITH BUFFALOES NEAR URUMIAH 



A WILD RIDE 83 

Kurds some months after I passed through it, the risk of the 
journey seems very real. 1 

The wind was high, and it drove sheets of blinding snow 
from the north, which we fortunately escaped whenever the 
path veered southward, although the drifts were often up to the 
horses' bellies. Over streams we proceeded and along precipices 
which seemed more sheer in the darkness, until the night became 
so pitchy black that there was nothing for us to do but throw 
the reins on the necks of the horses, swing our arms to keep 
from freezing, and trust to the horses' instincts to find the way. 
All this while my cane, umbrella, and hat-box with top hat 
were trundling at the side of my saddle and seemed ludicrously 
out of place amid these surroundings ; but I had had no chance 
to get rid of them (except by throwing them away as useless 
appendages) since I bade adieu to the last traces of civilization, 
where I had needed them in paying official visits. To cheer 
the situation and brighten my own spirits as well as those of 
the men, I whistled a tune and began to sing. All I could 
think of at the moment was the ' Star-Spangled Banner ' 
a strangely democratic pibroch for the highlands of Iran ! 
The guide responded with a Turkish ditty, which I answered 
in turn by a few lines of Hafiz that I knew in Persian, to the 
apparent entertainment of our leader, and thus we made 
the dark journey through what seemed to be the valley of the 
shadow of death. The lights of the mud houses on the out- 
skirts of Guchi at last began to twinkle through the gloom, 
and it was not long before we had descended from the only 
remaining hill, crossed a level stretch, and found ourselves 
lodged in the best house in the village. 

The master of the house was a tall, muscular man with heavy 
mustachio eyebrows which would have given his face a look of 
fierceness had it not been for his well-shaped nose and the 
kindly expression about his mouth. He was entertaining No- 

1 The Avesta alludes to such high- Zoroaster's time under the designa- 
wayrnen, bandits, and assassins in tions tayu, hazanhan, gada. 



84 AROUND LAKE URUMIAH 

Ruz guests, as the New Year season was approaching its height, 
and although the company was composed of a set of rough- 
looking fellows, resembling tramps and bandits, they seemed 
to have good hearts, and ' 1 was of their felaweshipe anon.' 

Some simple food and New Year's sweetmeats were provided 
for refreshment; my camp-bed was set up to furnish me 
a seat, while the guests squatted around the steaming samovar 
and enjoyed hot tea sweetened to a syrup by masses of sugar. 
Two little boys came in after the repast was over. They were 
clad in dirty rags, but were brothers of our host, who had sum- 
moned them to sing for our entertainment. Their bright faces 
beamed as they sang, which they did lustily and shrilly, beat- 
ing time rhythmically upon a rough tambourine, while the host 
joined with a zest in the music. Finally he passed the tam- 
bourine to me with a request to sing. I chose 'Yankee 
Doodle ' in preference to 4 Home, Sweet Home,' because of the 
tune, and not the sentiment, for in general I believe the 
Persian would prefer the 4 chop-sticks waltz ' to a melody of 
Rubinstein or a symphony of Beethoven. Apropos of song, I 
asked the company if they knew the story of Shirin and her 
sculptor-lover Farhad, so beautifully told by Nizami. 1 Several 
knew it well and recited portions both in Persian and Turkish. 

It was past midnight when the company broke up for the 
night and I found that I was to be one of five to occupy the 
rough-raftered, mud-walled room. I felt that I was in Media 
Past, for the A vesta presumes a condition of society in which 
a number shall occupy the same room, since it speaks of ' men 
lying down to rest in the same place, on a rug together, or 
a pillow together, whether there be two men by one another, 
or five, or fifty.' 2 The preparation of the natives for retiring 
consisted chiefly in loosening their belts and curling themselves 
up under a blanket, resting then undisturbed until the time 
came to give a shake, like a dog, in the morning. Three of the 
party who had not yet gone to bed remained squatting near 
i See pp. 188, 226. 2 Vd. 5. 27 = 7. 5. 




A GLIMPSE OF URUMIAH 




NORTHERN END OF LAKE URUMIAH 



PEASANTS AS HOSTS 85 

me and watched the proceedings with interest, first examining 
my pistol-belt as I unclasped it (now minus the revolver), 
admiring the leather, especially the mechanism of the clasp, 
then commenting on the quality as a whole, and passing the 
belt to the next. My russet riding-leggings, with their practical 
Yankee fastenings, called forth special approval, so I passed 
these around as further exhibits, and each of the three guests 
who were awake gravely tried one on. This done, we all fell 
asleep and remained undisturbed till after daylight. The 
inspection of exhibits was then renewed as I unpacked my 
dressing-case, brought out my folding mirror for shaving, 
and proceeded to load my camera with fresh films. These 
stages furnished exhibits X, Y, and Z, and I was glad when 
I could pack up again and escape from being stared at. 

Despite my oft-repeated zud, zud, tez, tez, 'quick, hurry up, 
the sun had nearly reached the zenith before I succeeded in 
getting two horses and a cart with which to resume the journey 
to Urumiah, hoping to reach it that night. When I walked 
out from the lodgings where we had spent the night, I found 
we wer& on the edge of the great Plain of Urumiah, having 
descended into it after the wild ride over the Karabagh passes 
the night before. To hurry matters up for departure I started 
to enter the courtyard where the men were engaged in hitch- 
ing up the cart which was to be our next means of trans- 
portation. One of the servants rushed forward to urge me 
not to enter because of the savage dogs. In the best Persian 
I knew I protested that I was not afraid of dogs, but dis- 
covered in an instant that these dogs were of a most vicious 
kind, and was forced to beat a retreat. The cart was finally 
ready, and we started. 

In the noonday air there was a slight suggestion of spring, 
and I noticed that the peasants were making a first attempt 
to turn up the soil with their plows. The Persian plow is a 
very primitive sort of affair. It consists of the crotch of 
a tree cut in such a manner that one of the two branches 



86 AROUND LAKE URUMIAH 

may be sharpened and shod with iron to serve as a plowshare, 
while the other, or main trunk, serves as the beam. Bullocks 
or cows are hitched to the unwieldy implement, and wheels are 
sometimes added to lighten the lumbersome affair. The soil of 
the great alluvial plain, however, yields readily to such primi- 
tive implements, because it is one of the most fertile districts in 
all Persia and justifies its right to the title ' Paradise of Iran.' 
The entire district, in fact, merits the praise which the Avesta 
bestows on the larger region of Airyana Vaejah, or Azarbai- 
jan, when it calls this 'the first and best of places created by 
Ormazd.' 1 

My efforts to reach Urumiah by nightfall failed signally at 
Karmabad, for there the guide absolutely refused to proceed 
farther on account of the dreadful state of the roads, which 
made it impossible to reach the city before dark. I tried every 
device from coaxing and bribery to commanding, but he and 
his companions were obdurate. Finally I had to yield and 
spend a cold night at the uncomfortable manzil, receiving a 
promise, however, that the start should be made at daylight on 
horseback. Only twice in my Persian experience, besides this 
occasion, did I fail to carry my point about proceeding on the 
march when the natives objected; but this time I found that 
the men were right, as the guide, with some satisfaction, showed 
me next day when we floundered through seas of mud and slush 
which might have proved dangerous to life in the darkness. 
It was with a veritable feeling of joy, toward noon that day, 
March 25, the sixth of my journey from Tabriz, that I neared 
the walls of the city of Urumiah, one of the several towns that 
lay claim to having been the birthplace of Zoroaster. 

1 Vd. 1. 2. All writers, ancient and Mustaufi, cited in Barbier de Mey- 

modern, speak of the richness of the nard's translation of Yakut, p. 26, 

soil and the abundance of the crops n. 3, and also Curzon, Persia, 1. 635. 
about Urumiah. See, for example, 




A BUTCHER AND A COBBLER, URUMIAH 




WOMEN WASHING CLOTHES, URUMIAH 



CHAPTER IX 
URUMIAH, A SUPPOSED EARLY HOME OF ZOROASTER 

* And there I shaped 
The city's ancient legend into this.' 

TENNYSON, Godiva, 3-4. 

'THEY claim that Urumiah was the city of Zardusht and 
that it was founded by the Worshippers of Fire,' so writes the 
Arab traveller Yakut, who visited the city in A.D. 1220, and a 
still earlier author Ibn Khordadhbah (about A.D. 816) calls it 
' the city of Zaradusht,' while Al-Baladhuri (A.D. 851) also 
notes that ' Urumiah is an ancient city of Azarbaijan, and the 
Magians think that their master, Zaradusht, came from there.' 1 
A half dozen other Oriental writers make similar statements 
associating Zoroaster's name directly or indirectly with 
Urumiah and pointing to its antiquity. Nevertheless the city 
is not mentioned in the Avesta or in the Pahlavi literature, 
for Anquetil du Perron was wrong in fancying that he recog- 
nized the name of Urumiah in the Zoroastrian prayer Airyema 
Ishyo; 2 although it does seem possible, as suggested in the 
preceding chapter, that its present name Ur-ml, Uru-miah, the 
latter element of which the natives often associate with md^ 
4 water,' may in some distant manner perpetuate the Avestan 
attribute uruy-dpa, urv-dpa, ' having salt (or warm) water,' 
which is applied to it in the ancient texts. 3 Most of the in- 
habitants, especially the Nestorians, call the city Urmi, the 
Persians Urumiah or Urmia, while European books employ 

1 See my Zoroaster, pp. 197-198 ; 2 Cf. my Zoroaster, p. 97, n. 1. 

also pp. 17, 30, 38, 48, 49, 96, 166, of 3 See p. 73, n. 3. 

the same work. 

87 



88 

Ouroomiah, Oroumiah, Urunriyyeh, and Urumia, besides other 
spellings. 

In its geographical situation the city is fortunate, lying in 
the alluvial plain of the ' Paradise of Iran,' and the climate is 
salubrious, though sometimes hot in summer after the rigorous 
winter. The river, which flows past the city on its southern 
side, and the abundant streams formed by the snow melting on 
the Kurdish hills to the west, assure a plentiful water-supply 
and excellent facilities for irrigation except during the extreme 
heat of summer. It is true that famine visited the city in 
1879, but that was at a time when the scourge swept over a 
large part of Iran. Systematic cultivation in recent years has 
done much to obviate for all time the recurrence of such a 
disaster. The country, for miles around, is covered in summer 
with gardens that produce melons and cucumbers in abun- 
dance ; with orchards laden with apples, pears, peaches, plums, 
apricots, quinces, cherries, and mulberries ; while the grapes of 
the vineyards are proverbial for their excellence. Wheat, 
barley, rice, and millet are among the products of the fields, 
and tobacco has been grown for many years, but its quality is 
suited rather for the common pipe, chibuk, than for the kalian, 
in which the natives smoke rather the tobacco of Shiraz. 

When I first saw Urumiah, at the end of March, there was 
nothing indicative of seed-time and harvest. The snow had' 
only just begun to melt, deluging the plain with floods of water 
and converting large areas into seas of mud. Through this 
slough of despond we had slowly to wade our horses, and it 
needed the sharp ring of the guide's song to which I often 
echoed a bravo, khaili khub, to cheer us on and keep the strug- 
gling animals in their course. A glimpse of the disappearing 
traces of the muddy way may be seen in the picture I took of a 
caravan of camels near the gate of Balau, as we entered the city 
from the north. They were stringing their way along, with 
dull-toned heavy bells, toward the caravansarai where they were 
to lay off their loads. I felt a special interest in these drom- 



FIRST IMPRESSIONS OF U RUM I AH 89 

edaries near Zoroaster's city because the prophet's name, 
Zarathushtra, is said to mean some sort of a camel (ushtra). 1 

The city of Urumiah is girt by a wall some three or four 
miles in compass, pierced by seven gateways and strengthened 
by a moat at the more vulnerable points. The value of this 
double defence had something of a test during the memorable 
Kurdish raid upon Urumiah in 1880, when the city was be- 
sieged and threatened with destruction through an attack by 
the hostile descendants of the ancient Carduchi, who had 
plundered the surrounding villages, burned, ravaged, and 
murdered throughout the entire borders and were checked only 
after much bloodshed and considerable damage to property. 2 

As one enters the town, Urumiah gives the impression of 
most Persian cities. Some of the streets are fairly broad, and 
a rough attempt has been made, here and there, to pave them 
with large round stones from the river-bed. There is no 
system of drainage, save the water-channels from the moat and 
river, which serve alike to receive refuse and furnish a washing- 
place for the women to launder their clothes. A hopeful sign, 
however, that better municipal ordinances may some day be 
established, with efficient authority, is seen in the fact that the 
butchering of animals in the public streets has been forbidden 
and a public slaughterhouse has been built near the Hazaran 
gate in the northeastern part of the city. No provision, how- 
ever, seems to be in force against shovelling the snow off the 
roofs into the streets and heaping up barriers that make the 
thoroughfare at times impassable ; nor is there any restriction 
against the use of burying-grounds in the city, where economy 
of space, time, and labor leads to using the same grave two 
or three times. The quick and the dead are one kin, and the 
Persian has little idea of hygiene in such respects ; I noticed 
especially in the hamlets of the rural districts that a preference 

1 The meaning 'plowing camel' has 2 For a full account of the events 

even been suggested ; see my Zoroas- connected with the Kurdish raid, see 
ter, pp. 147-149. Wilson, Persian Life, pp. 109-124. 



90 URUMIAH, A SUPPOSED EARLY HOME OF ZOROASTER 

was shown for placing the graveyard on the nearest hill, at the 
foot of which the village well was dug. As I followed the 
main street into Urumiah it led directly across a cemetery. 
There was no way of avoiding the graves, and the horses' 
hoofs often beat hollow over the excavation beneath the sod. 
In an adjoining plot a burial was in progress, and the mourners 
were still gathered about the half-filled grave, at the head of 
which was placed a rough stone without any inscription. It 
chanced to be the only funeral I saw during my stay in Persia, 
where the population is less dense and vitality less low than 
in plague-stricken India. 

Another quarter of an hour's ride, and I found myself dis- 
mounting before the door of the American Presbyterian Mis- 
sion, as the guest of the Rev. Dr. Benjamin Labaree. A 
welcome there awaited me which those can best appreciate 
who, weary, mud-bedraggled, chilled, and out of sorts, have 
endured the discomforts of 4 the road ' for six hard days through 
snow, slush, mire, and storm. 1 

In less than two hours after my arrival my host had ar- 
ranged for my first afternoon of research among the ash-hills, 
which was to be an excursion with his son to Degalah, the 
largest mound near the city. There are more than a dozen of 
these elevations directly in the vicinity of Urumiah, and it is 
stated that there are as many as sixty-four about the lake. 
The larger number of them are scattered over the Urumiah 

1 1 shall never forget this meeting were stripped of everything of value, 

with Dr. Labaree Sr. and young Mr. and the assassins escaped over the Turk- 

and Mrs. B. W. Labaree. Almost ish border. The bodies of these two 

exactly one year later, March 9, 1904, martyrs to the Christian cause were 

Mr. Labaree Jr. was brutally mur- afterward discovered and conveyed to 

dered by bandits and fanatics on the Urumiah, where they were buried in 

road from Dilman over which I had the same grave. The United States 

passed. His servant, a bright young government followed up the matter of 

native, was shot, and the body robbed the murder to its source and obtained 

even of the clothes, and Mr. Labaree from the Persian government some 

was carried away to a mountain ravine, reparation for the heinous crime, and a 

where he was savagely stabbed to death guarantee for the greater safety hence- 

with daggers and swords. His remains forth of American citizens in Persia. 




EXCAVATIONS IN DEGALAH ASH-HILL, NEAR URUMIAH 




SPECIMEN OF ANCIENT POTTERY FROM URUMIAH 
(In the author's collection) 



OF THF 

UNIVERSITY 
CAI 



THE ZOROASTRIAN ASH-HILLS 91 

plain and the plain of Sulduz to the south, but not to the 
north in Salmas. 1 They are all composed of immense deposits 
of ashes mixed with earth, the ashes having been added in many 
cases to a natural small elevation. ' In fact, there is scarcely 
an eminence on the plain which has not been increased, usually 
to a very great extent, by this means.' 2 The natives all agree 
in calling them 'hills of the Fire- Worshippers.' One must be 
careful, however, not to mistake for ash-mounds some of the 
numerous hillocks (tapah) about the lake, like the Gum 
Tapah at Mayan, not far from Tabriz, which imagination 
might easily crown with a fire-shrine. My first guide, know- 
ing my interest in the subject, obligingly called the Gum 
Tapah an Atash Gah (fire-temple), but it is a mere sand-heap 
and was probably never one of the Zoroastrian pynea. 

The village of Degalah directly adjoins Urumiah. The ash- 
hill is three or four hundred yards long, nearly as many broad, 
and a hundred feet or more in height ; but its dimensions are 
constantly being reduced, as the peasants within the past 
century have discovered the value of the alkaline quality of the 
ashes for fertilizing purposes and for producing saltpetre. As 
a consequence the hill has been burrowed into, tunnelled, 
trenched, undermined, and cut down in scores of places, and 
the soil carried off to spread upon the adjoining farms. The 
photographs taken by Mr. Labaree and myself will show some 
of the pits and hollows resulting from these excavations. The 
structure of the mound was therefore easy to examine. It 
consists of soft earth with stratum upon stratum of solid ashes 
at varying depths and several feet thick. There is little stone 
in the mass, but in former times some stone buildings stood on 
the top of the hill, and the village of Degalah is built largely 
from the stones of these, as I am informed by my colleague, 

1 On Gaur Tapah, ' Unbeliever's missionary among the Nestorians of 
Hill,' near Dizah-Khalil, on the north Persia, Mr. E. C. Shedd, cited by Dr. 
shore of the lake, see Ker Porter, W. H. Ward, Notes on Oriental An- 
Travels, 2. 606. tiquities, in Am. Journ. Archaeology, 

2 This statement is quoted from a 6. 280. 



92 URUMIAH, A SUPPOSED EARLY HOME OF ZOROASTER 

Dr. A. Yohannan, who was born there. I understand also that 
a foundation-wall of burnt brick was discovered some time 
ago near the bottom of the hill, 4 the bricks measuring at least 
six inches thick by eighteen to twenty-four inches long ' a 
statement which would agree with the so-called ' Gabar bricks ' 
of Zoroastrian structures which I found elsewhere in Persia. 1 

In their excavations the workmen are constantly unearthing 
fragments of pottery, sometimes whole vessels, terra-cotta fig- 
urines, coins, and other remains which show signs of consider- 
able antiquity. The specimens of earthenware are usually of a 
reddish or brownish clay, the commonest being a round pot 
with small handles or with a spout. They are generally with- 
out decoration, although a few have figures of men and horses, 
crudely drawn, or bands of color and other marks of ornamenta- 
tion upon the surface. 2 Some of the jars are two feet or more 
in height ; I saw such an amphora at a depth of more than 
twenty feet below the surface in one of the pits into which I 
went down. It was buried in an upright position in the earth, 
but was partly broken, so that we did not disturb it, except to 
scrape some of the debris from around it, which disclosed a few 
pieces of bones, grains of parched corn, and ashes in abundance. 
Potsherds by the hundreds were lying at the bottom and about 
the mouth of every pit, but I could not learn of a single 
instance where any inscribed tablet or cylinder had been found 
among the layers of earth and ashes. 

It is common, when speaking of this and the other ash-hills 
around Urumiah, to say that they are composed 'entirely of 
ashes,' 3 but from my examination in the present instance, and 
my investigations in others, this term is to be taken relatively. 

1 See the quotation by Mr. Shedd pies may be found in the hands of the 
in Dr. Ward's article (p. 286) previ- villagers or of residents in the city, 
ously cited, and cf. p. 255, below. 8 Mr. Shedd, op. cit. p. 286 ; and a 

2 A good collection of specimens native of Urumiah, Jonathan Badall, 
may be seen in the museum room of now in Yonkers, informed me that the 
the American Missionary College at hill of Lakki, thirteen miles north of 
Urumiah, and many individual sam- Urumiah, is composed 'wholly of ashes.' 



OF THE 

UNIVERSITY 




THE ASH-HILLS AT DEGALAH AND TERMANI 93 

I believe therefore that Dr. Ward, even though he had not seen 
them, was right in his impression that they are composed rather 
'of clay which has become mixed with ashes and saturated 
with nitrous salts of organic composition ' ; and he shows from 
an old Babylonian sculpture how such mounds could be built 
up. 1 There is every reason to assume that these elevations 
were surmounted by sanctuaries dedicated to the worship of 
fire, even if we do not agree in every detail with the natives, 
who unanimously attribute the vast accumulation of ashes to 
the accretions from the fire-temples, the ashes having been 
scattered over the hill age after age. 2 

On the following day a party was formed to visit another 
ash-hill and we rode out to Termani, six miles east of 
Urumiah. This mound culminates in an elevation some- 
what resembling a cone. A short distance from the conical 
rising it was possible to trace the general outline of what 
was once the foundation of an old building. The large size of 
the stones called forth comments from the natives, who 
expressed surprise as to how such large blocks could have been 
conveyed to the site where they stood. The hill itself has not 
been much excavated, but about fifteen or twenty years ago, 
when a well was being sunk near the top, an image of consider- 
able size was found. Unfortunately it was destroyed by the 
iconoclastic workmen, who were Mohammedans, as image- 
making is forbidden by the Koran. 3 On the hill I could see 
abundant traces of ashes everywhere, although they were not 
quite so plentiful as at Degalah. My judgment in this case, 
however, is based only upon an examination of the surface of 
the ground, as the mound had not been trenched and excavated 
like the other ; but the ground was strewn with potsherds 
which incidental diggings had brought to light, and the natives 

1 Ward, op. cit. p. 287. 8 See Koran (tr. Sale), chap. 2, pp. 

2 There is nothing of a volcanic 18, 23, etc., and the Mohammedan 
nature in the deposit, so far as my tradition against pictures and images 
limited geological knowledge allowed in Mishkat, bk. 12, chap. 1, pt. 1, and 
me to judge. bk. 29, chap. 5. 



94 URUMIAH, A SUPPOSED EARLY HOME OF ZOROASTER 

had numerous specimens of earthenware vessels thus unearthed. 
One of these, which is shown in the photograph, had lost part of 
its spout, but kept its small handle ; another presented a slight 
attempt at artistic finish, as the handle was twisted in a curious 
but rather graceful shape ; still another, which did not look 
quite so old as the specimen that I selected, somewhat re- 
sembled a modern teapot with holes perforated in the spout 
to serve as a strainer. 

A third mound which I visited was the hill of Ahmat, a 
short distance southeast of Termani. In one of the trenches 
there were fragments of a rather large urn, and I am told that 
specimens as large as a man are sometimes exhumed and that 
skeletons have been found buried in them. 1 The natives also 
informed me that, in excavating, they sometimes come across 
neatly made graves in these ash-mounds, with a stone slab 
covering the place where the body lay. The truth of this I 
afterward proved. 

The following day spent among the ash-hills was devoted 
to the mound at Geog Tapah, or Gog Tepe, which lies a little 
east-southeast from Urumiah. It was the fourth ash-mound 
that I visited, and was one of the largest. A Christian church, 
erected by the Nestorians, now crowns the summit of this 
ancient ash-hill, and the minister, Mr. Morehatch, an Assyrian 
Christian born near Urumiah, told me that when the workmen 
were excavating for the foundations of the church they came 
across an underground chamber built of stone and containing 
a carved hollow cylinder three or four inches high. The stone 
vault, he explained, had been filled up in order to make the 
foundation of the building more secure, and the image had 
been purchased and sent to America. On my return home I 
found that this cylindrical bas-relief is now preserved in the 
Metropolitan Museum of Art in New York City, and that my 
friend Dr. William Hayes Ward had given a detailed descrip- 

1 See also a similar statement the article, already quoted, by Ward, 
made by Mr. E. C. Shedd, cited in Am.'Journ. Archaeology, 6. 287. 




A TYPICAL MUD HUT AT TERMANI, NEAR URUMIAH 




AN OLD MILL NEAR URUMIAH 



ANTIQUE POTTERY AND AN ANCIENT CYLINDER 95 

tion of it in the publication referred to in the notes, supple- 
mented by Mr. Shedd's account of the chamber where the 
object was found. 1 I have examined the cylinder a number of 
times, and through the courtesy of the Museum authorities I am 
able to give a reproduction of it. 2 

In shape it resembles a large napkin-ring made of trans- 
lucent alabaster, and it measures 3^ inches (94 millimeters) in 
height by 2-J inches (59 millimeters) in diameter, the walls 
being about a quarter of an inch (6 millimeters) in thickness. 
The surface of the alabaster has been rendered somewhat 
opaque by exposure, as Dr. Ward observes in his description, 
from which I freely draw. The design of the carvings, in the 
opinion of this authority, is archaic Babylonian, and the 
figures represent the sun-god, Shamash, emerging from the por- 
tals of the east and accompanied by other divine personages. 
The god (the second figure on the right in the reproduction) 
carries a club on his right shoulder and holds a weapon in his 
left hand, as he mounts with his left foot the crest of a low 
hill. The hill is conventionally indicated by several rectangular 
blocks, which are multiplied also for the other figures to stand 
upon, and they form at the same time an ornamental base for 
the cylinder. Two bearded porters, with flowing hair and wear- 
ing low double-horned caps, fling open the gates through which 
the god advances. Behind the left-hand gate-keeper stands 
the demigod, Ea-bani, half man, half bull, facing full front and 
holding in his two hands a standard. Behind him again are 
three figures, on the other side of the cylinder, approaching 
the sun-god. The first of these is a man ; the second a woman 
in a flounced robe, who is considered by Dr. Ward to be 
the sun-god's wife ; and the third, a bearded divine figure 
clothed in a long skirted mantle. In the garments of all the 
figures I would call attention to the border of fringe, charac- 
teristic of the Median robe and noticeable on the sculptures of 

1 See Dr. Ward's article, Notes on Journal of Archaeology, 6. 286-301. 
Oriental Antiquities, in American 2 One third of the actual size. 



96 URUMIAH, A SUPPOSED EARLY HOME OF ZOROASTER 

the archers discovered by Dieulafoy at Susa and on the effigy 
of Cyrus at Persepolis. Dr. Ward believes that this cylin- 
drical bas-relief found at Geog Tapah is at least as old as 
B.C. 2000, and probably older, and in his opinion it is 4 a 
purely Babylonian product, which was conveyed, probably 
in some conquest of a very early period, to this distant land 
of the Minni.' 

Beneath the sandstone floor of the chamber where the 
cylinder was discovered some fragments of bones were found, 
but they were so decomposed that it was impossible to make 
anything out of them. 1 It is not infrequent, Mr. Morehatch 
told me from his own observations, to unearth from the Geog 
Tapah mound large earthen jars containing bones, which show 
that the custom of urn-burial was sometimes resorted to, or 
else that the use of astoddns was in vogue, as explained below. 
Malik Shimmon, my native host, at whose house we lunched, 
stated that on his own property two skulls had been dug up 
with brass nails driven into the ears. If this is to be taken as 
indicating death by execution, added light may be thrown on 
the interpretation of a passage in the Vendidad, where, among 
other comparisons of torture and punishment, the torments of 
hell are described as being as painful ' as if one should nail 
the bones of his perishable body with iron nails.' 2 

It is quite evident, from what has been said, that Geog 
Tapah was not only an ancient settlement, but that a part of 
the mound served also as a cemetery. At one place, where a 
deep road had been cut in the hill, I saw a number of graves 
exposed, but they were near the surface and apparently not 
very old, although in a ghastly condition to behold. Choosing 
a spot on the other side of the hill, where it was possible to 
excavate at a lower level on account of the conformation of the 



1 See Shedd, quoted by Ward, op. steter, SEE. 4 2 . 48 ; Le ZA. 2. 63, 

cit. p. 287. n. 43 ; and consult also Bartholomae, 

2 So Vd. 4. 61 , fSSbiS . . . ava-paSal, is Air. Wb. p. 879. 
probably to be understood ; see Darme- 




BAS-RELIEF CYLINDER FOUND NEAR URUMIAH 

(Now in the Metropolitan Museum, New York) 

(Two thirds actual size) 




OPENING AN ANCIENT GRAVE IN THE SIDE OF THE GEOG TAPAH ASH-HILL 



OPENING A HILLSIDE GRAVE 97 

mound, we proceeded to examine an ancient grave or sepulchre 
that had already been laid partly bare. Malik Shimmon sum- 
moned an old man to open the grave, and in a few minutes 
his long shovel had laid the sepulchre fully unclosed to view. 
It was a rough sarcophagus of stone, a part of the upper slab 
and a portion of the side slab of which (both whitish in color) 
were intact. The head of the grave appeared to be located 
toward the right, as I looked into the opening. There was a 
large jar deep in the earth above what was presumably the foot 
of the grave, but we preferred not to remove it, as it was broken 
and apparently quite empty. The grave itself was likewise 
empty, except for a few fragments of bones scattered about; 
in fact in most cases where the interment shows the greatest 
signs of antiquity the skeletons have been reduced to dust, or 
at most a few pieces of bones remain. Yet Mr. Shimmon, who 
had seen a number of these stone receptacles opened, said that 
sometimes three or four skeletons are preserved in a single 
sarcophagus, and the digger recalled an instance where as 
many as six were found. I had the same assurance also on 
other authority. 

Regarding the age and true nature of these repositories I 
am wholly undecided at the present time. If they were pre- 
Mohammedan and pre-Nestorian, as seems probable, it is open 
to question to which era we are to assign them, whether to a 
time when the Zoroastriaii religion prevailed or to a still 
earlier date. If we assign them provisionally to the Zoroas- 
triaii period, we can only explain them as astodans, or stone 
receptacles constructed to receive the bones after the flesh had 
been denuded from the body by the vultures, in accordance 
with the tenets of the Zoroastrian religion. 1 

Leaving this opened grave and passing by another which 
had recently been discovered a few rods distant, but not fully 
laid bare, we turned our attention to other sights and ascended 

1 For this custom in regard to the Modi, Astodan, a Persian Coffin, Bom- 
bones, see Vd. 6. 44-51, and consult bay, 1889 (brochure). 



98 URUMIAH, A SUPPOSED EARLY HOME OF ZOROASTER 

the brow of the hill to view the surrounding country. In the 
near foreground still another ash-mound, a small elevation, was 
pointed out; it was apparently the one called Chachili Hill, 
which I passed two days later on my journey southward from 
Urumiah. The village of Saralan, built partly on an ash-hill, 
lay on the same road, and there I saw the remains of a building 
that had been exposed in digging, and likewise pieces of a large 
oven (tandur, tanur, Av. tanura), together with fragments of 
a huge amphora, or wine-jar (lino). Not far from there, in 
the same general southerly direction, lies the village of Dizach- 
Takiah, or Diza-Takiah, built on one of the very largest of the 
hills that show traces of ashes in this neighborhood. My Nes- 
torian friend, Rev. Yaroo M. Neesan, at whose uncle's house we 
spent the night, told me that he himself had found at Diza- 
Takiah a small statuette with Assyrian affinities. In looking 
over some of the specimens of pottery that had been unearthed 
there in digging the foundations of his house, I noticed one 
particularly interesting old piece ; it was a large terra-cotta pot, 
evidently very antique and probably used as a milking vessel, 
according to my native host. 1 

It was at Geog Tapah, at the house of Malik Shimmon, 
whose hospitality I enjoyed, that I had my first truly Persian 
dinner. We partook of this meal seated on the floor in 
Eastern fashion, bolstered up by soft cushions and lost amid 
a wilderness of dishes with a variety of viands character- 
istic of the country. Among these was the clabber (mast), a 

1 The general subject of the ash- pottery has been published by Virchow, 
mounds around Lake Urumiah was Fundstucke aus Grabhugeln bei Ur- 
treated many years ago, I believe, in mia, Persien, in Zt. f. Ethnol. (Verh. 
a sketch by Mr. Abbott, but I have d. Berliner Anthrop. Gesellsch.}, 30 
not been able to find the brochure or (1898), pp. 622-527; 32 (1900), pp. 609- 
even to discover its exact title. Drs. 612. My own notes, though imper- 
Lehmann and Belck gave some atten- feet, will suffice to call the attention of 
tion to the matter of the ash-mounds Zoroastrian scholars anew to this field 
in their recent scientific tours through for archaeological research in north- 
Armenia and northwest Persia. Some western Persia, 
information regarding their finds of 



THE PERSIAN NEW YEAR 99 

caseous mixture that recalled to my mind the milk and cheese 
diet upon which the classical writers say that Zoroaster lived 
for years in the desert. 1 On our way back to the city we 
passed a number of interesting sights, one of which was an old 
mill, which interested me because of the primitiveness of its style 
and structure. 

As I happened to be in Urumiah before the season of the 
No-Ruz festival, or Persian New Year, was over, I had an 
excellent opportunity of becoming still better acquainted with 
the social life of the people. No-Ruz, 4 new day,' the oldest of 
all Persian festivals, does not fall in January, like our own New 
Year's, but comes in the springtime when the sun enters the 
sign of Aries at the vernal equinox. Iranian tradition dates 
the festival back thousands of years and says that it was first 
celebrated in the golden age by Jamshid, who lived before the 
Deluge. It was he who established the solar year ; and al- 
though the Persians, as Mohammedans, have otherwise adopted 
the lunar calendar current among the Arabians, they have never 
given up No-Ruz and its observances. The festival lasts over 
a fortnight and is celebrated with as much spirit as it was a 
thousand years ago in the time of Harun al-Rashid, when we 
find it alluded to in the story of the Enchanted Horse in the 
Arabian Nights as ' an ancient and solemn feast throughout all 
Persia, which has been continued from the time of idolatry 
and is observed not only in the great cities, but celebrated with 
extraordinary rejoicings in every little town, village, and ham- 
let.' 2 Holiday attire, the interchange of gifts, congratulations, 
and good wishes, together with merry-making, are the order of 
the season. Nor has fashion allowed the time-honored custom 
of paying visits on glorious Jamshid's day to lapse. The 
callers are welcomed with large trays of sweetmeats and sugar 

1 See my Zoroaster, p. 34, n. 2, connected with No-Ruz and the origin 
where the classical references are of New Year's presents, see Albiruni, 
given. Chronology of Ancient Nations, tr 

2 Compare Arabian Nights, p. 462, Sachau, pp. 199-204, London, 1879. 
Philadelphia, 1835 ; and for legends 



100 URUMIAH, A SUPPOSED EARLY HOME OF ZOROASTER 

confectionery ; and favors of this sort are sent from friends to 
friends, because tradition says that 4 he who tastes sugar on the 
morning of No-Ruz before speaking, and anoints himself with 
oil, will keep off all sorts of mishap during the greater part of 
the year.' Sugar and confectionery, of which the Persians are 
so extremely fond, have also ancient authority and royal sanc- 
tion, since among other good things associated with the first 
New Year's Day and the inauguration of the solar calendar by 
King Jamshid was the happy discovery of the sugar cane, and 
the king 'ordered the juice to be pressed out and sugar to be 
made thereof.' 

In making New Year's calls I had the pleasant privilege of 
going in the company of Dr. J. P. Cochran, the missionary 
physician at Urumiah, and our first visit was paid to Majidi 
Sultana, the newly appointed Vice-Go vernor, who two days 
before had received from the Shah his official appointment and 
the accompanying gift of a superb sword. He received us with 
formal ease and grace and, after the customary bows and salu- 
tations of etiquette had been made, conversed freely and pleas- 
antly in Turkish, which the other visitors spoke well, but he 
addressed me in French, as I was not familiar with Turkish. 

In his personality Majidi Sultana may be said to combine 
the soldier, courtier, and scholar, for he is a man of great per- 
sonal bravery, a stern commander, yet extremely gentle and 
finished in his manners, endowed with great fondness for his- 
tory and literature. His soldierly qualities are looked upon 
by the people with a respect approaching awe, and his military 
ability, quickness of decision, and promptness of action have 
won for him the post he holds. Urumiah is close to the 
Turkish border and the territory of the Kurds, and this district 
is particularly subject to danger from outlaws and marauding 
freebooters, as the Kurdish raid in 1880 proved. The Kurds 
in fact are even now a constant menace, and Majidi adopted 
the policy of employing a company of these warlike hillsmen 
themselves for the purpose of keeping order. Yet I have since 




MART MARIAM CHURCH, URUMIAH 
(Where one of the Magi is said to be buried) 




A TYPICAL GROUP OF MULLAHS AND SKIPS 



NEW YEAR'S CALLS 101 

learned that he nearly lost his life in holding a conference with 
the representatives of a rebel Sheikh from over the border. 
A parley had been agreed upon, but in the midst of it some 
threat of violence was made by the rebels ; before the band 
could carry out their purpose Majidi Sultana shot the leader 
on the spot and had the rest seized and blown from the can- 
non's mouth in the public square of Urumiah. 

In extending hospitality to us, Majidi Sultana was the 
polished gentleman above the soldier, and his home bore evi- 
dences of culture and scholarly tastes. Besides finely bound 
Persian and Arabic works his library contained a few standard 
French books, and he seemed to take a real interest in history. 
Knowing his breadth of view and freedom from fanaticism I 
had no hesitation in making inquiries about the Zoroastrian 
religion, and I soon found, from the questions which he himself 
put to me regarding the association of Zoroaster's name with 
Ardabil and Mount Savalan, that his range of reading in the 
Oriental writers on his shelves must have been considerable. 
The decoration of his rooms showed taste, and his collection 
of curios and antiques was an interesting one. At my depar- 
ture he promised to send two Kurdish guards to accompany 
me for three days when I should start on my journey south- 
ward toward Hamadan. 

After paying this visit to the governor the next ceremony 
was to call upon one of the native Khans, a lord of several 
villages. Tea, tobacco, and sweetmeats formed part of the enter- 
tainment. Following that came a visit to a Mullah, or Moham- 
medan ecclesiastic, which completed the afternoon, for Persian 
calls last long. The Mullah was a kindly old fellow and knew 
two words of French, bon jour, which he followed up with u^ 
Persian greeting, 'Your Worship is an iirumiriatio'n to my 
eyes!' In this latter salutation, however, there wss^ab^ele^ 
ment of pathos, for the poor man had become almost totally 
blind a few weeks before. He had been suffering from cataract 
and had submitted to an operation at the hands of a native 



102 URUMIAH, A SUPPOSED EARLY HOME OF ZOROASTER 

itinerant quack which had cost him his sight. I could but 
admire his nerve and courage as he quietly told how he held 
his head perfectly still without flinching while the charlatan 
thrust a needle into the pupil. We prolonged for some time 
our visit with the Mullah and his company of Seids, or Descend- 
ants of the Prophet, until it was time to go, and then we 
walked on foot through the streets, in order that I might see 
more of the town. 

Urumiah has no public edifice of any importance, but it has 
a church of great antiquity associated with the Biblical story 
of the Magi and indirectly with Zoroaster. This is the old 
Nestorian church of Mart Mariam, the Blessed Virgin, now 
a Russian sanctuary. Popular tradition says that in the crypt 
lie buried two, or at least one, of the Magi who came to wor- 
ship the infant Christ at Bethlehem. The church may, there- 
fore, be regarded as an Eastern rival of the Cathedral of 
Cologne, besides having Persian competitors in Savah, Avah, 
and perhaps Kashan. 1 

A legend regarding these Wise Men and the origin of the 
church was told me by one of the ladies of the American Mis- 
sion, who had it from her Syriac teacher at Urumiah. I recog- 
nized it at once as a form of the story which is found in the 
apocryphal New Testament Gospel of the Infancy, recounting 
how the Wise Men received from Mary the swaddling-clothes 
of the infant Christ as a gift, and on returning to their country 
made a fire with which to worship and consume the clothes ac- 
cording to their religion, but the blessed garments remained un- 
seared in the flames and became a sacred relic. A church was 
built in commemoration of the miracle, and it is this edifice 
%h'$ serves 2aS;|He Jburial-place of the Magi. For convenience I 
reproduce t he "narrative as it appears in the New Testament 
^ppor^phfl^ s&i4,& ,i is also of interest because of the ex- 
press statement that the Wise Men came from the East in ac- 
cordance with a prophecy delivered by Zoroaster (Zoradascht). 

1 See p. 412, below. 



A CHURCH WITH A LEGEND OF THE MAGI 103 

1 1. And it came to pass, when the Lord Jesus was born at Bethle- 
hem, a city of Judaea, in the time of Herod the King; the wise 
men came from the East to Jerusalem, according to the prophecy 
of Zoradascht, and brought with them offerings : namely, gold, frank- 
incense, and myrrh, and worshipped him, and offered to him their 
gifts. 2. Then the Lady Mary took one of his swaddling clothes 
in which the infant was wrapped, and gave it to them instead of a 
blessing, which they received from her as a most noble present. 
3. And at the same time there appeared to them an angel in the 
form of that star which had before been their guide in their 
journey; the light of which they followed till they returned into 
their own country. 

4. On their return their kings and princes came to them inquiring, 
What they had seen and done ? What sort of journey and return 
they had ? What company they had on the road ? 5. But they 'pro- 
duced the swaddling cloth which St. Mary had given to them, on 
account whereof they kept a feast. 6. And having, according to the 
custom of their country, made a fire, they worshipped it. 7. And 
casting the swaddling cloth into it, the fire took it, and kept it. 8. 
And when the fire was put ont, they took forth the swaddling cloth 
unhurt, as much as if the fire had not touched it. 9. Then they began 
to kiss it, and put it upon their heads and their eyes, saying, This 
is certainly an undoubted truth, and it is really surprising that the 
fire could not burn it, and consume it. 10. Then they took it, and 
with the greatest respect laid it up among their treasures.' x 

This story of the Gospel of the Infancy and the legend con- 
nected with the church in Urumiah seem to be old Nestorian 
traditions. 2 

Still another legend connecting Zoroaster's name with 
Urumiah is recorded. It is referred to by Spiegel, who states 
that in the mountain of Buzo-daghi, ' Calf Mountain,' to the 
northeast of the city, a cave is shown in which the Prophet 

J New Testament Apocrypha, In- New Testament, p. 38, London, 1820, 

fancy, 3. 1-10. See also Walker, states that ' La Crosse cites a synod at 

Apocryphal Gospels, pp. 100, 103, Angamala, in the mountains of Mala- 

Edinburgh, 1870. bar, A.D. 1599, which condemns this 

2 The Gospel of the Infancy is cur- Gospel as commonly read by the Nes- 

rent, for example, among the Nesto- torians in that country (i.e. India).' 
rians in India, and Hone, Apocryphal 



104 URUMIAH, A SUPPOSED EARLY HOME OF ZOROASTER 

Zardusht is said to have lived. 1 This would point apparently 
to the tradition of Zoroaster's living for a time as a hermit in a 
cave in the Persian mountains, but I myself have not been able 
to get any direct information regarding this story, and a native 
of Urumiah has since told me that he knows of the place only 
as sacred in the eyes of the Mohammedans as 'All's Spring.' 
I was later informed, however, that there is a rock cavern 
near Maraghah on Mount Sahand which is pointed out as the 
cave of Zardusht. 2 

Among the observances at Urumiah is one which the natives 
regard as a reminiscence of ancient sun-worship the blowing 
of a horn and the beating of a kettle-drum at sunset. The 
place where this curfew signal is given is called the Nakdrah 
Khdnah, 'Band-Tower,' and is situated near the Darwaz-i 
Ark, the gate where a citadel once stood. The horn that is used 
is more than six feet in length, and after the drum is beaten 
for the third time (uch tabil in Turkish), no person is al- 
lowed abroad in the streets except under liability of arrest. 
A similar custom is in use also at Isfahan, Teheran, Meshad, 
Bokhara, and elsewhere, and I am inclined to think that it has 
nothing more to do with sun-worship than has a sunset gun or 
a curfew bell. 3 

The bazaars in Urumiah are in no way remarkable ; but 
some of the shops carry a fair supply of European articles, one 
of the best shops being kept by an Armenian. The streets 
have none of the press and throng of people that mark the 
more densely populated towns. The number of inhabitants is 
variously estimated between fifteen thousand and forty thousand, 
the difference in the estimates being largely due to whether the 
surrounding villages are taken into account or not. The major- 
ity of the population are Persians ; the remainder is composed 

1 Spiegel, tfranische Alterthums- to the caves see p. 61, n. 1, above, and 
kunde, 1. 131, n. 3, Leipzig, 1891. p. 173, n. 1, below. 

2 For traditions regarding Zoroas- 8 For the usage in Teheran and other 
ter's hermit life, see my Zoroaster, places, see p. 267, below, and Curzon, 
pp. 34, 189 e, 194, n. 1. With regard Persia, 1. 164, 174, 309, 350 ; 2. 27. 



THE NESTORIANS 105 

of Turks, Afshars, Assyrian Nestorians, some Armenians, a 
few Jews, and a colony of Europeans. 

A special historic interest is attached to the so-called 
Nestorian Christians. They are not of Persian blood, but 
originally Syrians, or rather Assyrians, a term which they 
themselves prefer, or Chaldseans, as the French call them. 
They are descendants of the ancient followers of the Christian 
bishop Nestorius, who was excommunicated in the fifth century 
for holding unorthodox views concerning the divinity of Christ, 
not regarding him as the God-man, but separating his human 
personality from his divine nature. The adherents of Nes- 
torius spread first into Persia, then far and wide through Asia, 
carrying their sectarian doctrines with them. In recent years a 
number of these Persian Nestorians from Urumiah have come 
as immigrants to America, which they regard as a second 
Eldorado, and where they have found occupation in carpet 
shops, hat factories, and other industries. There is a colony of 
them in New York and in Yonkers, and as a rule they have 
proved themselves honest and faithful workers, anxious to 
avail themselves of the educational opportunities offered by 
the night-schools, and keeping up their connection with the 
Christian church. 

A description of Urumiah would be incomplete without a 
notice of the Christian missions and their work in this difficult 
and dangerous field. America, France, England, and Russia 
are all represented in the cause, and some work is done by 
Germans and Swedes in the villages of the Urumiah plain. 
The Americans were the first on the ground, in 1835, under 
the American Board of Foreign Missions. For seventy years 
they have labored with devoted zeal, teaching, preaching, help- 
ing the poor, and ministering to the sick, for the medical dis- 
pensary plays an important part in the activity of the Mission. 
The buildings and grounds of the Mission itself, though simple 
and unassuming, are an object-lesson in making a place neat, 
homelike, and attractive ; and a separate department is set 



106 V RUM 1 AH, A SUPPOSED EARLY HOME OF ZOROASTER 

aside as an office for the printing-press which aids in the evan- 
gelical and educational work. Special facilities for female edu- 
cation are given in the Fiske Seminary, where girls receive a 
good schooling ; and opportunities for higher study are offered 
to young men in Urumiah College, which was founded by 
the missionaries many years ago. It is situated outside the 
city, about twenty minutes' ride to the southwest, in a richly 
wooded enclosure, like a park, which served as a place of 
refuge for a large number of native Christians during the 
Kurdish raid. Here are gathered the various buildings of the 
institution halls, plain but serviceable, recitation-rooms, simple 
but neat, a small museum, a library and offices, and last but not 
least the clinic and medical dispensary under Dr. Cochran's 
supervision. 

The hospitality which I enjoyed also at the English Mission 
gave me an opportunity to judge of that good work as well, 
and the same may be said of the representatives of the other 
Christian creeds in their respective spheres. At the cost of 
great personal sacrifice, and even risk of life, they are doing 
their share to fulfil the commandment which bids them to 
preach the gospel to all nations. To the Mission friends, one 
and all, who were so kind to me during my stay I feel deeply 
grateful. 

My heart was quite full as I mounted my horse to join 
my pack-caravan, .led by Shahbas, the chdrvdddr, Safar, my 
faithful servant, some attendants, and the two guards provided 
for my safety on the journey by the governor, Majidi Sultana. 
They were both well mounted, but Safar's horse resembled the 
rake which the Clerk rode in the Canterbury Tales. The 
horse upon which Shahbas sat cushioned for his pack-saddle 
(pdldri) was a heavy mattress of stuffed straw looked better 
fed, but he had an uncanny fashion once in a while of getting 
his left forefoot out of joint, then hobbling a few feet till 
he stopped or fell, unless chance meanwhile twisted the dis- 
located joint back into place. The pack-horse was a sturdy 



MY CARAVAN 107 

gray stallion, and his color led to our calling him Kabud, liter- 
ally 'blue.' My own animal was a small but good horse, so 
I dubbed him Rakhsh, after the famous charger of Rustam, 
much to the amusement of the Persians that heard the name. 
When all was ready the signal was given, and away our caval- 
cade started on the twelve days' journey to Hamadan. 



CHAPTER X 
BY CARAVAN AND CAVALCADE 

'Where beasts and men together o'er the plain 
Move on a mighty caravan.' 

WORDSWORTH, Descriptive Sketches, 8. 

4 MESHADI, Mesh-a-di, Mesh-a-a-d-i-i ! ' in a gentle crescendo 
and with musical accent cries the voice of our Persian servant 
to awaken the naib, or master of the caravansarai. A muffled 
ball, bali, c yes, yes,' responds from a distant corner of the 
walled enclosure, and in a few minutes from somewhere in 
the darkness there peers into the mud-built sleeping-room 
the owner of this Oriental title. In reality the appellative 
Meshadi designates a Moslem who has made the pilgrimage 
to Meshad, if not to Mecca. In practice, however, it is often 
applied loosely, like our colonel, major, or professor, and fre- 
quently has little more dignity than the colloquial American 
'boss.' Yet Meshadi, Husein, or whatever his name maybe, 
is probably a tall, dignified individual, often handsome, with 
rather tine features, a chiselled nose, and a broad forehead 
surmounted by a high Persian cap. In his veins still flows 
some of the blood that made great the race of Cyrus ; but he 
is slow, frequently shiftless, at the same time generally pos- 
sessed of an eye to the main chance, but always kindly in his 
quiet way. 

The camel-drivers are already up and starting before day- 
light. It is high time to be off. The dull dong, dong not 
ding, dong, for there is no variety coming from the huge 
bells attached to the dromedaries and mules, tells that the cara- 
van-train, or Persia's Twentieth Century Express, is on the 
move. But he who wishes to travel 'fast,' as there are no 

108 




ti - 

fc "3 



PQ 




OF THF 

UNIVERSITY 

OF 




PREPARATIONS AT DAWN 109 

railroads in Persia, must have his own cayalcade of horses, as 
I did, eked out sometimes by mules or even by donkeys. 
If the route lies on the beaten tracks, one may resort to post- 
horses and go chdpdr. But posthaste in the kingdom of the 
Shah does not mean much, as a rule, and patience unbounded 
is a necessary prerequisite for the traveller who does not wish 
to grow gray with anxiety and loss of temper at delays. 

The sun is not yet up ; murky darkness still floods the 
caravansarai. The small sleeping-room above, or bald khdnah, 
as the upper chamber is called, is lighted by the flicker of a 
primitive palm-oil lamp, or by a candle, if the voyager has 
had the forethought to purchase one at the last bazaar. All 
this while, however, the man in charge of the wretched hos- 
telry has been silently standing at the rickety door waiting 
for orders. But your first inquiry is 4 Where are the horses ; 
are they ready ? ' asphd kujd and, asphd hdzir hastand ? The 
unfortunate necessity of asking this question early will be 
learned with time, if time exists in almost clockless Persia. 
Only too often it happens, unless one has taken the precaution 
to have the animals tied or properly hobbled, that the horses 
have wandered off during the night to a distance of three or 
four miles, grazing on the plain. With camels there is per- 
haps less chance of this, as the wary camel-driver has forced his 
ungainly beasts to lie down in a crowded circle around the 
bundles of hay that serve as a magnet to keep the nose of the 
ship of the desert duly pointed toward the fodder pole. With 
the fleet little Persian horses the case is different. The keen 
strain of Arab blood keeps up their love of the desert, and at 
the slightest provocation, unless controlled, they scamper miles 
away on the plain. 

The horses must, therefore, be recovered, and the ncCib of 
the caravansarai summons his head man to perform the task. 
He responds with the conventional chashm, implying, ' My 
eyes are the forfeit if I fail to do your Worship's bidding,' and 
disappears in the darkness. By experience it will be found 



110 BY CARAVAN AND CAVALCADE 

well to promise a silver tip to the whipper-in's lash, if he will 
only secure the renegade horses betimes, so that the day's 
journey may begin. This attractive promise is sufficient to 
call forth a lazy stretching also on the part of a second, a third, 
or a fourth attendant, each of whom slowly throws off the 
lethargy of sleep and declares that he stands ready to saddle 
and give extra barley to the horses on their return. In fact, 
one of the group gives assurance that the animals are already 
near-by. God save the phrase ! 4 Near-by ' may mean half 
an hour or an hour or more of the traveller's patience, who 
thinks vainly of American speed and the time-table rush of 
the latest Chicago Limited. 

The intervening time, however, has not been vacant, though 
darkness still prevails. Our own excellent and faithful man- 
servant Safar, who has learned through Western contact the 
meaning of the word ' hustle ' and has caught the enigmatic 
force of its indefinable spell, has been busy every moment. In 
an instant he has kindled a fire with a skill that is the despair 
of one who looks upon the combustionless combination of damp 
fagots, sedge grass, brier bush, and perhaps pancakes of manure 
fuel. Fresh water (ab-i tamlz) has been brought by invisible 
hands evoked out of the darkness. The samovar, or Russian 
tea-urn, which forms part of the advance guard of encroaching 
Russian civilization, is already simmering. A moment later, 
while you are adjusting your riding-leggings, the faithful Safar 
has an improvised breakfast of some sort ready to be served on 
the mud floor of the room. 

The bread consists of huge leathern aprons of dough, which 
I have already mentioned in describing the extraordinary pro- 
ductions that result from a combination of Persian wheat and 
old Iranian ovens. The loaves are not loaves in our sense, but 
enormous flat pancakes, two feet long, a foot or more wide, and 
of the thickness of a griddle-cake. In baking they are deftly 
slapped against the side of an earthen jar or oven sunk in the 
floor of some dingy living-room or of a real Persian bakery in 



AN EARLY BREAKFAST 111 

the bazaar. For travelling they are the most convenient article 
of food imaginable, for they serve not only to be eaten, but also as 
a wrapper to fold up the knife and fork, a chicken, sweetmeats, 
or what not, just as one would use heavy brown paper. This 
Persian bread, called nan or nun, according to the dialect, is 
generally moist when served and often a bit soggy; but when 
allowed to grow dry and crisp, it is excellent to the taste, 
though sometimes fatal to the digestion. 

But all this talk or description of breakfast is after all only 
a digression ; for if one is in haste, the breakfast often consists 
only of a raw egg, some bread, and a good cup of tea, drunk 
from a glass crammed half full of sugar. When time allows, 
fortune may add a bit of roast meat (kabdb) or chicken, and 
two or three ginger-snaps as a special treat. The latter are 
the cherished remains of the parting box, or present for the 
roa.d, given instead of a stirrup cup by the good housewife 
of the missionary whose hospitality it has been a privilege 
to enjoy. 

Time has all the while been elapsing. The stray horses 
have at last been captured, we do not know how far away. 
Their thin -shod hoofs have a cheery ring as they approach the 
mamilt or halting-place of the caravansarai. The pack must 
therefore be made ready to load upon the draught-horse. This 
is the signal for commanding the mafarashband to be prepared. 
This piece of baggage consists of two huge oblong carpet 
pockets shaped like chests and bound round with stout goat- 
hair ropes. All the utensils, including the folding camp-bed, 
must be tumbled into these receptacles, with an eye, however, 
to equal balance. When both are firmly lashed, the sturdy 
chdrvddar, or caravan-leader, skilfully lifts the first and after- 
ward the second on his back and staggers beneath the burden 
down the unequal mud steps or lowers it from the roof to the 
court below, and deposits his load on either side of the pack- 
horse or mule. With signals, hoists, grunts, pulls, tugs, shoves, 
and punches, the loads are finally lifted on the animal's back 



112 



B 1* C A R A VA N 



CA VA L CA D E 




LOWERING LUGGAGE FROM THE 
HOUSE-TOP AT DAWN 



and properly adjusted. Then 
conies the tying. A Persian 
can give points to a jack tar 
in the matter of knotting a 
rope ; but despite all the 
skill the load sometimes does 
slip, and it is well to add 
or subtract a tip at the end 
of the day according to the 
way in which the pack was 
fastened. 

Saddling likewise takes time, for everything except money 
goes slowly in Persia. All the native members of the caval- 
cade, man-servant, muleteer, guide, and armed guards, have 
their horses' girths passed through iron rings and cinched. 
Only the foreign Sahib, or Master, with his European saddle 
has recourse to buckles. Delays are still the order of the hour, 
and nothing but an oft-repeated zud, zud, tez, tez, 'quick, 
quick, hurry up,' followed by a threat and then a promise of 
silver, furnishes the means of securing even the slowest haste. 
Among the animals of the cavalcade the Sahib's horse is 
supposed to be the best, which is not always saying a great 



THE START 113 

deal ; but usually better still is the horse of the head guard, 
who owns his own steed, which he proudly claims to be an 
Arab. The pack-horse generally has no rider. The servant 
takes the next best horse, and the shdgird, or postilion (if one 
may dignify this individual by so high a title), takes what is 
left. But the last shall be first, as experience will prove, for 
the postboy usually manages to reserve for himself one of the 
better horses out of the poor lot, and if he does not, there is 
little gain for the traveller. No train goes faster than its last 
car, and the caravan-leader is a bad man to leave behind. 

At last we are ready to start. Nearly two hours of time 
have been used up, largely with unnecessary delays. ' Quick, 
hurry up ! ' is again the command. But payment for the 
night's lodging must not be forgotten. It is paid to the over- 
seer of the manzil. A gratuity given to him generally means 
the same as a tip to all as to the head waiter in the Vienna 
cafe system. If the gift be large enough to meet the general 
approval, a combined good wish, khudd hdfiz, 4 God be mindful 
of you,' greets the departing company, and a word of praise, 
khaill khub, ' very good,' is bestowed upon the best horse and 
serves as an omen of promise for the journey. 

As we ride out through the low door of the courtyard, dark- 
ness is just beginning to vanish. For safe travelling it is 
always well to have the light of day. We find ourselves 
a moment later in the midst of a passing caravan. We join 
the motley assembly of camels, horses, mules, donkeys, mule- 
teers on foot, pilgrims, and merchants, that make up this 
heterogeneous company. The never ceasing dong, dong of the 
big copper bells, the rasping bray of the little donkeys that are 
wide awake, but staggering beneath crushing loads, the dull 
pace of the wag-eared mules prodded on by their drivers, the 
chatter of the men, and the peculiar odor of the scraggly- 
coated camels, make a deep impression on the senses. The 
gray streaks of dawn now lighten in the horizon. The sky 
takes on a more silvery hue. Night withdraws her star- 



114 BY CARAVAN AND CAVALCADE 

bespangled fan and reveals the blush of dawn. The night- 
ingale's song is hushed before the carol of the lark. And out 
of the cavern of the hills the sun rises in splendor to shed its 
glory as of old over the ancient land of Iran. 1 

Such is the composite picture left on my mind by many 
weeks of caravaning through Persia ; but few of my earlier 
journeys were started amid sunshine and the carol of birds. 
Winter still reigned during most of the time I was in Azarbai- 
jan, and persistently refused to give up his sway ; the calendar 
pointed to the end of March when I left Urumiah, but spring 
was a long way off, and snow and bad roads were a constant 
source of delay on the journey and were destined to be so for 
a fortnight and more to come. 

For the first three days after leaving Urumiah I had the com- 
pany of a Nestorian friend, Rev. Yaroo M. Neesan, whom I 
had met in America a number of years before, when I began 
my Persian studies, and who is now a priest in the Anglican 
Mission at Van. He knew the country round about as only a 
native can, and his fund of information and stock of anecdotes 
were inexhaustible. I still can see the humorous twinkle in 
his eye as he put on his cartridge-belt and pistol, before we 
mounted, and said, patting it affectionately, 'It is more 
blessed to give than to receive (bullets) ! ' 

To beguile the time as we rode along, he told story after 
story in a manner that would have delighted Chaucer's heart. 
It was now an account of the old woman whose cleverness saved 
the citadel of Van in time of siege by bidding her beleaguered 
countrymen adopt the strategy of pouring down from the 
citadel showers of fine white ashes. These the enemy took for 
flour and abandoned the siege, because it seemed useless to 
beleaguer a fortress so well supplied with food that its oc- 
cupants could thus lavishly throw away flour. Next it was 
the tale of the witty vizir whose artful dodge made Shah 
Abbas pay for the dish of cherries which he himself had eaten ; 
1 See my letter in the Evening Post, New York, October 3, 1903. 



STORIES BY THE WAYSIDE 115 

or, once again, an exciting narrative of Yaroo's own experience 
when attacked by bandits some years before, while returning 
from Sauj Bulak a reminiscence called forth by passing a 
pile of stones heaped up by the roadside to mark the spot 
where a native had been murdered a month before. The soft 
piping of a Kurdish shepherd in the hills turned the conversa- 
tion to pastoral life. Neesan had tended the flocks in his youth, 
and he told how the shepherds, as in the Bible, knew every 
sheep by name, how they would sometimes seek shelter from 
the cold by sleeping in the mud enclosures built to protect 
the sheep at night on the plain, and he described some of the 
primitive shepherd customs handed down from the earliest 
antiquity. All the birds that hovered about, the long- tailed 
magpie, thrush, and crested lark, seemed to be familiar friends, 
while the eagles and kites soaring above our heads were old 
acquaintances of his. And so the journey proceeded. 

Saatlu, near Diza-Takiah, was our resting-place for the first 
night. Here the courtesy of our native host, who was a rela- 
tive of Mr. Neesan, appeared to contrast strangely with the 
fierce appearance given him by a huge dagger at his belt and 
the gun which lay by his side at night to forestall a surprise or 
raid. The equipment of his arsenal seemed to be plentiful, but 
no less abundant was the supply of provisions in his larder. 
From these he gave us in profusion at supper that evening, and 
it was then that I noticed for the first time the fondness of the 
Persians for melted butter (roghari). A new significance was 
thus imparted to the Avestan passage which describes how 
zaremaya raoghna, c clarified butter of springtime,' is the first 
food which the faithful who have died in Zoroaster's creed 
receive when they enter into Paradise. 1 

Before daybreak I was awakened by a messenger who 
entered my room armed with gun and pistol. He had 
come from Urumiah and had ridden through the night to 
deliver to me a cablegram which had reached the city after 

i Yt. 22. 18. 



116 BY CARAVAN AND CAVALCADE 

sunset. The city gates were closed, and the guards, taking 
him for a horse-thief, had refused to let him pass until he pre- 
sented his credentials, when he was allowed to hasten on his 
way to overtake my caravan. The cable had become much 
garbled in the course of transmission, but happily the cipher 
contained good news from my distant home. 




A PERSIAN TELEGRAM 

By this time day had broken and after a good breakfast in 
Persian style our pack was put in order by the servants, and 
we were ready for an early start. Neighing horses, barking 
dogs, and an oft-repeated khudd hdfiz from our host and his 
friends sped us onward upon our journey along the shores of 
the blue lake of Urumiah. 

The village of Mahmadyar, a place of about one hundred 
houses, was our next night's lodging after ten hours on the 
road, and we again enjoyed native hospitality, this time in a 
large living-room filled with men, women, children, and smoke. 
Owing to the cold weather and the rain, all were huddled 
together around the urn-shaped hole in the mud floor contain- 




A CARAVAN OF NOMADS 




KB* 



CHACHILI HILL, NEAR URUMIAH 



A NIGHT AT MAHMADYAR 117 

ing some smoldering embers. A baby in its mother's arms 
was choking with the croup, and another small tot was gasp- 
ing with whooping-cough. The rest of the children, however, 
looked healthy and kept constantly running in and out, chatter- 
ing and playing, while the women were occupied with their 
work. The women, like the men, were dressed in a common 
blue cotton cloth, usually with their feet and limbs bare to the 
knees, and wore a red kerchief about the head. Their faces 
were not veiled, because the village women, especially among 
the Kurds, do not conceal their features, and as these women 
were Armenians and Nestorian Persians, they moved about with 
greater freedom from restraint in the presence of the men. 
The men, meanwhile, were holding a conference and drew 
nearer together to discuss a matter requiring important con- 
sideration. A band of Kurds had come down and carried off a 
young Armenian girl from the village, and had disappeared 
with the captive over the mountains. The deliberation of the 
council resulted in maturing a plan for restitution or revenge, 
and an appeal was also sent to the governor at Urumiah ; but 
I never heard what the outcome of the affair was. 

Early in the morning my caravan was on the march, and I 
had an opportunity of making what I believe to be an interest- 
ing identification in connection with Zoroaster's ministry 
locating the district where he most probably made his first 
convert. This was near the village of Khor Khorah, between 
Mahmadyar, Daralak, and Miandoab. From tradition we 
know that the first adherent won by Zoroaster for his religion 
was his own cousin, named Maidhyoi-maonha in the Avesta, 
and Medhyo-mah in Pahlavi. 1 The scene of the conversion 
is located by the Pahlavi writings of Zatsparam ' in 
the forest of reedy hollows, which is the haunt of swine 
of the wild boar species.' 2 Now we know that Zoroaster 
passed much of the earlier part of his life in the region of Lake 

1 On the tradition of Zoroaster's first cf . Rosenberg, Livre de Zoroastre, p. 24. 
convert, see my Zoroaster, p. 54, and 2 Zatsparam 2. 38. 



118 BY CARAVAN AND CAVALCADE 

Urumiah (Av. Chaechasta), and that the southern shores of 
this lake, from times of antiquity, have been covered with great 
tracts of reeds. The map to-day shows a ' Forest of Reeds,' 
some sixty miles in extent, and I had already heard accounts of 
this when I was in Tabriz and Urumiah. The 'forest of 
reedy hollows ' in the Zoroastrian tradition is evidently iden- 
tical with this. The highroad which leads around the lake 
and toward Ragha (now Rei) near Teheran, where Zoroaster's 
mother is said to have been born, passes along its very edge. 
The region abounds in 'hollows' and 'reeds,' and I saw 
immense masses of the slender stalks, some of them fifteen or 
twenty feet high, cut from the ' forest,' and I heard a number 
of stories from people who had been lost among them in the 
marshes. In this region, likewise, ' the swine of the wild boar 
species abounds,' and the animals are hunted for sport by the 
natives. They are not eaten, however, by the Mohammedans, 
to whom the flesh of the hog is forbidden by the Koran, but by 
the Armenians, who have no such religious scruples in the 
matter and derive actual benefit in the way of food from the 
chase. 1 The surroundings of the place combine with what we 
know of Zoroaster's life from traditional sources the A vesta, 
Zatsparam selections, Zartusht-Namah, and other works to 
make it reasonably probable that this 4 Forest of Reeds,' south 
of Lake Urumiah, is the region at least to which the passage 
refers, even if we cannot identify the precise spot. 2 
, At noon that same day, after luncheon at Daralak, I was 
obliged to lose Mr. Neesan's good company, as his missionary 
duties called him to Sauj Bulak. I parted from him with 
regret, and the afternoon was lonely without him, despite the 
novel experience of camel riding, which I tried as a change 
from horseback. The dromedary which I mounted was one of 

1 Cf. Koran (tr. Sale), chap. 2, p. 18 ; experiences in this * forest of grass and 
5, p. 73, etc. reeds,' compare Perkins, Eight Years 

2 See my article in JAOS. 25. 183- in Persia, pp. 193-194 ; consult also 
184 ; and, for a description of some Wilson, Persian Life, p. 106. 




S ARAL AN, A VILLAGE OF MUD HOUSES 




KHOR KHORAH, NEAR THE FOREST OF REEDS 
(Probably the district where Zoroaster made his first convert) 



TRADITION OF ZOROASTER'S FIRST CONVERT 119 

the finest specimens of camel flesh that I saw in Persia, but I 
was glad to change back again to my horse, as the creature's 
gait was atrocious and one needs to be a good sailor to navi- 
gate with success the ship of the desert. 

Sunset found my caravan entering Miandoab, a town of sev- 
eral thousand inhabitants, well named 'between two rivers' 
(midn do db) from its situation between the Jagati and the 
Tatavu. 1 Here I received a pleasant visit from an Armenian- 
Persian for whom I had brought letters from Urumiah. As 
we chatted on various subjects I told him that I had been un- 
able thus far to meet with any followers of Zoroaster's creed 
in Azarbaijan. He replied that this was not surprising to him, 
for he knew of but one Fire- Worshipper (Atash-Parast) in the 
district of Miandoab, and that man called himself a Babi, 
a follower of the religious reformer whom I have previously 
mentioned. It may be possible that this ' Babi ' may have 
actually turned from Zoroastrianism to Babism, but more likely 
that he had chosen to conceal his religion under the garb of 
that eclectic faith for fear of persecution from fanatical Mo- 
hammedans, at whose hands non-acknowledgment of the Mos- 
lem creed may mean cruel oppression, and apostasy from 
Islam, death. 

Miandoab itself lies in the undulating region southeast of the 
lake, at a height of four thousand two hundred feet above sea- 
level ; and from this point the territory begins to be hilly and 
leads steadily upward through fine scenery in the high passes 
of the Mian Bulagh Mountains to Sanjud. Before reaching 
these we had to cross the Jagati River, which we did next day 

1 1 did not know until after my in Zt. f. Ethnol. ( Verhandl. Berl. An- 

return to America that at Dashtapah, throp. Gesellsch. ) 26 (1894), pp.479 seq. 

near the reed forest in the neighbor- (cf.Streck, Zt.f.Assyr. 14. 144). There 

hood of Miandoab, there is a cunei- is also a rock-hewn cave in the same 

form inscription on a rock, and there region (Wilson, Handbook of Asia 

is said to be one also in the vicinity of Minor, p. 324, and de Morgan, Mission 

Sauj Bulak ; cf. Wilson, Persian Life, Scientifique en Perse, 4. 294-296). 
pp. 99, 105 ; Belck, Inschr. Taschtepe, 



120 BY CARAVAN AND CAVALCADE 

in a huge scow that was large enough to ferry our horses and 
all across. I expected to make an extra stage' in that day's 
journey, as I had become thoroughly accustomed to early 
starts and long marches ; but I only reached Kashavar, a vil- 
lage with an old fort, at which we arrived about three in the 
afternoon. Here I came into violent collision with my cara- 
van-leader Shahbas, because of his negligence and laziness. 
Every one who travels in Persia is familiar with the struggle 
one usually has to go through with his chdrvdddr early in the 
journey, until it is settled once for all that the master is 
master and will tolerate no slackness, although ready promptly 
to reward good service. Shahbas had yet to learn this lesson 
and be cured of some of his shortcomings, and the first dose of 
the medicine, which varied from gentleness and coaxing, bribes 
and rewards, to sternness and chastisement, was administered 
at Kashavar. This needed to be repeated a few times after- 
ward, and with excellent effect, so that by the end of a month, 
when I dismissed Shahbas to return to Urumiah, he had be- 
come quite a model, as far as a Persian muleteer can be, and 
was well rewarded for his improvement. 

My man-servant Safar, on the other hand, who had been 
capable from the outset, showed himself more reliable and 
valuable every day. The effect of his training among the 
missionaries manifested itself in many ways, and his quickness 
and intelligence were a delight to me. To improve his Eng- 
lish and my Persian, we used sometimes at night to read and 
translate a simple story-book which had -been given him, and 
with his English vocabulary his manners also seemed to im- 
prove. He learned how to be polite in his address, and his 
first blunt ' yes ' soon gave place to a more courteous ' yes, sir,' 
and he knew what it meant to say 4 thank you ' - a phrase not 
common among the Persians of the lower classes. His busi- 
nesslike methods in looking after my interests came more and 
more to the front ; he found ways I should never have 
thought of for saving me money, and was quite impatient at 



FROM MIANDOAB TO SANJUD 121 

my readiness to pay rather than debate a price. His honesty, 
moreover, was unimpeachable, and he was likewise always help- 
ful in arranging the details of the journey. The next stages 
of the march proceeded well, except when my horse Rakhsh 
was taken with a chill and cramps in the mountain pass after 
leaving Sain Kalah, 1 and I feared I should have to leave the 
animal dead on the road where he lay. Shahbas, however, 
worked over him, and we managed to pull Rakhsh through, so 
that we reached Sanjud all right, though late in the afternoon. 

Sanjud, where I passed that night, April 2, is situated on the 
side of a steep ravine amid masses of rugged mountains. The 
approach to it from the north lies over high ridges, through deep 
defiles, and past precipices, beneath which there run streams 
transformed into torrents by the rains and melting snows. 

The night was a cold one, but the panes of oiled paper in the 
windows shut out some of the wind, and the fire burned 
brightly. Around it were gathered a number of villagers, 
who made pretexts to come and interview the farangi, most 
often to beg remedies for various ailments. One old man, a 
Kurd, stayed all the evening. He seemed to be a worthy 
personage and offered me the company of his two sons to act 
as guides the next day, as Shahbas did not know the road to 
Ahmadabad and Takht-i Suleiman. The two appeared to be 
reliable fellows and of a sturdy type. I looked them over 
by the flickering light of a lamp made of an earthen bowl, 
filled with the oil of the castor plant, into which was stuck a 
piece of cotton twist to serve as a wick, for there was not even 
a tallow dip in the village. To be sure that they would come 
back in the morning, I took from the father a befr, or pawn, 
which he deposited as a guarantee for their appearance, and 
they departed with the promise to return betimes inshdllah! 
Before sunrise they both were on hand, ready to guide the way 
over the heavy passes. 

1 In the vicinity of Sain Kalah there the site of an ancient fire-temple ; see 
is reported to be a mound of ashes, Bishop, Journeys in Persia, 2. 197. 



122 BY CARAVAN AND CAVALCADE 

The snow lay two feet deep on the side of the hills and some- 
times three or four in the gullies. Twice that day the horse 
which I rode went down under me in its treacherous depths. 
The glare from its white surface, when the sun came up, was 
dazzling and set up a painful inflammation in my eyes, one of the 
first stages of snow-blindness. In places where the snow had 
melted, the mud was an obstacle to progress, so our advance 
was slow during most of the morning and afternoon. As we 
crossed the last ridge of hills the snow lessened, but the rough- 
ness of the road increased, since the path led for some distance 
along the rocky and precipitous bank of a river. The horses 
insisted upon walking as near the edge of the cliff as possible, 
and it was sometimes a marvel that they kept their footing. 
Rocks now and then almost totally barred the way and we had 
to descend to the river's brink or else surmount them. It was 
no wonder that the load on the pack-horse often became loose, 
causing a halt. I remember one jutting crag in particular, 
which stood like a giant buttress directly in our path. It was 
impossible to go around in front of it, and the only way was to 
climb up back of it. The angle seemed almost perpendicular, 
and to sit in the saddle was out of the question ; so we had to 
dismount and scramble up past the crag as best we could, 
tugging the struggling horses behind us. As darkness began 
to fall, the mountain gorges became dangerous because of rob- 
bers, and at each turn in the road it interested me to watch 
the two guards lower their guns to the saddle-bow, the one 
riding a short distance ahead, the other dropping behind. 
Thus we pushed on until after sundown, when we reached 
Ahmadabad, our manzil for the night. 

The place where we lodged adjoined a cow-stable, but it was 
a well-constructed building, the only one with two stories in 
the village. The master of the house was entertaining his 
brother and some friends as guests for the night, but he offered 
me a welcome if I was willing to share the room with the party, 
to which I gladly consented. I found that I had really to con- 



OF THE 

UNIVERSITY 

OF 

iFQfifii! 



A ROUGH ROAD TO AHMADABAD 123 

tribute my share also toward the entertainment, however, by 
being constantly on exhibition before the company, who plied 
me with questions till past midnight, though I should have 
been glad to sleep and to rest my inflamed eyes. At last all 
were ready to retire ; the Persians crept under the quilts that 
covered the kursi> or brazier, made like a low wooden table 
over the fire -hole in the middle of the room, while I took to 
the blankets of my camp-bed. 



CHAPTER XI 

TAKHT-I SULEIMAN, AN ANCIENT FORTIFIED TOWN 
IN RUINS 

* We scrutinize the dates 
Of long-past human things, 
The bounds of effaced states, 
The lines of deceased kings ; 

We search out dead men's words, and works of dead men's hands.' 
MATTHEW ARNOLD, Empedocles on Etna. 

THE sun had not yet drawn his sword from the scabbard of 
night to adopt an Oriental phrase before I was again in 
the saddle and ready to start from Ahmadabad to Takht-i 
Suleiman, or 4 Solomon's Throne,' a ruined fortified city which 
Rawlinson claimed to be the capital of Media. I obtained as 
guide a peasant who was more familiar with the site than my 
own guides, and we set off at a brisk canter as the sun began 
to show over the hills, disclosing to the view Mount Zindan, 
4 Solomon's Dungeon,' toward which we directed our course. 

The path lay along the crest of a volcanic ridge. This eleva- 
tion, formed partly of grayish-brown lava from the crater of 
Zindan, extends for two or three miles like the prostrate form 
of some huge giant whose head is Mount Zindan. There is no 
longer any life in the monstrous frame, but from the cavernous 
mouth of the hill there still arise blasts of fetid air, which lead 
one to wonder whether Mount Arezurahe Griva, ' Neck of the 
Demon Arezura,' the entrance to hell in the Avesta, may not 
have been an extinct crater like this prison of the Dlvs. 

The hard basaltic conglomerate of the ridge echoed with a 
hollow sound as we galloped over it, and a score of tiny warm 
calcareous springs bubbled up from a row of miniature craters 

124 








A MOUNTED GUARD 







A TYPICAL PERSIAN VILLAGE 



^ OF THE 

UNIVERSITY 

OF 



MOUNT ZINDAN 125 

and showed the igneous origin of the ridge. It took me per- 
haps half an hour to reach the base of Solomon's Prison Height, 
which rises to an elevation of nearly four hundred and fifty 
feet above the plain and is capped by a massive cone of con- 
glomerate produced by the discharge from the interior of the 
volcano. 1 The ascent to the summit of the rocky mass is very 
steep. On reaching the top I found the rim of the crater to 
be about three hundred feet in circumference, and as I peered 
over into the dizzy abyss I could but think of Milton's lines : 

* And in that lowest deep a lower deep, 
Still threatening to devour, opens wide. ' 

The natives, in fact, regard the funnel-like shaft of Zindan 
as a bottomless pit, and there is no question that its depth must 
be several hundred feet, judging from the time required for a 
stone to reach what seemed to be the bottom. Crawling as 
near to the edge as I dared, and looking down, I could discern 
near the bottom a patch of snow, which showed that that part 
of the crater is not warm now, but there was perhaps a still 
lower depth, invisible, from which issued the sulphurous fumes 
that were exhaled to the air above. 

The view from the peak of Zindan is a fine one. On all 
sides rise lofty mountains. One of these to the north, called 
Takht-i Bilkis, ' Throne of the Queen of Sheba,' towers sky- 
ward to the height of ten thousand feet, and on its summit (so 
legends say) King Solomon built a summer palace for his 
beloved. The ridges to the east are less high, but they com- 
bine with the others to form a huge caldron and shut in the 
plain, out of the midst of which rises a low hill of scarped rock 
crowned with the ruins of Takht-i Suleiman. The real size of 
the hill of the Takht is dwarfed by the height of the surrounding 

1 Throughout I have taken Zindan, the hill as a ' cinter ' cone and assigns 

and the ridge formed from it, to be of to the mountain an aqueous origin ; so 

igneous origin. Rawlinson (Journ. do also Wilson, Persian Life, p. 162, 

Roy. Geog. Soc. 10. 53-54), however, and Gordon, Persia Bevisited, p. 62, 

describes the scarped, rocky crest of following Rawlinson. 



126 TAKHT-I SULEIMAN, AN ANCIENT FORTIFIED TOWN 

mountains, but its appearance and the shape of the enclosure 
on its summit recalled to my mind the description of the 
Vara of Yima Khshaeta (Jamshid) in the Avesta, although 
tradition does not locate the Vara in this particular region. 1 

Leaving the horses at the base of Mount Zindan I started on 
foot with my guides over the snow toward the deserted ruin, 
half a mile or more distant. The sun was now some degrees 
above the horizon, and the glare from the crystal plain became 
excruciating to my inflamed eyes. I held my black hat before 
my face for relief, but it availed little against the blaze of light, 
and I hurried forward as fast as I could to reach my goal. As 
I approached them, the huge ramparts and colossal bastions 
of the massive pile, between thirty and forty feet high, came 
out in their true proportions. The entrance to the stronghold 
was once made by four great gateways, roughly facing the 
cardinal points. Not all the details of these were clear, since 
the snow was much drifted against them and the western ram- 
part is falling into ruins. On scaling the battlements at this 
point I could make out, more or less distinctly, the general 
outline of the entire oblong enclosure, which is about three 
quarters of a mile in circuit ; yet it seemed to me hardly pos- 
sible to reconstruct, even in imagination, any walls within the 
great stronghold itself, unless we should convert into such 
structures what I regarded as the outlines of a drive or cause- 
way. I mention this fact now because of its relation to the 
problem of the identification of the site of Ecbatana, which I 
shall touch on later, pointing out how important a part is played 
by the battlements. 2 

Within the enclosure the first feature that strikes the eye is 
the remains of two walls of stone and brick that once formed 

1 Cf. Bd. 29. 14 ; 29. 5 ; 32. 5. says, * Of the seven walls, one alone is 

2 See pp. 151-166, below. I have to be traced ; and even here the Median 
since found that Canon George Raw- structure has perished and been re- 
linson, following his brother, Sir Henry, placed by masonry of a far later age' 
in the belief that Takht-i Suleiman rep- ( Five Great Monarchies, 3. 27). 
resents the ancient Median Ecbatana, 



THE MOUNTAIN STRONGHOLD 127 

part of a rather high structure facing eastward. The side 
walls of this building, though shattered and crumbling, are 
still standing, and they show that the edifice could not have 
been less than a hundred feet long and forty feet wide. To me, 
even in the distance, they did not give the impression of great 
antiquity, certainly not of antedating the Mohammedan con- 
quest, and this fact was proved by the fragments of Arabic 
letters inscribed upon what had once been the portal. 1 

Some thirty yards to the north-northeast are the remains of 
a second structure, an arched and vaulted building partly 
sunk below the ground and now largely a mass of rubbish and 
debris. Its domelike outline rising amid the snowy surround- 
ings may be seen in my photograph, and its general position 
with reference to the other remains in a rough sketch I made 
of the general elevation at this point. Conspicuous are the 
two arched portals through which one descends to the vaulted 
chamber of brick below, and also a small tower projecting 
above the northern elevation. The bricks which were used in 
the construction of the edifice are nearly a foot square, 2 while 
the walls themselves are four or five feet thick and convey the 
impression that the vaulted chamber and its arched recesses 
once served as a place for preserving a precious treasure. 3 The 
natives call the building the hamdm, 4 bath'; 4 but its true iden- 
tification, as a great fire-temple of most ancient date, was first 
made clear by Rawlinson, as will be explained hereafter. 

The other remains of buildings within the citadel of Takht-i 

1 1 have since found my observations 2 Resembling the bricks I afterward 
borne out by Rawlinson (JEGS. 10. saw at the Fire-Temple near Isfahan 
51, 66), who cites a statement of Mus- and also in the walls of Rei; cf. pp. 
taufi to show that the Mongol king 253, 255, 435, below. 
Abaka Khan (d. 1282), son of Hulagu 3 Rawlinson, JEGS. 10. 51, de- 
Khan, restored this edifice. Compare scribes the walls as fifteen feet thick, 
also Wilson, Persian Life, p. 162. and he adds some details regarding 
The site of the structures, however, the condition of the dome and the cen- 
was probably occupied originally by tral chamber when he saw it ; his 
much older buildings, perhaps the an- description should be consulted, 
cient palace of the Keianian, Arsa- 4 So also Ker Porter was informed ; 
cian, and Sasanian kings. see his Travels, 2. 560. 



128 TAKHT-I SULEIMAN, AN ANCIENT FORTIFIED TOWN 



Suleiman are a group to the extreme north, which the natives 
call the 4 bazaar,' and two others near the rampart on the south- 
ern and southwestern side. One of these latter the guide 
called the 'market'; the other, somewhat to the southwest, 
which still has a towerlike shaft of its wall standing, he termed 
the 'kitchen.' But neither of these structures gave me the 
impression of being very old, although it was difficult to 
examine them because of the depth of the snow. 1 




RAWLINSON'S PLAN OP TAKHT-I SULEIMAN 

One of the most interesting features of the enclosure of 
Takht-i Suleiman is natural and not the work of human hands. 
It is a deep pond, some three hundred yards in circumference, 
situated south of the central part of the Takht and plainly 
owing its existence to some action of a volcanic character. The 
depth of this reservoir is evidently great, for its rich emerald 
hue shades into an azure that resembles the color of the deep 
sea. 2 The water is exquisitely clear, but somewhat saline or 



1 For some description of them see 
Ker Porter, Travels, 2. 660, and Raw- 
linson, JUGS. 10. 51. 



2 Rawlinson, JEGS. 10. 50, gives 
the depth as twenty-six fathoms. 



THE POND AND THE GATEWAY 



129 



calcareous to the taste, and it possesses a peculiar petrifying 
quality, imparted to it by the strong impregnation of calcium, 
so that its banks are encircled by a stony rim formed by the 
incrustation from the water wherever it overflows. In the 
plain a serpentine dike, called Azh Dahak by the natives, after 
the name of the mythical dragon, has also been built up 
by the deposit left as the stream flows down in a sinuous 
course from the height. When I saw the lake, the chief outlet 
seemed to be over the road leading out of the southeastern 
gateway, and the water was pouring through this in large 
quantities. Despite the outflow, which is constant, the volume 
of water, like the Avestan stream of Ardvi Sura, appears never 
to diminish, being constantly replenished from some unknown 
source. 1 

The great gateway at this point of the battlements is in an 
excellent state of preservation, and I even saw, if my memory 
serve me, the masons' signs marked on the bastions to the right 
and left of it, such as I saw on the wall of the terrace built by 
Cyrus near Murghab. The height of the rampart is here no 
less than forty feet, and its structure gives a good idea of the 
masonry of the rest of the wall. The large oblong blocks 
of which it is composed measure more than two and a half feet 
by one and a half, and are laid in a horizontal line with narrow 
upright stones fitted carefully between them, somewhat in this 
fashion. 2 I did not see elsewhere in Persia, either at Kangavar, 



Murghab, or Persepolis, stones so peculiarly set, although in 
the latter cases the blocks were more massive than those of 
Takht-i Suleiman. 



1 1 have since found that Rawlinson 
(JRGS. 10. 48) conjectures that the pond 
is connected by an underground syphon 
with some obscure mountain source. 



2 My notes may be supplemented 
by Rawlinson {op. cit. p. 47), who 
also refers to the extreme nicety with 
which the stones are fitted. 



130 TAKHT-I SULEIMAN, AN ANCIENT FORTIFIED TOWN 

Wishing to have a full view of the Takht from the eastern 
side as well, I walked along the citadel height, through the deep 
snow, to a point about the middle of the rampart, facing the 
rising sun. The descent from the battlement was here ex- 
tremely steep, but with the aid of my guides I found a place 
by one of the bastions where the snow was drifted almost to 
the level of the wall, and, as the crust was hard, I slid down and 
tramped off to a considerable distance in the direction of the 
eastern hills, so as to view the Takht in its full extent, includ- 
ing the northern end. The light at the time was unfavorable 
for photographing, so that two of my pictures failed ; but the 
third, which takes in a part of the middle section, especially 
the domed roof of the famous fire-temple, came out fairly well, 
and, when supplemented by my rough sketch of the side eleva- 
tion, will convey some idea at least of Takht-i Suleiman as I 
saw it. The noble contour of the walls was clear to me, but 
the impression which I received from this side of the Takht, as 
from the others, was not favorable to the theory of identifying 
it with Herodotus' description of the walls of Ecbatana. 

Up to this point I have given my own observations and 
memoranda without alluding to any historic accounts of 
Takht-i Suleiman and without any detailed references in 
the text to Sir Henry Rawlinson's claim that Takht-i Suleiman 
represents the Atropatenian Ecbatana and the site of the 
ancient capital of northern Media. 1 Although I knew in a 
general way of his theory, I purposely refrained from reading 
his monograph until I had studied the locality myself and had 
also visited Hamadan. In fact, I wrote out my notes for this 
chapter when on the steamer returning to America, and did 
not consult the famous essay until I reached home. Nor did I 
see Canon George Rawlinson's arguments, summing up his 
brother's treatise, until I had visited Hamadan and had 
already formed my own judgment in favor of the traditional 

1 Rawlinson, Notes on a Journey to the Atropatenian Ecbatana, in JUGS. 
Takhti-Soleiman and on the Site of 10. 1-158, London, 1841. 



IDENTIFICATION OF THE RUINS 131 

view that Hamadan, and not Takht-i Suleiman, really occupies 
the site of ancient Ecbatana. 1 Although I do not accept Sir 
Henry's view that the Takht is the site of Ecbatana (and I am 
not alone in my dissent), 2 I fully accept his unquestionable 
identification of Takht-i Suleiman with the Shiz of the Arab 
writers, and probably also with the Gazna and Ganzah of the 
Persians, the Gazaka or Canzaca of the classical writers and, I 
may now add, the Ganjak of the Pahlavi texts. 3 

In dealing with the subject, I shall first present the grounds 
for Rawlinson's identification of the ruins with the ancient Shiz 
and the fire-temple of Adhargushnasp. Afterward I shall 
touch on their probable identity with Gazna, Ganzaca, Gan- 
jak, etc., as Rawlinson believed. Some of the citations which 
he uses to prove his point will be found also in my book on 
Zoroaster, where I had independently given them for another 
purpose; but I reproduce the more important of them here 
because of their value in this connection and because they give 
an excellent idea of Takht-i Suleiman. 4 

The first statement which I cite, and which was used also by 
Rawlinson, is taken from an account of the stronghold written 
by Yakut (about A.D. 1220). Yakut speaks of Shiz as located 
among mountains that abound in ore and mineral wealth, 
and describes the lake, whose petrifying qualities I have 
mentioned, and speaks of the buildings, especially the an- 
cient fire-temple, within the circuit of the city's walls. Yakut's 
account reads as follows : 

1 See George Rawlinson, Five Great the city, and the same is true of Gazn, 
Monarchies, 3. 24-28, London, 1865. Jazn, Ganjah, Ganzah (Lat. Canzaca 

2 George Rawlinson mentions the or Ganzaca}, which is generally be- 
fact that his brother's view was com- lieved to be another designation for 
batted by Quatremere in the Memoires the same place. 

deV Academic des Inscriptions et Belles 4 See my Zoroaster, pp. 195, 197, 

Lettres, 19. 419 seq. 201, 202, 204 ; and compare Gottheil, 

3 The name Shiz, as stated above References to Zoroaster, in Classi- 
(p. 73), is derived through an Arabic cal Studies in Honour of Henry Dris- 
corruption of Clz from Caefasta, the ler, pp. 40-45. Most important is 
Avestan name of Lake Urumiah. This the monograph by Rawlinson, JRGS. 
title designates the district as well as 10. 65-158. 



132 TAKHT-I SULEIMAN, AN ANCIENT FORTIFIED TOWN 

* Shiz, a district of Azarbaijan. Its name is a form of Chis, 1 
out of which the Arabs have made Shiz. It is said that Zaradusht, 
the prophet of the Fire- Worshippers, came from there. The chief 
place of the district is Urumiah. . . . Here is what Mis'ar ibn 
Muhalhal (A.H. 330 = A.D. 940) says about Shiz : "... This town is 
situated between Maraghah, Zanjan, Shahrzur, and Dinavar, in the 
midst of mountains containing mines of gold, quicksilver, lead, silver, 
orpiment, and amethysts. ... A wall encloses the city, and within 
its circuit is a pool whose bottom cannot be sounded. I dropped a 
line in it more than fourteen thousand cubits, but the lead did not 
find any resting-place and remain steady. 2 The area of the lake is 
about one quarter of an acre. Earth soaked with water from it im- 
mediately becomes hard stone. Seven streams of water flow from the 
lake, each of which turns a mill before flowing out under the wall. 
At Shiz there is also a large fire-temple, which is held in great 
veneration. From it are lighted the fires of the Magians from the 
east to the west. On the top of the dome there is a silver crescent 
which is a talisman. Many rulers 3 have tried to remove it, but have 
not succeeded. One of the extraordinary things connected with the 
temple is, that a fire has been kept burning in it for seven hundred 
years without any ashes having been found ; 4 nor has the fire gone 
out for a single hour. Hurmuz ibn Khusrushir ibn Bahrain built the 
city of plaster (Ms) and stone. Near ('ind) the fire-temple stand 
lofty palaces and large and magnificent buildings. Whenever an 
enemy advances to take the city and plants his ballista against its 
walls, the stone from the machines falls into the pool which we have 
mentioned ; and if he move the ballista back [from the wall], even 
as far as one cubit, the stone falls outside the wall. . . . " 5 So 
far Abu Dulaf Mis'ar ibn Muhalhal, the poet ; but I cannot vouch 

1 Variant readings are Jls, <7Ts, //Is, 3 Lit. Amirs ; Kazvini has * con- 
and .Tins. In reality Shiz is a corrup- querors.' 

tion of Ct2, Avestan Caetasta ; see p. 4 This is evidently the ' imposture 

73, above, and my Zoroaster, p. 197. of the coals' (irXdvy &v0pdicwv) alluded 

2 This extravagant number in the to by the Church Father Georgius 
Arabic text is reduced in the Persian Cedrenus ; see p. 141 below, 
version to 4000, which in itself is ex- 6 I omit here a legend which recounts 
aggerated enough. See Rawlinson, how the city was built by Hurmuz on 
JUGS. 10. 68, n. 4 ; Barbier de Mey- the spot where a fire had appeared at 
nard, Diet. geog. p. 368, n. 1. The the grave of a messenger whom he had 
true depth is not more than 156 feet, sent to worship Christ and who had 
as given by Rawlinson, op. cit. p. 50. died on his return. 





THE RUINED FIRE-TEMPLE AT TAKHT-I SULEIMAN, BURIED IN THE SNOW 




TAKHT-I SULEIMAN FROM THE EAST 
(After a sketch by the author) 



YAKUT'S DESCRIPTION 133 

for the truth of his statement, for things are told on his authority 
which are inaccurate and untrue. I have incorporated it here as I 
found it ; but Allah knows best. Some one else has related that in 
Shiz there is the fire of Adharakhsh, a temple honored of the Magians. 1 
It was customary for their kings, when they ascended the throne, to 
make a pilgrimage thither on foot. The people of Maraghah and of 
this neighborhood call this place Gazna; but Allah knows best.' 2 

Yakut has furthermore a separate short article on G-azn 
(Kazri) and its fire-temple, which, if taken in connection with his 
previous assertion that G-azn, Gaznd (^GaznaJi) is the Persian 
form and Shiz the Arabic form, shows that he regards the two 
places as one, and their identity is generally accepted. 3 Thus, 
in his brief note on Gaznd, lie says : 

' Gazna is a small town about six farsakhs from Maraghah. In 
it there is a place of worship of the Magians and an old fire-temple, 
and also a great and very old palace built by Kei Khosru.' 4 

And again, in his statement under Jaznak, Yakut says : 

' Jaznak is a small but very flourishing town in the vicinity of 
Maraghah, and one may see there the ruins of edifices erected by 
the ancient kings of Persia, and also a temple of fire/ 5 

The Persian geographer Kazvini (about A.D. 1263) repeats 
almost word for word the same statement regarding the lake, 
the walls, and the fire-temple at Shiz, and adds that Zoroaster 
came from this city of Azarbaijan. 

1 Spelled Adhrkhsh, without vocali- Studies in Honour of Henry Drisler, 
zation, but vocalized as Adhar-khush, p. 42, cited in my Zoroaster, p. 200. 
('good fire') in the edition of Masudi, For aid in translating the passage 
Kitab at-Tanbih, ed. De Goeje, in Bib- from the Arabic, I am indebted to iny 
liotheca Geographorum Arabicorum, friend and former pupil, Dr. William 
8. 95. For other readings of the Popper. 

name (all of which are evident corrup- 8 See especially Noldeke, Geschichte 

tions of Adhar-gus[nasp]), see Raw- der Perser und Araber, p. 100, n. 1, 

linson, JUGS. 10. 104. accepting Rawlinson's view, which is 

2 See Yakut, ed. Wustenfeld, 3. 354, maintained also by Marquart, Erdn- 
and the translation by Barbier de ahr, p. 108, and others. 

Meynard, Diet. geog. p. 367 seq. ; * Yakut, p. 488 ; cf. also Gottheil, 

also the version by Gottheil, Refer- References to Zoroaster, p. 44. 
ences to Zoroaster, in Classical 6 Yakut, p. 161. 



134 TAKHT-I SULEIMAN, AN ANCIENT FORTIFIED TOWN 

1 Zaradusht, the prophet of the Magians, takes his origin from 
there (i.e. Azarbaijan). It is said that he caine from Shiz. He 
went to the mountain Sabalan, living apart from men. He brought 
forth a book the name of which was Basta (i.e. Avesta).' 1 

Still earlier Arab and Persian writers mention Shiz and its 
renowned fire-temple, as I have shown in my Zoroaster (p. 198), 
to which work I may refer, and I therefore give only the main 
points here. Thus Ibn Khordadhbah, who wrote about A.D. 
816, and whose father is said to have been a Magian, enumer- 
ates the important cities of Azarbaijan and includes : 

' Janzah (i.e. Ganjah), the city of Abarwiz (i.e. Parviz), Jabra- 
wan, Nariz, Urmiah, the city of Zaradusht, Salinas, and Shiz, 
in which there is a fire-temple Adharjushnas[f], which was vener- 
ated by the Magians ; and it was customary when one of their mon- 
archs ascended the throne, for him to make a pilgrimage [thither] 
on foot from Madain (i.e. the Sasanian capital Ctesiphon).' 2 

Al-Hamadhani (about A.D. 910) repeats the same statement 
elsewhere : 

1 J a n j a h, Jabrawan, and Urumiah, the city of Zaradusht, and 
Shiz, in which there is the fire-temple of Adharjush- 
n a s [f ], which is held in high veneration by the Magiaus.' 3 

In still another passage the same author says that the fire 
of Adhargushnasp belonging to Kei Khosru (and originally 
located elsewhere in Azarbaijan) was removed to Shiz. 

' Adharjushnasf, the fire of Kei Khosru, was in Adharbaijan; but 
Anushirvan removed it to Shiz.' 4 

Masudi (died A.D. 951), in his account of the various fire- 
temples, in Meadows of Gold, chapter 68, says : 

1 See Gottheil, References to Zoro- between Ganjah and Shiz, as will be 
aster, pp. 40, 42, n. 2 ; Rawlinson, noted below, p. 136. 

JRGS. 10. 68 ; and my Zoroaster, 8 Al-Hamadhani, Kitdb al-Buldan, 

p. 201. ed. De Goeje, Bibl. Geog. Arab. 5. 

2 See Ibn Khordadhbah, ed. De 286 ; see also my Zoroaster, p. 198. 
Goeje, Bibl. Geog. Arab. 6. 119 ; and * Al-Hamadhani, p. 246 ; see also 
transl. p. 91. It is to be observed Gottheil, References to Zoroaster, 
that Ibn Khordadhbah distinguished p. 45. 



A FAMOUS FIRE-TEMPLE 135 

'A fourth fire-temple is found in the country of Shiz and Arran; 
it was originally consecrated to those idols which Anushirvan de- 
stroyed. Others say that Anushirvan, having found in this temple 
an altar on which the sacred fire was burning, transported it to a 
place called al-Birkah ("the basin," near Shiraz). The [ancient 
Keianian] king Kei Khosru built a temple which was known under 
the name of Kusujah' * 

The still earlier writer, Asmai (died A.D. 831), a scholar 
at the court of Harun al-Rashid, mentions Shiz and its fire- 
temple in his account of the campaign of the Sasanian king 
Khosru Parviz against the Byzantine emperor Heraclius 
(623-627). 

1 The king [Khosru] went on till he arrived at the city of Shiz, 
where there is a very great fire-temple which remains to this day. 
Khosru remained constantly at prayer in this temple, while he 
ordered his army to form an entrenched camp ; and he abode for a 
month at Shiz to refresh himself and his troops, and employed him- 
self in collecting provisions and establishing bazaars.' 2 

Tabari, the historian (840-923), speaks of the rich gifts 
made to the fire-temple (of " Adhargushnasp) at Shiz in Azar- 
baijan by the Sasanian king Bahrain Gor, after his victory over 
the Turkmans, about A.D. 420. 

' Bahram's route, on returning from that campaign, lay through 
Azarbaijan. Accordingly he sent to the fire-temple in Shiz the 
rubies and other jewels which were in the crown of the [vanquished] 
Khakan, and also his own sword, inlaid with pearls and jewels, as 
well as many other ornaments. The Khatun, or wife of the Khakan, 
he made an attendant in the temple.' 3 

The native Persian lexicon, Farhang-i Anjuman-i Ardi Nasarl^ 
a modern but valuable compilation, cites an older work which 

1 See Masudi, Les Prairies d'Or, 10. 76, from whom the passage is quoted, 
tr. Barbier de Meynard, 4. 74, Paris, 3 Tabari, translated in Noldeke, 
1865. The name Kusujah is evidently Gesch. der Perser und Araber, pp. 
a scribal error for Ganzah. Shiz is 100, 102, 104 ; also Rawlinson, JRGS. 
mentioned by Masudi also at 2. 235. 10. 77, and compare Justi, Grundr. 

2 Asma'i, cited by Rawlinson, JUGS. iran. Philol. 2. 527. 



136 TAKHT-I SULEIMAN, AN ANCIENT FORTIFIED TOWN 

likewise speaks of the fire of Adhargushnasp at Shiz and the 
association of Zoroaster's name with the place. Thus : 

' The author of the Haft Ikllm says that Shiz is the name of a 
city between Maraghah and Zangan, and there was a great fire-temple 
there, called Adharjus[nasf], which was held in great esteem by 
the Persian kings, who used to come to the mountain on foot. The 
origin (asl) of Zaradusht was from there, and he went to the moun- 
tain Sabalan and composed there the book Abasta and brought it 
to King Gushtasp.' l 

Taking all these quotations together, especially Yakut's 
statement about the mountains that contain ore, the lake whose 
waters produce petrifaction, and the famous temple of fire, it is 
clear that Takht-i Suleiman, or ' Solomon's Throne,' is actually 
the site of the ancient city of Shiz, and that when standing on 
its terrace we have the ruins of the historic pyrseum of Adhar- 
gushnasp before us. We may therefore accept Rawlinson's 
identification of the site of the so-called ' Throne of Solomon ' 
(Takht-i Suleiman) as complete and convincing, just as we now 
know that the ruins in Southern Persia which the natives call 
'Throne of Jamshid ' (Takht-i Jamshld) are really the historic 
Persepolis of the Achsemenians. 

I shall now turn to the second part of the discussion, the 
arguments of the same scholar to show that the city referred to 
in the classics as Gazaka or Ganzaca, and in Armenian as 
Gandzag, is likewise identical with Shiz and consequently with 
the present Takht-i Suleiman. We have already seen that 
Yakut regarded G-azn and Shiz as identical places, even if the 
earlier Ibn Khordadhbah differentiated them. 2 To prove the 
correctness of the identification, Rawlinson makes use of data 
connected with the campaign of Khosru Parviz against Bahram 

1 This new allusion, which I add 2 See quotation from Ibn Khordadh- 

to Rawlinson's material, has recently bah given on p. 134, above, and com- 

been made accessible by Von Stack- pare also the remarks by Noldeke, 

elberg, Persische Sagengeschichte, in Geschichte der Perser und Araber, 

WZKM. 12. 233. p. 100, n. 1. 



ANCIENT GAZN AND SHIZ 137 

Chobin (A.D. 589), and the wars of the Byzantine emperor 
Heraclius against Khosru (A.D. 624), and draws deductions 
also from the movements of Mark Antony against the Parthians 
six centuries earlier. I shall not attempt to reproduce Raw- 
linson's arguments in detail, but shall simply state that most 
scholars accept his identification, together with the view that 
the ruins represent the site of what was also called Phraaspa, 
and the fortress of Vera in Media. 1 It may be taken as prac- 
tically certain that if Ganzaca and Shiz in antiquity were riot 
absolutely one and the same city, the former cannot have been 
located far distant from the latter, and we may perhaps distin- 
guish between the city in the plain and the fortress on the 
height, following the statement quoted by Rawlinson from 
Strabo, who says, ' The summer residence of the kings is at 
Gazaca, situated in a plain and in a strong fort named Vera, 
which was besieged by Mark Antony in his expedition against 
the Parthians.' 2 As a matter of conjecture, in that event, I 
might refer to the modern place named Ganjabad, a short dis- 
tance to the northeast of Takht-i Suleiman (about lat. 37 15', 
long. 47 45') see the map at the end of this volume or to 
a place marked Ganja between Sanjud and Tikantapah on the 
map in Curzon's Persia? 

By way of supplement, however, to Rawlinson's monograph, 
I present one or two new matters of evidence on the subject of 
Ganzak, Shiz, and the fire-temple Adhargushnasp. The testi- 
mony is in the form of Zoroastrian traditions, relating espe- 
cially to the earlier king Kei Khosru (Av. Kavi Haosravali), 
who is supposed to have lived about B.C. 800, and his enemy, 
Afrasiab the Turanian (Av. Tuirya Frahrasyari). 

1 Rawlinson, JUGS. 10. 71-111, read ' and [the winter palace] is in 
113-115 ; Noldeke, Geschichte der a strong fort,' contrasting the resi- 
Perser und Araber, p. 100, n. 1 ; dences according to the seasons, but 
Marquart, Eransahr, p. 108; Justi, this is less good. 

Grundr. iran. Philol. 2. 527, 542, 544 ; 3 Streck, Zt. f. Assyr. 15. 332, com- 

Darmesteter, Le ZA. 1. 155, n. 12. pares Ganzaka or Gazaka with G-izin- 

2 Strabo, Geog. 11. 13. 3 ; cf. Raw- (fikissi and notices the forms of the 
linson, JUG 8. 10. 113. Some critics name Jiz, Shiz. 



138 TAKHT-I SULEIMAN, AN ANCIENT FORTIFIED TOWN 

A Pahlavi treatise written early in the ninth century A.D. 
and entitled Shatrolhd-l Alrdn, from the account which it gives 
of the cities of Iran, says that the town of 4 Ganjak ' or 4 Gan- 
zak ' in Azarbaijan was founded by Afrasiab of Turan. 

Trasiak of Tur (Afrasiab) founded the town of Ganjak in the 
region of Ataropatakan (Azarbaijan).' l 

There is further evidence to associate Afrasiab, for a time at 
least, with this region, since Firdausi narrates that after his 
defeat in Turan by Kei Khosru, Afrasiab fled to Azarbaijan and 
took refuge in a cave on a high mountain near Bardah, a place 
located east or northeast of Lake Urumiah, somewhere between 
Ardabil and Maraghah, and frequently mentioned by the 
mediaeval Persian and Arabic writers, including Yakut, who 
says it is 'nine farsakhs from Ganjah.' 2 The fugitive, accord- 
ing to the legend, was discovered by a hermit, whom Firdausi 
calls Horn (and the Avesta calls Haoma), and then sought to 
escape into the waters of Lake Urumiah (miswritten in the 
manuscripts of the Shah Namah as Khanjast for Chechast, Av. 
Chaechasta) . 3 Afrasiab's place of concealment, however, was 
revealed, and he was captured and slain by Kei Khosru, who 
thereupon proceeded to the fire-temple Adhargushnasp, to 
give thanks for his success. 4 

Somewhat earlier in his narrative Firdausi had given an 

1 See Jamaspji Minocheherji, Pah- above, and cf. Ker Porter, Travels, 2. 
lavi Texts 1, Shatrolha, 58, Bombay, 495-496 ; Rawlinson, JRGS. 10. 45 ; 
1897, and the translation of Modi, Wilson, Persian Life, pp. 73-74. 
Shatrolhd-l Alrdn, p. 117, Bombay, 8 See the earlier form of the legend in 
1899; also Blochet, Villes de Vlran, the Avesta, Ys. 11. 7, and for the strug- 
in Eecueil de Travaux relatifs a gle between Franrasyan (Afrasiab) 
la Philologie, 17. 176, Paris, 1895. It and Haosrava (Kei Khosru), seconded 
is possible also in MKh. 27. 44 to by Haoma, compare Yt. 9. 17-23 ; 
read instead of Dujako, in the Pahlavi 17. 37-43. See also next page and 
characters, Ganjako (i.e. Canzaca), note that there is a 'cave' at the 
but it would not give so good an inter- ruined city of Shahr-i Afrasiab near 
pretation of that particular passage; Samarkand. 

see West, SEE. 24. 62, n. 2. * See Shah Namah, ed. Vullers- 

2 Yakut, p. 92. For the caves in Landauer, 3. 1386-1398 ; tr. Mohl, 4. 
the vicinity of Maraghah, see p. 104, 155-169. 



AFRASIAB AND KEI KHOSRU 139 

account of the founding, by Kei Khosru, of the fire-temple of 
Adhargushnasp, which he locates in the ' Castle of Bahman ' 
(Dizh-i Bahman). The context would imply that this was on 
the 4 frontier ' of Ardabil, l where it is located also by Yakut, 2 
but Rawlinson urges that the description answers to Takht-i 
Suleiman. 3 The lines about the founding of the pyrseum 
after the castle was stormed run as follows : 

' Within the vast rampart the king found a city with gardens, 
public squares, mansions, and a palace. At the spot where the light 
had flashed out and the darkness had been utterly dispelled, Khosru 
gave orders that a domed temple should be erected to a height 
towering to the sky. It was ten noose-lengths square and was 
surrounded by high vaulted chambers, covering a circuit of half a 
horse-course. There he brought and installed the fire Adhargush- 
[n]asp, around which sat Mobeds, astrologers, and wise men.' 4 

In keeping with Firdausi's account of Afrasiab's tempo- 
rary escape from Horn and Kei Khosru by concealing himself 
in Lake Urumiah, and in harmony also with the Avesta, which 
locates the struggle of Afrasiab and Kavi Haosravah ' behind (or 
in the sight of) Lake Chaechasta (Urumiah),' we have still 
other Zoroastrian traditions preserved in the Pahlavi writings. 
These books associate the name of Kei Khosru with the fire of 
the warriors, Adhargushnasp, and locate this pyrseum in the 
neighborhood of Lake Urumiah or on Mount Asnavand. 5 Thus 

1 Pers. to, dar-i Ardabil ba-marzl 4 Shah Namah, ed. Vullers-Lan- 
kih anja Dizh-i Bahman ast, see Shah dauer, 2. 761 ; tr. Mohl, Livre des 
Namah, ed. Vullers-Landauer, 2. 756, Hois, 2. 441 ; and Pizzi, II Libro del 
and compare tr. Mohl, Livre des Hois, JRe, 3. 78. 

2. 435, ' a Ardebil . . . sur la frontiere ' ; 6 See my Zoroaster, pp. 100, 48 ; 
compare also Pizzi, II Libro dei Be, and consult also p. 70, n. 1, above. 

3. 72, 'al confin d' Ardebil.' On the Avestan phrase 'behind (or 

2 Yakut, p. 125. So also Sadik Is- in the sight of) Lake Chaechasta,' 
fahani, p. 14. see Bartholomae, Air. Wb. s.v. pasne 

3 Rawlinson (JRGS. 10. 82, n. 3) (for pasne). Darmesteter, Le ZA. 2. 
was mistaken in saying that Ardabil is 631, n. 92, locates the scene of the 
not mentioned ; it occurs in the text conflict to the north or northwest of 
(Vullers-Landauer, 2. 756) a few pages Lake Urumiah, calling attention to 
before the fire-temple is described. both Lake Sevan and Lake Van as 



140 TAKHT-I SULEIMAN, AN ANCIENT FORTIFIED TOWN 

the Pahlavi Bahman Yasht (about the seventh century 
A.D.) states: 

' The fire Ataro-gushnasp, near (or by) the deep Lake Chechast, 
which has warm water that drives away the demons.' 1 

Similar is the statement found in the Selections of Zat- 
sparam (about A.D. 881): 

' Two fountains of the sea were opened out for the earth : Che- 
chast, a lake which has no cold wind and near (or on) whose 
shore rests the triumphant fire Gushnasp, and the other, Sovar 
(near Tus).' 2 

The Bundahishn, one of the oldest Pahlavi works extant, 
describes how the fire Ataro-gushnasp aided Kei Khosru when he 
was engaged in putting down idol-worship about Lake Chechast, 
and its shrine was in the same locality near Mount Asnavand. 

1 When Kai-Khusrob was engaged in extirpating the idol-temples 
of Lake Chechast, it (i.e. the fire Ataro-gushnasp) settled upon the 
mane of his horse and drove away the darkness and gloom, and made 

possibly representing Av. Vairi Hao- four parasangs from Lake Urumiah in- 

sravah (Sir. 2. 9 ; Yt. 19. 56 ; Ny. stead of fifty, which is the number 

5. 5) in contrast to Vairi Caetasta given in the other text. 
(Lake Urumiah), noticing also the 1 BYt. 3. 10. The Pahlavi word 

village of Khosrova near Dilman, to which I have rendered * near (or by) ' 

which I have referred above, p. 70, is pavan (.pa), which West renders 

n. 1. See also West's note on Bd. 'on,' SEE. 5. 218. For the Pah- 

22. 8, in SEE. 5. 86, n. 7. Spiegel, lavi text see Noshervan, Pahlavi 

jZrdnische Alterthumskunde, 1. 653- Zand-i Vohuman Yasht, p. 14, 1. 12 

654, suggests Lake Sevanfor VairiHao- (= p. 17, 1. 1, translit.), Bombay, 1900. 

sravah, but notices also that Masudi On the reading garm, 'warm'- (for 

(2. 131) assigns some of the scenes of 'medicinal,' West, SEE. 5. 218), 

the Honi-Khosru-Af rasiab conflict to consult Rosenberg, Livre de Zoroastre, 

the town of Serav, between Ardabil p. 74. 

and Tabriz (see Yakut, p. 306). It 2 Zsp. 6. 22, transl. West, SBE. 

would be tempting to regard Vairi 5. 173. Again the Pahlavi for 'near 

Haosravah, 'Lake of Khosru,' as our (or on) whose shore' has munas 

present lake at Takht-i Suleiman pavan (pa) bar, see text, ed. West, 

adjoining Khosru's fire-temple, and in Avesta, Pahlavi, etc., Studies hi 

Mount Asnavand (Sir. 2. 9, etc.) as Honour of D. P. B. Sanjana, first 

Mount Zindan; the Iranian Bunda- series, p. Ixxi, Strassburg, 1904. 
hishn locates the Lake of Khosru at 



A ROMAN EMPEROR DESTROYS THE SHRINE 141 

it quite light, so that they might extirpate the idol-temples ; in the 
same locality the fire Gushnasp was established at the " appointed 
place " (i.e. shrine) on the Asnavand mountain.' 1 

The location of Takht-i Suleiman is some ninety miles 
distant from Lake Urumiah (Chechast), it is true, but near 
enough, perhaps, to answer to the general description given 
in the Pahlavi works, when compared with the other evidence 
in the case, even if we be not tempted to the extent of conjec- 
turing that the lake adjoining Khosru's ancient fire, Adhar- 
gushnasp, in the Takht, is also the Lake of Khosru, Vairi 
Haosravah, referred to in the Avesta and Pahlavi texts, or 
go so far as to surmise that the extinct volcano of Mount 
Zindan, so near at hand, might possibly be Mount Asnavant 
(Asnavand) , which is invoked together with Chaechasta in the 
Zoroastrian prayers. 2 

I shall cite a single other passage relating to Takht-i Sulei- 
man, or Ganzaca, and its palace and fire-temple ; but it is one 
of special interest because of its connection both with 
Zoroastrianism and ChristianhYy. 3 Georgius Cedrenus, the 
Church Father, in narrating the progress of the war of 
the Byzantine emperor Heraclius against the later Sasa- 
nian king Khosru Parviz, or Chosroes, describes how Khosru 
had his own image enthroned amid emblems of the celestial 
bodies to which he did homage a sight which so outraged 
the Roman emperor and his Christian soldiers that he burned 
the fire-temple and reduced the entire city to ashes. 



7, translated by West, 8 BE. pies for idol- worship around Lake 

5. 64. Here the Pahlavi text reads Chechast; see West, 8 BE. 24. 15. 

pavan (pa} Asnavand kof . . . bard 2 See Sir. 2. 9 ; Ny. 5. 5. The 

(be) val (6) ddd-gah, cf. Justi, Bunde- sanctity, not the size of the lake, would 

hesh, p. 41, Leipzig, 1868, lithogr. ; entitle it to consideration. But all 

Westergaard, Bundehesh, p. 41, Co- this would be merely an attractive 

penhagen, 1851 ; Unwalla, The Pah- hypothesis upon which a theory might 

lavi Bundehesh, p. 48, Bombay, 1897. be built. 

The Pahlavi treatise Mainog-i Khirad, * Cited by Rawlinson, JUGS. 10. 

2. 96, also alludes to the achievement 52, 78. 
of Kei Khosru in exterminating tern- 



142 TAKHT-I SULEIMAN, AN ANCIENT FORTIFIED TOWN 

1 The Emperor Heraclius took possession of the city of Gazaca, 1 
in which was the temple and the treasures of Croesus, king of Lydia, 
and the imposture of the burning coals. 2 On entering the city he 
found the abominable image of Khosru, an effigy of the king seated 
under the vaulted roof of the palace 3 as though in the heavens, and 
around it the sun, moon, and stars, to which he did homage with 
superstitious awe, as if to gods, and he had represented angels bear- 
ing sceptres and ministering unto him. 4 And the impious man had 
arranged by cunning devices to have drops falling from above, like 
rain, and sounds resembling roaring thunder to peal forth. All 
these things Heraclius consumed with fire, and burned both the 
Temple of Fire and the entire city.' 5 

Viewed in the light of all this testimony from the historic 
past, the desolate scenes within the walls of Takht-i Suleiman 
take on new life. The so-called 4 bath ' resumes once more its 
ancient glory as the great fire-temple of Adhargushnasp, whose 
dome, once crowned with a silver crescent, has crumbled into 
decay. The deep-sunk vaulted arch beneath appears as the 
shrine where, eight hundred years before Christ, the Keianian 
king Khosru, robed in white, 6 prayed for victory over Afrasiab, 
and, on his triumphant return from battle, offered thank- 
offerings worthy of a king. The same example of thanks- 
giving was followed by the Sasanian king Bahram Gor, nearly 
five centuries after Christ, when he deposited in this shrine 
the treasures won in his victory over the Turanians and made 
the vanquished queen a priestess in the shrine. Here, likewise, 
two centuries later, the Christian emperor Heraclius destroyed 
the idolatrous images of Khosru Parviz. Some day the archae- 
ologist's spade may bring to light evidences even more tangible, 
if greater proof be needed, of the identity of this famous ancient 
shrine. As I retraced my steps from the ruins of Takht-i 

1 Gk. TTJV Ta.$a.Kbv ir6\iv. Georgius Cedrenus (c. A.D. 1100), 

2 See p. 132, n. 4, above. 1. 721-722, ed. Migne, Patrolog. Graec. 
8 Gk. TraXaTlov. 121. 789-790, Paris, 1864. 

4 For the sceptred angels, compare 6 Firdausi, Shah Ndmah, tr. Mohl, 

the carving over Khosru's arch at Tak-i 4. 155 ; compare also Yt. 9. 21 ; 17. 41. 
Bostan ; see p. 221, below. 



LEAVING THE HISTORIC RUINS 143 

Suleiman, I felt certain that I had visited the historic site of 
Shiz, designated also by other names, but I reserved my judg- 
ment regarding the claim that the walled enclosure might 
represent the Atropatenian Ecbatana, until I should have 
visited Hamadan. 



CHAPTER XII 
RAMADAN, THE ANCIENT ECBATANA 

' Orchards stretch their bloomy span 
Bound the walls of Hamadan ; 
Purples deepen on the grape, 
Lyric brooks make blithe escape, 
Yet are all the glories gone 
That the Lord of Macedon 
Saw, ere drew the revel on 
And the Bacchic orgy ran 
Round the walls of Hamadan.' 
CLINTON SCOLLARD, Hound the Walls of Hamadan. 

THE four days' riding from Takht-i Suleiman were fatiguing 
ones, with forced marches of twelve hours a day in the saddle. 
The first night was spent at the small hamlet of Nasarabad in a 
mud hovel whose roof was adorned by the skull of a horse, which 
served apparently as a talisman to induce sleep, for I rested well. 
Toward evening on the second day we halted at Bab-i Ro- 
shani, 4 Gateway of Light,' which looked its name as we entered 
the village at sunset, flooded with golden light from the clos- 
ing portals of day. The third halt was at Kultapah, where I 
passed an uncomfortable night, but was ready at daybreak for 
an early start toward Hamadan. 

Of the many hard stages that made up the journey of one 
hundred and fifty miles from Urumiah, over high, snow-covered 
mountains, this last was one of the most difficult, for, although 
April had advanced a week, the drifts in the passes were particu- 
larly deep, and the streams in the plains at the foot of the hills 
swollen and extremely hard to ford. My men and horses had stood 
the strain for twelve days well, but I had trouble with a guide, 

144 



OF THF 

UNIVERSITY 

OF 

-: ; .-, , liv ., 




DISTANT VIEW OF HAMADAN AND MOUNT ALVAND 




THE MUSALLAH, HAMADAN 



GLIMPSES FROM AFAR 145 

Lutf ullah, whom I hired at Bijar a couple of days previously, to 
serve also as a guard, because the Muharram festival was at its 
height and a Christian's life is not altogether safe in some of 
the fanatical villages at this time, when the Shiite Mohamme- 
dans work themselves up to a religious frenzy in their lamenta- 
tions at the martyrdom of Husein, lacerating themselves with 
swords, knives, and sharp stones and throwing themselves into 
a state of pious mania. As a guide Lutfullah proved very 
unsatisfactory and on the last day was so cruel to his horse 
and so disobedient to me that I was obliged to resort to force 
and eject him from the courtyard of the caravansarai. He took 
up a position outside the gateway, crouched half sulkily, half 
threateningly in a corner with his rifle, and awaited my coming 
out. I knew it was no time for hesitation, so I leaped on my 
horse and rode through the gate. For a moment he wavered, 
casting a sullen, menacing glance, then suddenly his face 
changed, he dismissed his anger, came quietly to my side, and 
from that time forward was absolutely obedient and devoted 
to me, and proved so useful a servant that I kept him a number 
of days longer than I really needed him. 

The plain through which the road approaches Hamadan from 
the north was beginning to put forth signs of spring ; the dark, 
moist earth showed a tinge of green; the gardens and orchard 
leas were taking on a trace of color ; and the purple-stalked 
willows such as I have seen only in the shading of the 
French impressionist school told that days of sunshine were 
at hand. A hazy light spread a soft glow over the landscape 
and lent an air of enchantment to all around, when far in the 
distance I caught my first glimpse of Hamadan, that city of by- 
gone glory. 

Although the city was in sight at noon, it took nearly five 
hours to reach it, and this gave me a good opportunity to study 
its situation and the surrounding country. Nature has given 
Hamadan a location that is in many ways remarkable. A level 
plain spreads like a garden before the feet of the city for a 



146 HAM AD AN, THE ANCIENT ECS AT AN A 

distance of fifteen miles in length and ten in breadth, and pre- 
sents a smooth expanse of territory which is well watered and 
responds readily to cultivation, yielding barley, wheat, fruit, 
vegetables, and the poppy plant. Mountains and hills form a 
palisade on either side to enclose the plain. On the southern 
and western side the mass of peaks called Alvand towers six 
thousand feet above it, or twelve thousand feet above sea-level, 
and stretches for miles away, guarding the approach in that 
direction and keeping watch by night over the sleeping town 
like some giant sentry. Night after night, during my stay at 
Hamadan, I used to watch for the moon to light up the figure 
of this silent sentinel muffled in ice and snow. 1 On either side 
back of the city are lofty hills, so that the town looked from 
the distance, as I rode toward it, as if it were set on the side 
of a height. Directly adjoining the city is the elevation called 
Musallah, the acropolis in ancient times, but the true propor- 
tions of this citadel height did not come out till I drew near 
and saw from its outline how much more probable it is that 
this, rather than Takht-i Suleiman, is the hill to which Herodo- 
tus alludes in his description of the walled city of Ecbatana, 
the capital of ancient Media. 2 

About five o'clock in the afternoon my train of tired animals 
slowly entered Hamadan, weary from almost a fortnight of 
uninterrupted marching from Urumiah. 

Not a sign of the ancient glory of Ecbatana, that was once 
the home of kings, struck my eye, nor was there a trace of that 
solemn grandeur which is noble in its decay at Persepolis and 
Pasargadse. I saw instead only crooked streets, alleys where 

1 The name Alvand or Elvend is as pilgrimages to the grave of one of the 

old as the Avesta, where it is found as followers of the Prophet, named Sahib 

Aurvant ; see Yt. 19. 3, asta aurvanto Zaman, who is buried on the top of 

fdnkavo, ' the eight spurs (?) of Aur- Alvand, and near the grave is a spring 

vant.' In the Pahlavi Bundahishn it known as Chashmah-i Malik, ' Foun- 

is Alvand (Bd. 19. 3, alvant). The tain of the King.' 
Greeks called it Mount Orontes 2 See the description of Herodotus, 

The Mohammedans make p. 162, below. 



MT. ALVAND IN THE DISTANCE 




OF THF 

UNIVERSITY 

OF 




THE MODERN CITY 147 

ran channels of dirty water, rows of shabby houses with flat 
mud roofs, and not a vestige of beauty anywhere. The wind- 
ing street which I followed led past a Mohammedan burying- 
ground located in the heart of the town. I shall never forget 
the stench from the shallow graves. Nothing but the severity 
of the winters and the healthy position of the town itself saves 
Hamadan from pest. Street-cleaning ordinances appear not to 
exist, and nature's scavengers, the birds of prey, seem sadly to 
neglect their duty, for the carcass of a dead dog was lying in 
one of the frequented lanes during the entire time of my two 
visits at Hamadan. 

The outline of the city is roughly that of a parallelogram, 
running from north to south. 1 Mount Alvand lies at a dis- 
tance of about three miles on the southern side; the Musallah, 
or acropolis hill, is adjacent to the eastern section of the town, 
of which it forms a part. 2 The town itself is divided into four 
quarters, or wards, for administrative purposes, each under a 
separate magistrate (kadkhudd), who is accountable to the gov- 
ernor, but whose office is practically hereditary. Through 
the middle of the city runs the insignificant river Alusjird, 
spanned by several single-arch stone bridges and some wooden 
ones. In the spring it is swollen by the melting snows from 
Alvand, but in summer it is nearly dry, although it suffices to 
turn the Persian water-mills on its banks. This is evidently 
the stream to which Ctesias refers when he says that Semira- 
mis, queen of Assyria, on visiting this place, where she built a 
palace, found the city poorly supplied with water, and caused 

1 For much information regarding the hill ; but it is said to be called also 
'the localities of the city, and for a the Hill of Ahasuerus (Wilson, Persian 

sketch map of the town, I am indebted Life, p. 157), and I once heard it called 

to Aram Zohrabian, a native of Kama- Fortress of Dara (Darius). But this 

dan. For several notes I have also to term was given by Mirza Sahak, a Per- 

thank the courtesy of Mr. H. L. Rabino, sian of considerable education, who 

British Consular Agent at Kerman- may have had some knowledge of the 

shah. story of Darius Codomannus and his 

2 Musallah, ' citadel,' is the name overthrow by Alexander the Great, 
by which the people generally know 



148 HAM AD AN, THE ANCIENT ECBATANA 

a channel to be dug at vast expense to conduct the water from 
a lake on the other side of Mount Orontes into the town. 1 

The population of Hamadan is estimated conservatively at 
twenty-five thousand, which is probably less than in ancient 
times, when the city was a great metropolis. 2 The majority of 
the inhabitants are Iranian in blood, of the old Median stock, 
with some slight infusion of the Arab strain, and there are 
many Turks, both in the city and in the neighboring villages. 
The Turkish characteristics, however, are by no means so 
noticeable as at Urumiah, and my servant Safar, who spoke 
Persian fluently, now began to make jokes at the expense of 
Shahbas, the caravan-leader, because his speech was confined 
to the Turkish dialect of Azarbaijan and he could not converse 
freely with all the people. Among the inhabitants are also a 
considerable number of Armenians, about three hundred of 
whom occupy a special section of the town, and there is likewise 
a Jewish quarter, in the southern section, the community being 
estimated at five thousand souls, a figure not unlikely, since 
there have been Jews in Hamadan from the earliest times. 
The Europeans in the city are few in number. They are 
engaged chiefly in the work of the American Presbyterian 
Mission, which established two schools here for boys and girls 
in 1880, and of the Alliance Israelite, which began work among 
the Jews of Hamadan in 1900. 

A tour of the bazaars will give the best idea of the Oriental 
population and of the city's commercial status. For the most part 
the bazaars are of the familiar type, vaulted over, and they con- 
tain more than five hundred busy shops. As a matter of course 
they are interesting, although only of secondary importance 
when compared with those in some of the large Persian towns, 
because the line of caravan trade turned for a time somewhat 

1 See the fragments of Ctesias pre- 2 The population is put at forty thou- 

served in Diodorus Siculus, 2. 13. 5, sand in the Sixty-seventh Annual Re- 

ed. Gilmore, pp. 51-52, London, 1888, port of the Board of Foreign Missions 

and transl. Booth, Diodorus the Sici- of the Presbyterian Church in the 

lian, 1. 110-111, London, 1814. U.S.A. p. 233, New York, 1904. 



IN THE BAZAARS AND AT HOME 149 

aside from Hamadan, and cholera and famine decimated its 
inhabitants. Nevertheless the commerce has been restored in 
recent years, and the merchants speak of the town as 4 the store 
(ambdr) of Persia.' Among the chief articles of trade are the 
leather goods, as Hamadan is a city of tanneries and the Hama- 
danis are renowned for their skill in dressing, tanning, and work- 
ing hides and skins, and in manufacturing from them useful and 
ornamental goods and articles. Saddles, straps, trunk-covers, 
and pointed shoes are on sale at the various booths, while the 
soft white felts (namad) used for mats, saddle-cloths, shepherd 
coats, and helmet-shaped caps, can hardly be excelled in 
Persia. l I still have as a memento of Hamadan's industrial 
work a white felt saddle-cloth (ghdslali) and two carpet saddle- 
bags (khurjlri) purchased for me by Safar in the bazaar. The 
caravansarais, which adjoin the bazaars or are located in the 
vicinity, number more than fifty and do a flourishing business 
in providing accommodation for the throngs of merchants and 
pilgrims that pass through the city. There are likewise fully 
sixty public baths which derive an added income from the same 
source. 

During my stay in Hamadan I was a guest of the American 
missionaries and thus enjoyed a taste of home life again after 
the wearisome days of halting at caravansarais. At the mission 
house I made also my first acquaintance with the true Persian 
cat of higher breed. Persia is sometimes called the land of 
cats, but the best specimens come from the mountains of 
Kurdistan and are a great contrast to the ordinary village cat, 
which skulks about the caravansarai and is often a nuisance to 
the traveller because of its thievish propensities, leading me 
sometimes to say to Safar that I should prefer the mice if I 
had my choice. The two Hamadan tabbies to which I refer 
were superb creatures, larger than the largest Angora cats. 

1 Additional and detailed informa- Diplomatic and Consular Beports, 
tion regarding the trade of Hamadan Persia, no. 3189, annual series, pp. 
may be obtained from Rabino, British 41-46, London, 1904. 



150 HAM AD AN, THE ANCIENT ECBATANA 

One of them was pure white, the other was partly black, but 
in temperament both were exactly alike, and their tricks and 
gambols together were an endless source of amusement to 
me. 

The most modern features of Hamadan are a telegraph office 
and a bank. The banking establishment is a branch of the 
Imperial Bank at Teheran and has for its director an Arme- 
nian Persian. The residence of this gentleman, to whose 
courtesy I was also indebted, was partly furnished in European 
style, adapted to Oriental needs, and it gave a good example of 
how ready the Iranian is, under certain conditions, to combine 
Western civilization with Eastern life. But it is not the mod- 
ern phase of Hamadan that is interesting ; the chief interest 
in the place centres in its antiquity, and I shall devote the 
rest of the chapter to the historic associations of the city. 

In the first place, the name Hamadan can be traced back to 
the earliest times. During the Sasanian era it was known as 
Hamatdn? and in the ancient Persian inscriptions as Hag- 
matdna? which means literally ' a place of meeting or convening, 
concourse of many ways,' and Hamadan to-day is a meeting- 
place of as many highways from various parts of the kingdom 
as when it was the Median capital. 3 In the Babylonian in- 
scriptions it appears as Agamatanu.* The Greek writer Ctesias, 
who knew Persian well, renders the name correctly as Agbatana 
(*A7/8aTam), although most of the Greeks called it Ekbdtdna 
('E/c/3drava), with initial E, not A, and giving to the penulti- 
mate vowel the wrong quantity (a for a). 

1 Pahlavi Hamatdn, Bd. 12. 12 ; less important places, including Tabriz, 
22. 6. Urumiah, Kazvin, Teheran, Isfahan, 

2 OP. H M g a m a t a an a , Bh. 2. 76, Kermanshah, and other cities known 
77, from ham, 'together,' and gam, in antiquity. 

'to go,' i.e. 'Co-ventry' (?). Bar- 4 Babylonian A-ga-ma-ta-nu, Bh. 

tholomae, Air. Wb. p. 1744, regards the 60, and Nabonid, Annaleninschr. Av. 

etymology as uncertain. 2. 3, 4. Cf. also Bang, Melanges 

8 From the top of the Musallah, Charles de Harlez, p. 8, Leiden, 1896 ; 

for example, it is possible to count and Streck, Armenien und Westper- 

roadways leading to twelve more or sien, in Zt. f. Assyr. 15. 367. 




PERSIAN CATS AT THE MISSION HOUSE, HAMADAN 



A FAMOUS CITY OF OLD 151 

The antiquity of these allusions to the name shows how 
ancient the history of the city really is. Its existence as early 
as the twelfth century before Christ is believed to be vouched 
for by the mention of Amadana in an Assyrian inscription of 
Tiglath-Pileser I (c. B.C. 1100), 1 and we certainly have in- 
scriptional evidence to prove that it must have existed in the 
time of Ramman-nirari, husband of Semiramis (Sammuramat), 2 
or, according to some authorities, before the end of the ninth 
century B.C. Ctesias, who was court physician to Artaxerxes 
Mnemon for seventeen years (B.C. 416-399), and well acquainted 
with Persian traditions, says that 'when Semiramis came to 
Ecbatana, which is situated in a low, level plain, she built a 
palace and bestowed more care and attention upon it than she 
had done at any other place ' ; and Ctesias describes the water- 
courses which she caused to be constructed from Mount Alvand 
to Hamadan. 3 According to the same author, moreover, when 
Sardanapalus, the last Assyrian king, was overthrown by 
Arbaces, and the Median kingdom established, Arbaces con- 
veyed to Ecbatana all the gold and silver treasure which was 
taken from Nineveh after its capture ; but the historic accuracy 
of this report is, perhaps, open to question. 4 

The more generally accepted account of the founding of 
Ecbatana is that given by Herodotus, who ascribes its origin 
to Deioces, the first great ruler of the Median Empire, about 
B.C. 700. The description of the place is not only picturesque, 
but important for the purpose of identification, because of the 
allusion to the walls. 

1 Spiegel, Erdnische Alterthums- and the translation by Booth, Diodorus 
kunde, 2. 246, and Browne, Literary the Sicilian, 1. 124. Compare also 
History of Persia, p. 20. Rawlinson, JRGS. 10. 125, and 

2 See Justi, Geschichte /raws, in consult, for Arbaces, Justi, Ge- 
Grundr. iran. Philol. 2. 404. schichte Irans, in Grundr. iran. Phi- 

3 These statements of Ctesias are lol. 2. 407-408. For other traditions 
preserved in Diodorus Siculus, 2. 13. 5. regarding the founding of Hamadan, 
See p. 148, n. 1, above, for references. see Modi, jShatroihd-i Airan, or Cities 

4 For the text of Ctesias, see Gil- of Iran, p. 151, Bombay, 1899. 
more, Fragments of Ktesias, 29, p. 90, 



152 HAM AD AN, THE ANCIENT ECBATANA 

'Deioces built the massive and strong- walled city now called 
Ecbatana ('Ay/Jarcwx), the walls being arranged in circles one within 
the other. The rampart is planned in such a manner that each circle 
rises higher than the one preceding it by the height merely of the 
battlements. The nature of the ground, which is a gentle hill 
(xoXtuvos), is favorable for carrying out such a design ; and, as 
there are seven circles in all, particular care was taken to have the 
royal palace and the treasury within the innermost circle. The cir- 
cuit of the outer wall is nearly as large as that of the city of Athens. 
Of this first circle the battlements are white ; the second, black ; the 
third, red; the fourth, blue; and the fifth, orange. The battle- 
ments of all the circles are decorated in this manner with colors, 
but those of the two last are incrusted, the one with silver, the other 
with gold. Such were the palace and the surrounding fortifications 
which Deioces built for himself; but the rest of the people he 
ordered to build their houses round about outside the wall.' l 

The Old Testament apocryphal book of Judith attributes the 
founding of Ecbatana (called Achmetha in Aramaic, Ezra 6. 2) 
to a king named Arphaxad, whose historic identity has not been 
satisfactorily proved, and gives an elaborate description of the 
city's walls, towers, and gates. I quote the account in full, 
although in the eyes of most critics it has no greater value 
than that which we should attach to an Oriental romance. v 

'1. In the twelfth year of the reign of Nabuchodonosor, who 
reigned in Nineveh, the great city ; in the days of Arphaxad, which 
reigned over the Medes in Ecbatane, 2. And built in Ecbatane 
walls round about of stones hewn three cubits broad and six cubits 
long, and made the height of the wall seventy cubits, and the 
breadth thereof fifty cubits: 3. And set the towers thereof upon 
the gates of it, a hundred cubits high, and the breadth thereof in 
the foundation threescore cubits : 4. And he made the gates thereof, 
even gates that were raised to the height of seventy cubits, and the 
breadth of them was forty cubits, for the going forth of his mighty 
armies, and for the setting in array of his footmen: 5. Even in 
those days King Nabuchodonosor made war with King Arphaxad in 
the great plain, which is the plain in the borders of Kagau. . . . 

1 Herodotus, History, 1. 98, 99. Cf. 194, London, 1862 ; H. C. Rawlin- 
also G. Rawlinson, Herodotus, 1. 191- son, JRGS. 10. 120-127. 



RAMPARTS AND BATTLEMENTS 153 

13. Then he marched in battle-array with his power against King 
Arphaxad in the seventeenth year, and he prevailed in his battle : 
for he overthrew all the power of Arphaxad, and all his horsemen, 
and all his chariots, 14. And became lord of his cities, and came 
unto Ecbatane, and took the towers, and spoiled the streets thereof, 
and turned the beauty thereof into shame.' l 

According to Polybius, the Greek historian, who wrote in 
the second century B.C., the magnificence of Ecbatana was such 
as to call for special notice, especially regarding the splendor 
of its palace and the temple of Aena or Anaias. This name is 
none other than a disguised form of Anahita or Ana'itis, the 
Zoroastrian goddess of the waters, who is celebrated in the 
Avesta and whose worship was especially popular among 
the Persians after the time of King Artaxerxes II (fourth cen- 
tury B.C.). 2 From other classic sources we know that this 
goddess had a famous temple at Ecbatana. 3 The description 
in full is as follows : 

' It was originally the royal city of the Medes, and vastly superior 
to the other cities in wealth and in the splendour of its buildings. 
It is situated on the skirts of Mount Orontes, and is without walls, 
though containing an artificially formed citadel fortified to an aston- 
ishing strength. Beneath this stands the palace, which it is in some 
degree difficult to describe in detail, or to pass over in complete 

1 Judith, 1. 1-14. Yakut, p. 597, ing no real historic importance, but as 

asserts that Hamadan was founded by inserted to lend an air of antiquity to 

one of the great-grandsons of Noah, the Apocryphal story; see Cheyne, 

and some of the older Occidental Encyclopaedia Biblica, s.v. Arphaxad, 

scholars used likewise to point to the Judith. The value of the description 

name of Arphaxad, grandson of Noah, is open to the same attack, 
in Genesis 10. 22; 11. 10-14, and 2 See Windischmann, Die persische 

associate with it the tradition that Anahita oder Ana'itis, p. 5, in Abhandl. 

Media was so called from Madai, son of kgl. bayr. Akad. 'Wiss. Bd. 8, Abthl. 1, 

Japhet (Gen. 10. 2); see Ker Porter, Munich, 1856; compare also pp. 237- 

Travels, 2. 94. Some attempts have 242, below. 

likewise been made to identify Ar- 3 See Isidorus Characenus, Man- 

phaxad, as founder of Ecbatana, with stones Parthicae, 6, and Plutarch, 

Deioces or with his son Phraortes ; Artaxerxes, 27. 3 ; cf. Windischmann, 

but the trend of modern scholarship is Die persische Anahita, pp. 6, 13. 
to regard the name Arphaxad as hav- 



154 HAM AD AN, THE ANCIENT ECBATANA 

silence. To those authors whose aim is to produce astonishment, and 
who are accustomed to deal in exaggeration and picturesque writing, 
this city offers the best possible subject ; but to those who, like myself, 
are cautious when approaching descriptions which go beyond ordi- 
nary notions, it presents much difficulty and embarrassment. How- 
ever, as regards size, the palace covers ground the circuit of which is 
nearly seven stades ; l and by the costliness of the structure in its sev- 
eral parts it testifies to the wealth of its original builders : for, all its 
woodwork being cedar or cypress, not a single plank was left un- 
covered ; beams and fretwork in the ceilings and columns in the 
arcades and peristyle were overlaid with plates of silver or gold, 
while all the tiles were of silver. Most of these had been stripped 
off during the invasion of Alexander and the Macedonians (B.C. 335), 
and the rest in the reigns of Antigonus (B.C. 325-301) and Seleucus 
Nieator (B.C. 312-280). However, even at the time of Antiochus's 
arrival (i.e. Antiochus the Great, B.C. 210), the temple of Aena still 
had its columns covered with gold, and a considerable number of 
silver tiles had been piled up in it, and some few gold bricks and a 
good many silver ones were still remaining. It was from these that 
the coinage bearing the king's impress was collected and struck, 
amounting to little less than four thousand talents ($4,730,000).' 2 

It is important to note that Polybius says that the city was 
' without walls, though it contained a strongly fortified citadel 
beneath which the palace stood,' and that he says that the 
palace was built of wood, because this would render it pecul- 
iarly liable to destruction as contrasted with the stone used 
at Persepolis. 

If we now ascend the hill known as the Musallah, or citadel, 
adjoining the city of Hamadan, which is built partly upon its 
western and northwestern slope, we can understand how it 
might be possible for the town itself to be without walls, as 
Polybius asserts, and for the fortified ramparts and battlements 
to be confined, in his own words, to the c artificially formed cita- 
del fortified to an astounding strength, beneath which the 

1 Four fifths of a mile in circum- burgh, The Histories of Polybius 
ference. Translated, 2. 26-27, London, 1889. 

2 Polybius, Hist. 10. 27 ; see Shuck- See also p. 168, below. 



OF THE 

UNIVERSITY 

OF 




THE MUSALLAH HILL 156 

palace stood. ' 1 The ascent of the Musallah, I may observe, is 
steep enough for the Persian horses to start up it at a sharp 
gallop, as they always do when approaching a marked incline. 
The height, in my judgment, would answer far better than 
Takht-i Suleiman to Herodotus' description of the place as a 
Ko\a>v6s and would fulfil all the requirements of his sketch. 2 
There is room for the seven circles of walls, and the Median 
people could have 'built their houses round about outside the 
wall,' as Deioces bade them (Herodotus, 1. 99), on the very 
site still occupied by the city. Even at the present time the 
ridge of the Musallah is crowned by the remains of massive 
walls, some fifteen feet thick and twenty high, composed of 
clay, slate, brick, and small stones, and forming a parallelogram 
running northeast and southwest, in the midst of which the bed 
rock is here and there visible. Although no one would ascribe a 
great antiquity to these redoubts (in fact Agha Mohammed Shah, 
at the close of the eighteenth century, is said to have destroyed 
all remnants of antiquity at Hamadan), 3 nevertheless they may 
occupy the same position as those which crowned the KoXowo? 
in Median days and be similar to them in structure, for things 
change slowly in Persia. 4 

Herodotus, moreover, does not say that the walls of Ecba- 
tana were of stone, though this might be inferred, and it is 
almost certain that the color decorations were confined to the 
battlements raised upon them. It is true that the passage in 
Judith speaks of 4 hewn stone '; but even if we are to press the 
phrasing of that historical romance, it may be possible that the 



1 Canon George Rawlinson, Five 127, and George Rawlinson, Herodo- 
Great Monarchies, 3. 23, expresses tus, 1. 191-192, note, London, 1862. 
doubt ' whether the Median capital 3 See Ker Porter, Travels, 2. 102 ; 
was at any time surrounded by walls.' Wilson, Persian Life, p. 157. 

2 It is not clear to me how Sir Henry 4 Such is also the view of de Morgan, 
Rawlinson could have taken exception Mission Scientifique en Perse, 4. 248- 
on this score, as my careful examina- 249 ; it is worth noting that he refers 
tion of the site on two occasions con- also (p. 249) to the ruins of two Gabar 
vinced me ; see Rawlinson, JRGS. 10. towers. 



156 HAM AD AN, THE ANCIENT ECBATANA 

blocks have been carried off in later times for building pur- 
poses, for one sees in the city a number of large hewn stones 
built into the foundations of houses and possibly taken from 
such a wall, if stone, rather than the earthen breastworks, be 
insisted upon. 1 It must also be said that Canon George Raw- 
linson, when summing up his brother's claim that Takht-i 
Suleiman represents the so-called northern capital of Media, 
acknowledges that 4 of the seven walls, one alone is to be traced 
at Takht-i Suleiman ; and even here the Median structure has 
perished and been replaced by masonry of a far later age.' 2 

The other claims advanced by Sir Henry, such as the state- 
ment of Herodotus (1. 110) that the country north of Ecbatana 
is mountainous and covered with forests, seem to me equally 
applicable to Hamadan as to Takht-i Suleiman, and a compari- 
son of the two sites in detail convinced me that Hamadan has 
the right and title to being the sole heir of Ecbatana, that the 
Musallah was its citadel, and that the ruins of Takht-i Sulei- 
man, although occupying a site that may be equally old, but 
not so important as a metropolis, have a different history from 
Ecbatana. 3 

In view of all these facts I felt convinced, when standing 
upon the height of the Musallah and overlooking Hamadan, 

1 1 am informed by Aram Zohra- has changed in some slight degree, as 

bian, of Hamadan, that about a year there are evidences of ruins around the 

after the time when I was in Persia, the northeast side of the Musallah, where 

remains of a so-called Ganj, 'treasury,' old bricks are dug up and gold coins 

were discovered in the Armenian quar- are found. In notes sent me by Mr. 

ter of Hamadan and some magnificent Rabino it is claimed that the site of 

hewn stones were laid bare. the city has been changed several times 

2 George Rawlinson, Five Great and the present site of the town can- 
Monarchies, 3. 27, and n. 11. Consult not be more than five hundred and fifty 
also de Morgan, Mission Scientifique years old, but this seems to me doubt- 
en Perse, 4. 248-249. ful, and archasological researches alone 

8 My view has been anticipated, I can settle the question. In the Musal- 

find, by de Morgan, Mission, 4. 238- lab itself, however, there is little or no 

249, whose chapter on Hamadan should chance for excavation, as the earth is 

be consulted. It is possible that exca- almost entirely washed away, so that 

vations for archaeological purposes may the bed rock is visible, 
show that the position of Hamadan 



THE HISTORIC CAPITAL OF MEDIA 157 

that I was surveying the site of the capital of ancient Media. 
Here within the fortress there once stood the royal treasury 
mentioned by Herodotus, 1 and here the stronghold where 
Arbaces deposited the silver and gold found in the king's 
coffers at Nineveh. 2 Hither also the untold riches of Croesus 
were conveyed by the conqueror Cyrus ; 3 and to this strong- 
hold Alexander, following the example of his victorious prede- 
cessors, transported the wealth he had plundered from Susa, 
Persepolis, and Pasargadse. 4 Here likewise 'at Achmetha, in 
the palace that is in the province of the Medes,' was found the 
decree of Cyrus giving orders for the rebuilding of the temple 
at Jerusalem, a command which was carried into effect by 
Darius and his successor Artaxerxes. 5 The height served also 
as a Persian Bastille in ancient days, and here in a dungeon 
the Acheemenian kings caused offenders against the state to be 
confined and executed. Within its walls, for example, Darius 
put to death the Median leader Fravarti, who, like several 
others, had set up a claim to the throne while the king was 
engaged at Babylon, and had headed an army to support his 
claim. The pretender was defeated in battle at Raga, the 
modern Rei near Teheran, was taken prisoner, and met with a 
fate that is told in the Great King's own words in the rock 
inscription at Behistan : 4 Fravarti was seized and brought to 
me. I cut off his nose and his ears, and cut out his tongue, 
and put out his eyes. He was kept in chains at my door ; all 
the people saw him. Afterward I caused him to be crucified; 6 

1 Herodotus, History, 1. 98. furthermore McCrindle, Invasion of 

2 Ctesias, Fragments, 29 (cited by India by Alexander the Great, pp. 34, 
Diodorus Siculus), ed. Gilmore, p. 90 ; 126, n. 1, London, 1896. 

tr. Booth, Diodorus the Sicilian, 1.124. 5 Ezra 5. 17 ; 6. 1-3. 

3 Herodotus, History, 1. 153. 6 Or impaled' (OP. uzmayapatiy) . 

4 Arrian, Anabasis, 3. 19. 7, and cf. All these mutilations as punishments 
3. 18. 10 ; 3. 19. 2 ; also Quintus Cur- are still practised in Persia, except 
tius Rufus, Alexander, 5. 6. 1-10 ; that the barbarous mode of execution 
Diodorus Siculus, Hist. 17. 71; Plu- has given place to others. So also 
tarch, Alexander, 36-37 ; Strabo, Geog. Hiising, Elamische Studien, in Mitteil. 
15. 3. 9, cf. also 15. 3. 23. Compare Vorderasiat. Gesellsch. 3. 315. 



158 HAMAD AX, THE ANCIENT ECBATANA 

and the men who were his principal followers I imprisoned in 
the fortress (dido) at Ecbatana.' 1 

Here on the citadel height, in addition to the dungeon, the 
treasure-house, and the temples of the ancient divinities of 
Iran, 2 there once stood the palace of the ancient Median kings. 
Deioces, the founder of the empire of Media, sought seclusion 
within its fortified walls, which he himself had built, when he 
assumed the crown and withdrew himself from the ordinary 
presence of the people. 3 Probably within the same walls 
Astyages, according to legend, received the youthful Cyrus, 
who was destined later to wrest from him his crown and trans- 
fer the supremacy of Media to Persia. 4 The Persian, kings of 
the Achsemenian. line kept up the old-time prestige of the 
Median city and its citadel by making Ecbatana their summer 
capital. Here in their royal abode overlooking the plains and 
surrounded by high mountains they must have found a delight- 
ful change from Susa, whose warm climate was suited only for 
a .Aviiiter residence, or from Persepolis, their spring and summer 
home. 6 To Ecbatana likewise Alexander returned toward the 
end of his short career, in order to celebrate the glories of his 
campaigns in eastern Iran and India. 6 The Parthian dynasty 
(B.C. 250-A.D. 226) still made the city a favored place, 7 and 
Antiochus found in the palace and temple an enormous amount 
of gold and silver bullion to turn into coin to pay his soldiers. 8 
The Sasanian rulers (A.D. 226-651) were perhaps less partial to 
Hamadan, but the place was still so important when the Arabs 
captured the city (645) that they regarded this event as second 

1 Bh. 2. 73-78. Foy, Kukri's Zeit- * Herodotus, History, 1. 121-130 ; 
achrift, 35. 39-42, understands this Xenophon, Cyropaedia, 1. 3. 1-18. 
latter to mean, 'I hanged his prin- 5 Xenophon, Anabasis, 3. 5. 15, and 
cipal followers before the fortress.' Cyropaedia, 8. 6. 22; Strabo, Geori. 

2 The temple of Anaitis referred to 11. 323 ; Quintus Curtius Rufus, Alcx- 
above, p. 153, and probably here like- ander, 5. 8. 1. 

wise the shrine of the Persian ' vEscu- 6 See pp. 163-165, below. 

lapius ' alluded to by Arrian, Andb. 7. 7 Quintus Curtius Rufus, Alexander^ 

14. 6. 5. 8. 1. 

8 Herodotus, History, 1. 99. 8 See p. 154, above. 



SCENES OF BY-GONE GLORY 159 

only to the triumph at Nahavand over the army of the house 
of Sasan. 1 Early in the tenth century Hamadan was stormed 
by Mardavij ibn Ziar of Gilan; 2 in the thirteenth century by 
Tamerlane ; and once again, five centuries later, sacked and 
pillaged by Agha Mohammed Shah. It is no marvel that this 
ancient home of kings and scene of great events is but a shadow 
of by-gone glory. Its fate is best told in the verses of the 
short poem from which I quoted at the beginning of the 
chapter. 

' Gone the great sun-temple where 

Golden stair rose over stair ; 

Gone the gilded galleries, 

Porticoes and palaces ; 

And the plaintive night winds plead 

For the memory of the Mede, 

Sob for alien ears to heed, 

Pilgrim train and caravan, 

Round the walls of Hamadan. 

Nought of all the radiant past, 

Nought of all the varied, vast 

Life that throbbed and thrilled, remains 

With its pleasures and its pains, 

Save a couchant lion lone, 

Mute memorial in stone 

Of three empires overthrown 

Median, Persian, Parthian 

Round the walls of Hamadan.' 

The famous but battered stone lion referred to in the verses 
as the only monument that has lasted through the long ages of 
Hamadan now lies near the foot of the Musallah not far from 
the road leading to Isfahan. It is one of the landmarks of 
Hamadan and is regarded as a guardian genius of the town. 
Even a thousand years ago it was spoken of by Masudi as very 
ancient, and he describes it as standing by the 'Lion Gate' 

1 See Yakut, p. 598, and Justi, chap. 130, ed. Barbier de Meynard, 
Geschichte /raws, in Grundr. iran. 9. 21-22, Paris, 1877 ; and compare 
Philol. 2. 546. Horn, Gesch. Irans in Islam. Zeit, in 

2 See Masudi, Les Prairies d'Or, Grundr. iran. Philol. 2. 564. 



160 II AM AD AN, THE ANCIENT EC BAT AN A 

(Bab al-Asad) on a low hill overlooking the road to Rei and 
Khorasan. 1 He speaks of its lifelike appearance and compares 
it in size to some great bull or a crouching camel, adding that 
it was carved after Alexander's return from Khorasan (as native 
tradition ascribes the founding of Hamadan to Alexander) and 
set up as a talisman to protect the walls of the city and its 
inhabitants, which were destined to be safe as long as the lion 
was not thrown down or broken. The overthrow of the lion 
was accomplished, he tells us, about his own time, when the 
army of Mardavij stormed Hamadan, as stated above, 2 and this 
event was accompanied by utter disaster, which fulfilled the 
prophecy. A legend almost as old, recorded by Yakut (about 
A.D. 1220), says that the image was set up by Belinas as a 
talisman against the severe winters of Hamadan. 3 If effica- 
cious, we can hardly imagine what would be the rigor of the 
climate without its influence, for Hamadan is one of the coldest 
places in Persia. In the absence of authentic history we can 
only resort to legend and tradition regarding this lion monu- 
ment and its origin, but popular belief has certainly surrounded 
the sculptured stone with a deep veneration in the eyes of the 
people. Dozens of superstitions are attached to it. Mothers 
hold up their babes to pat the huge beast or kiss its face, 
barren women touch its brow to remove the curse of sterility, 
and pilgrims lay offerings of stones, some of them carved, upon 
its head as a coronet or on the block below its mouth. 

From the standpoint of art the lion is rather effective in the 
distance, as the mutilation of the stone does not then show, 
and I was impressed by the lifelike appearance of the image as 
I first rode toward it, an effect which is enhanced by the 

1 Masudi (died 951) devotes a para- Belinas also placed other talismans, 
graph to this monument in his no longer in existence, to the right of 
Meadows of Gold, chap. 130, see Les the statue to protect the people of 
Prairies d'Or, ed. Barbier de Mey- Hamadan against snakes, scorpions, 
nard, 9. 21-22. insects, and floods. Belinas is com- 

2 See p. 159, above. monly explained as a corrupt Oriental 
8 Yakut, p. 606, who adds that form for Plinius, Pliny. 




THE LION OF HAMADAN 







THE LION'S FACE 
(Votive stone offerings under the mouth) 



THE FAMOUS STONE LION 161 

yellowish sandstone out of which the figure is carved. The 
head is massive, and the heavy waves of the mane are realistic 
in appearance, but it is difficult to catch the exact expression of 
the face in its present prone position, although the chin is well 
marked and the jaws are partly open. A deep hole in the 
forehead mars the expanse of the brow, and the face is smeared 
by dirty hands and is greasy from the oil which pilgrims pour 
upon it. Between the shoulders and in the back there are eight 
or more holes, due to erosion, and the rain settling in these cavi- 
ties has tended to enlarge them, so that several are big enough 
for the fist to be inserted. Although the legs of the creature 
are broken off at the shoulders and thighs, the body is entire 
and not split by a crack as the reproductions in some books, 
since the time of Flandin, would lead one to suppose. 1 A care- 
ful examination of the sculpture shows that the lion originally 
sat in an upright posture with the forelegs straight and with- 
out any curve from the shoulders except the natural rounding 
at the haunches. In other words, it was a lion sejant, not 
coucTiant. The right hip is lower than the left, and the tail, 
though missing, curved around the left flank, as is shown by a 
perceptible groove in the stone at that point. From head to 
tail the image measures between eleven and twelve feet (3.40 m.), 
the head itself being nearly forty inches in diameter (1 m.). 

The present position of the lion, about an eighth of a mile 
from the foot of the Musallah, and facing south, is probably due 
to chance. Both Masudi and Yakut speak of the sculpture as 
being near a gate of the city, and, judging from a modern mud 
tower which guards the road at this point, it is possible that 
there once was a gate near by, or that the lion possibly guarded 
an entrance to the citadel height at this spot. 

Concerning the age of the statue we can only make guesses, 
reckoning back from the time when Masudi spoke of it a 

1 See, for example, Flandin and Rawlinson, Five Great Monarchies, 
Coste, Voyage en Perse, Ancienne, 3. 92 ; Justi, Geschichte des Alten 
1. pi. 25 ; Texte, p. 17 ; George Persiens, p. 5, Berlin, 1879. 



162 HAM AD AN, THE ANCIENT ECBATANA 

thousand years ago. On the whole I agree with those who 
attribute a great antiquity to the sculpture, assigning it even 
to the times of the ancient Median kingdom, when it may have 
anticipated the lion of the royal Persian emblem. 

Not far from the lion and in the southeastern section of the 
city, there is a towerlike structure which is generally called 
by the people Burj-i G-urbdn, or Kurbdn, * Tower of Sacrifice,' 
but was pointed out to me as the Zardushtldn, or Zoroastrian 
fire-temple, by Mirza Sahak, an intelligent Persian. One 
morning, after coming down from the mound of the citadel, 
I paid a visit to this partly ruined turret. I found a structure 
decagonal in shape and built of ordinary Persian bricks, not the 
large sun-dried bricks as at Rei and in the fire-temple near 
Isfahan. It reminded me, in appearance, of another towerlike 
building which I had visited in the northern part of the city, 
the Gumbad-i Alavian, and like the latter it gave me the 
impression of being no older than the thirteenth century, the 
time of the Mongol sway. I could not see anything particu- 
larly Zoroastrian about it, nor was there anything in its interior 
to support such a view. The ten wall-spaces on the inside were 
set off by pointed niches, and there were four small window 
openings high up, which gave a good light. The woodwork 
around these openings had been partially burned away, evi- 
dently by some accidental fire. There were also one or two 
charred pieces of beams here and there in the walls. Other- 
wise the interior was empty, and the only opening in the floor 
was an irregular descent into what was probably once a tomb. 
Within the walls of the building the natives on certain occa- 
sions sacrifice a camel, which accounts for the name, ' Tower of 
Sacrifice,' by which the structure is known. 1 

1 This rite may be a survival of some tenth day of the twelfth Mohammedan 

ancient rite, like the animal sacrifices month of Zi-hi)ja, in commemoration 

in the Avesta, Yt. 5. 21, 25, 33, etc., of Abraham's offering up his son Isaac 

and the festival l ld-i kurbdn, 'id-i azhd, (or, according to the Mohammedans, 

in which sheep are sacrificed even by Ishmael), as I am informed by Kho- 

Zoroastrians as well as Moslems on the dabakhsh Bahram llais, of Yezd, in a 



TOWERS AND SHRINES 163 

The Gumbad-i Alavian, to which I have alluded as being in 
the northwestern part of the town, is a shrine perhaps five or 
six centuries old, belonging to the Seljuk era, and sacred as a 
tomb ; but its only interest for us consists in the stucco work 
of the interior. This is artistic alike in design and in execu- 
tion, and there are also some scrolls of fine workmanship and 
Kufic inscriptions from the Koran, which are very ornamental. 

Probably older, as far as concerns the site which it occupies, 
is the Kalah-i Kohnah, 4 Ancient Citadel,' on the southern side 
of Hamadan, lying in the direction of Mount Alvand (see sketch 
map). If archaeological researches should be undertaken in 
the neighborhood of this structure they might possibly yield 
interesting relics, as Hamadan is rich in antiquities, and the 
soil of the plain, when systematically dug, washed, and exam- 
ined, as is done by some enterprising natives, yields a consider- 
able harvest of old coins, seals, jewels, and other valuables, 
which repays the labor involved. 1 Gold- washing, in fact, is a 
regular occupation at Hamadan, and systematic washing of the 
fields has been carried on for the past twenty years. The yield 
in coins and nuggets has decreased, but the amount of gold-dust 
that is found gives a fairly good profit. 

I have spoken of Alexander the Great in connection with 
Hamadan, and we know from history that he twice visited this 
ancient capital of Media, once when pursuing the vanquished 
Darius Codomannus, and afterward when returning from 
Bactria and India. His name is still well known among the 
people as Iskandar and various legends about him are preserved 
to the present time. A building, for example, near the bridge 
leading over the stream to the Musallah, is said to occupy the 
site of the 4 Governor's Palace,' where he is supposed to have 

letter dated May 14, 1905. See p. 371, 1 See also the remarks by Wilson, 

below, and cf. also Pietro della Valle Persian Life, p. 157 ; and compare the 

(1617), Viaggi, 1. 536; Travels, ed. reproductions of the finds in the collec- 

Pinkerton, 9. 36 ; Tavernier, Travels, tion of de Morgan, Mission Scien- 

p. 143. tifique, 4. 250-251. 



164 HAM AD AN, THE ANCIENT ECBATANA 

stayed, and evil tales regarding his habits of excessive drink- 
ing still linger among the people. They repeat a story that 
one night in his cups he boasted how rapidly he was conquer- 
ing the world. 4 Sire,' said one of his generals, ' it is through 
your father's soldiers that you win such successes.' Angered 
at the rebuke and inflamed with wine, he caused his remonstra- 
tor to be put to death forthwith. On the following day, not 
knowing what he had done, he called for the general and then 
learned of the crime he himself had committed. The site of 
the supposed tomb of his victim is pointed out on the street 
overlooking the stream and is called Grabr-i Iskandar, 4 Grave 
of Alexander,' for the common people more generally believe 
that the World-Conqueror himself is buried there, rather than 
that it is the grave of Alexander's officer. The so-called 
sepulchre is nothing more than a recess in a rounded bastion of 
clay, mortar, and stone, that now forms part of the foundation 
of a mud house which is occupied as a dwelling and is entered 
by a small door, a foot and a half wide and two feet high. 
I made no attempt to inspect the supposed crypt more closely. 1 
The story as it is told in Hamadan seems to contain a remi- 
niscence of the death of the general Clitus, whom Alexander 
slew with his own hand in a fit of drunken madness because he 
ventured to rebuke the conqueror an event generally said to 
have taken place at Samarkand combined with a tradition 
of the loss of his favorite, Hephsestion, who died at Hamadan 
and whose death Alexander mourned in a wild despair. 2 
Plutarch describes the circumstances attending upon this 

1 Those natives of Hamadan who and the body was accordingly interred, 

maintain that Alexander is really This note I have on the authority of 

buried in their city narrate a legend to Mr. H. L. Rabino of Kermanshah. The 

the effect that he gave orders that after symbolism in the legend can easily be 

his death his body should be carried recognized. 

with outstretched arms, holding earth 2 See Plutarch, Alexander, 60, 51, 

in the hand, about the kingdoms which 72, ed. Bekker, Leipzig, 1858 ; transl. 

he had conquered. His corpse should Langhorne, 5. 256-269, 282-283 ; cf. 

be buried wherever he withdrew his McCrindle, Invasion of India, p. 43, 

hand. This happened at Hamadan London, 1896. 




OF THF 

UNIVERSITY 

OF 






-** '^ 




BRIDGE OVER THE RIVER AT HAMAD AN MOUNT ALVAND IN 
THE BACKGROUND 



LEGENDS OF ALEXANDER 165 

latter event. Alexander had returned to Ecbatana from 
India, and on reaching the ancient capital, of which he was 
now the victorious lord, he gave himself up to celebrating his 
successes with all the wanton luxuriousness of the East, for 
his habits had been growing more and more Asiatic, much to 
the distress of his hardier Macedonian leaders. The rejoicings 
were accompanied by games and public festivities conducted in 
regal fashion. In the midst of these celebrations, which Plu- 
tarch pictures as little better than drunken orgies, Hepheestion 
died. 4 Alexander's grief knew no bounds,' he says. 4 He 
immediately ordered the manes of the horses and mules to 
be shorn as a sign of mourning, and tore down the battlements 
of the towns in the vicinity; 1 he caused the unfortunate physi- 
cian who had attended Hephsestion to be impaled, and forbade 
the flute and all other music in the camp for a long time.' 2 
Plutarch then describes how Alexander ravaged the country 
round about, taking vengeance on the people for the death of 
his minion, putting to the sword all those that were of youth- 
ful age and calling this 'a sacrifice to Hephsestion.' He 
designed a superb tomb for the dead favorite, which was to 
cost ten thousand talents arid be executed by the celebrated 
Greek architect Stasicrates ; but whether the mausoleum was 
ever built or whether the body was less magnificently interred 
at Ecbatana or perhaps even embalmed and carried with Alex- 
ander to Babylon, is not known. Nor is it at all certain that 
the so-called sarcophagus of Alexander preserved in Constanti- 
nople was really the coffin of the great conqueror. 3 

Another tomb, which is less known, but in reality more inter- 
esting, is that which contains the body of the great physician 
and philosopher, Ibn Sina, or Abu AH ibn Sina, better known 

1 This is the statement of Plutarch, traditions about Alexander, compare 
72, T&V irtp 7r6Xew>, but generally Ker Porter, Travels, 2. 99-101. 
understood to refer to the walls of EC- 3 For a picture of the so-called 
batana ; see ^Elian, Hist. 7. 8; cf. Lang- Alexander sarcophagus at Constanti- 
horne, Plutarch's Lives, 5. 283, n. 190. nople, see Skrine and Ross, Heart of 

2 Plutarch, Alexander, 72. For other Asia, p. 9, London, 1899. 



166 HAMAD AN, THE ANCIENT ECBATANA 

to the West as Avicenna. This remarkable man flourished 
about A.D. 1000 and was one of the finest intellectual forces 
that the Orient ever produced. He was a native of Bokhara 
in Turkistan, but lived long in Persia, spending his last days 
at Hamadan. His famous work on medicine, written in 
Arabic, but based on Greek authorities, ranks as a standard in 
the East and, through the medium of translations, enjoyed 
such distinction in Europe, some hundreds of years ago, 
that Chaucer refers to it in the Pardoner's Tale as a work 
familiarly known to his readers. 1 Avicenna's metaphysical 
writings, which were likewise affected by Greece, being influ- 
enced by Aristotle and Neo-Platonism, also found their way 
indirectly into Europe through the so-called Arabian philosophy 
of the Moors in Spain and thus affected scholastic philosophy. 2 
In addition to his renown as a physician and philosopher 
Ibn Sina had some repute as an occasional writer of verse ; 
some of his quatrains anticipate Omar Khayyam by a century. 
I paraphrase a stanza which is particularly Khayyamesque in 
tone and looks almost as if 'that large infidel' might have 
written it. 

1 From Earth's dark Centre unto Saturn's Gate 
I've solved all Problems of this World's Estate, 
From every Snare of Plot and Guile set free, 
Each Bond resolved saving alone Death's Fate.' 8 

1 Chaucer, Pardoner's Tale, 889- ' From the abysm of the dark earth to 
891. the height of Saturn, 

2 See Browne, Literary History of I have made all mysteries of the world 
Persia, p. 381. resolved ; 

8 To show I am not over-influenced I have leaped free from the fetters of 
by FitzGerald (quatrain 31) I append all machinations and guile ; 

the Persian text of the quatrain and Every bond has been resolved, except 
translate it literally : the bond of Death.' 

az ka'r-i gil-i sldh td auj-i ziihal Instead of gil there is a reading gul, 

kardam hamah mushkildt-i gltl rd hal 'from the root of the dark rose,' and 

blrun jastam zi-kaid-i har makr u hll 'dlam, 'world,' as a synonym for gltl 

har band kushddah shud magar band-i in the second line. The text of this 

ajal quatrain is given by Eth6, Nachrichten 




TOMB OF AVICENNA (!BN SINA) 




ToMl! OF KSTHKK AM) MoKDKCAI 



THE TOMB OF AVICENNA 167 

The tomb itself is a simple brickwork building, rectangular 
in shape, and surrounded by an unpretentious walled court- 
yard which is haunted by dervishes, pilgrims, and loiterers. 
A carved and inscribed slab covers the dust of the great thinker, 
and by his side rest the remains of his contemporary, Sheikh 
Abu Sai'd, the Persian mystic poet and author of quatrain 
verses in allegorical and symbolic style, who is said to have 
been acquainted with Ibn Sina. 1 A modern inscription written 
inside the tomb records the fact that this final resting-place of 
' His Holiness Sheikh Abu Sa'id and the Prince of Sages, Bu 
Ali Sina (Avicenna),' had fallen into ruins and had been 
restored by the Princess Nigar Khanum of the royal line of 
the Kajar family in the year 1877 (A.H. 1294). 2 

Still another poetical shrine, situated not far from the Gum- 
bad-i Alavian in the northwestern section of the city, is the 
tomb of another pre-Khayyamite, the dervish poet Baba Tahir 
Uryan (d. 1019), a native of Hamadan, whose verses are in 
especial favor with the Persians because of their sweetness and 
their moral tone, even though tinged with the tender melan- 
choly which marks the dervish character. 3 

Among the various tombs in the city by far the most inter- 
esting, because of its traditional claims upon the student of the 

von der Kgl. Ges. d. Wiss. zu Gottingen, And leapt out free from bonds of 

p. 558, Gottingen, 1875, and by Pizzi, fraud and lies, 

Chrestomathie Persane, p. 89, Turin, 'Yea, every knot was loosed, save that 

1889 ; for a German version of the of death.' 

stanza consult Horn, Geschichte der 

Persischen Litteratur, p. 150, Leipzig, 1 See Eth( ' **!>"***" Litteratur, 

1901. For the phraseology of Omar's in Grundr ' iran ' Phil 1 ' 2 ' 275 ' 

quatrain (no. 303) which resembles 2 A verse from Hafiz on springtime 

Ibn Sina's in many respects, see Whin- and the Divine Love is added ' and 
field, Quatrains of Omar Khayyam, thanks are S iven to God for the rest - 
pp. 204-205, London, 1883, who ren- ration of the shrine ' I am ind ^ted 

ders Khayyam's stanza : - to Mr ' Rabino for a CO W of the Persian 

inscription. 

4 1 solved all problems, down from 3 See Heron -Allen and Brenton, 

Saturn's wreath Lament of Baba Tahir, London, 1902 ; 

Unto this lowly sphere of earth be- and Ethe', Neupersische Litteratur, in 

neath, Grundr. iran. Philol. 2. 223. 



168 HAM AD AN, THE ANCIENT ECBATANA 

Bible, is the sepulchre alleged to be the burial-place of Esther 
and Mordecai. This is situated in an old Jewish cemetery 
south of the centre of the city and is said to occupy the same 
site as the original tomb, which was demolished when Tamer- 
lane sacked Hamadan. 1 The building is a small brick struc- 
ture with a high pointed cupola that has lost most of its stucco 
and tiles. The entrance is an unpretentious arch pierced by a 
very low door, which is made of a single stone turning heavily 
on rough-hewn pivots carved from the stone itself and set in 
deep sockets. 2 I had to crouch to pass through, and then found 
myself in a low winding passage leading into the crypt. The 
dingy walls of this vaulted room are so discolored by the smoke 
from tapers and ill-fed lamps used to light the hundreds of 
pilgrims who visit the shrine (Mohammedans as well as Jews) 
that the texts and graffiti in various languages are hardly 
noticed. Side by side in the middle of the chamber are the 
two graves, each covered by an ark-shaped sarcophagus made of 
ebony, one slightly smaller than the other, and inscribed with 
Hebrew letters. These and the inscriptions on the walls con- 
tain texts from Esther and eulogies of the Jewish heroine and 
of Mordecai, together with various other records. 3 Fragments 
of parchment scrolls of the Scriptures, crumbling, but too sacred 
to destroy, are scattered about, placed here for safe-keeping as 
in the Jewish Crenizahs. Regarding the authenticity of the 
graves scholarly criticism is of the opinion that the Jews of 
Hamadan are the victims of a pious delusion and that the tradi- 
tion that the tomb represents the sepulchre of Esther and 
Mordecai (whose names are inscribed on the cenotaphs) has no 
historic foundation. The Hebrews themselves in the city 

1 Cf. Ker Porter, Travels, 2. 108. but the latest scholarly treatment of 

2 There are socket-holes of similar these inscriptions is by Israel Le'vi, 
pivots in the square building opposite lievue des Etudes Juives, 36. 237-255, 
the tombs at Naksh-i Rustam, which is Paris, 1898, and by Kaufmann, op. cit. 
described at p. 302, below. 37. 303-304. A picture of the two 

8 Ker Porter, Travels, 2. 107, de- sarcophagi may be seen in Flandin, 
scribed them nearly a century ago, Voyage en Perse, Moderne, pi. 69. 



THE TOMB OF ESTHER AND MORDECAI 169 

never question the authenticity, however, and firmly believe 
that miracles are wrought at the sepulchre, especially at the 
time of the Purim festival. 1 To them the Biblical account of 
Esther is not a work of fiction, but a record of fact, of scenes 
enacted in this city where Xerxes, with whom King Ahasuerus 
has been historically identified, had his summer residence and 
carved an inscription on Mount Alvand. Haman, his minister, 
is a living reality in their eyes, and they are capable of meas- 
uring the persecutions which he inflicted on their people by 
their own sufferings endured from time to time in Persia. 
Under these circumstances we can imagine what must have 
been the rejoicings of their ancestors at Susa, if not at Hamadan, 
in Bible times when the appeal of their beautiful heroine 
touched the heart of King Ahasuerus (Xerxes) and caused 
him to spare their people and to hang their enemy on the 
very gallows that had been prepared for Mordecai. 2 

Other stories from the Scriptures, besides these, haunt our 
memory when in Hamadan. Here at Ecbatana, as described 
in the apocryphal book of Tobit, 3 lived Sara, daughter of 
Raguel, who was under the fatal ban of the demon Asmodeus 
until freed by Tobias. 4 If this story be dismissed as too 
legendary, we still may advert to another Scriptural fact, 
known from later history, namely, that Antiochus Epiphanes 
(about B.C. 164) came to Ecbatana after the outrages he com- 
mitted at Persepolis, and was here smitten by the disease that 
brought about his death as a direct visitation of the curse of 

1 See the article by Sidi of Hamadan, Tobit 3. 7 ; 6. 6 ; 7. 1 ; 14. 12, 14. 
in Revue des Ecoles de V Alliance Is- See also Moulton, The Iranian Back- 
raelite, no. 8, pp. 64-68, Paris, 1903. ground of Tobit, in the Expository 
Sidi endeavors to refute the claims Times, 11. 257-260. Again Sir Henry 
against the authenticity made by Israel Rawlinson, JRAS. 10. 136-137, pleads 
Le>i (see reference in preceding note) that Takht-i Suleiman, not Hamadan, 
and points to the miraculous manner fulfils the conditions described in Tobit. 
in which women are freed from barren- 4 The demon Asmodeus is generally 
ness by performing certain rites in the regarded as identical with Aeshma the 
monument. Daeva, in the Avesta, but this is doubted 

2 Esther 7. 10 seq. by some. 



170 HAM AD AN, THE ANCIENT ECBATANA 

God, alluded to in 2 Maccabees 9. 1-3. In addition to this 
I may again recall the Biblical statement to which I have 
already referred, describing how the edict of Cyrus the Great 
in favor of the Jews was found at Hamadan. In connection 
with the Bible, likewise, although not confined to Hamadan, 
I observed a parallel to the custom of cities of refuge ; for, as 
I walked through the streets of the town, I saw several pre- 
cincts marked off by chains over the entrances and gateways 
to indicate that the places were asylums of refuge, like those 
recognized in the Pentateuch. 1 One of these last khdnah, as 
they are called, is the Imam Zadah Yahya, not far from the 
Musallah Hill. 

A visit to Hamadan would be incomplete without a sight of 
the Ganj Namah inscriptions carved by Darius and his son 
Xerxes on one of the rocky peaks of Mount Alvand, southwest 
of the city, on the summer route to Tuisirkan. The distance 
is about an hour's easy ride on horseback, but owing to the 
snow and muddy roads it took me double the time to reach the 
place. The inscriptions are carved in two niches in the face of 
a granite rock, about a hundred feet above a small stream 
which flows past the base of the hill. The situation is a pic- 
turesque one, and the approach, under ordinary circumstances, 
is not difficult, but I had to clamber up the hillside through 
knee-deep snow which extended almost to the base of the inscrip- 
tions and formed an artificial terrace under the lower edge. 

The position of the rock is such that the tablets face directly 
east. The niches are rectangular in shape, measuring about five 
feet by eight and a half, and sunk about a foot deep in the rock. 
The inscription of Darius is in the niche to the left and is 
slightly higher "than the Xerxes tablet. Both are about of the 
same size and proportions and both show signs of weathering in 
places, for the rain and melting snow have proved destructive 
to the stone, despite the fact that the framework of the recess 
serves somewhat to protect it. The Darius inscription has 
1 See, for example, Numbers 35. 6, 11-15 ; Deuteronomy 4. 41-43. 





THE GANJ NAMAH INSCRIPTIONS IN WINTER 



KJI 



THE WORD 'GOD' (BAGA) IN ANCIENT PERSIAN CUNEIFORM LETTERS 



THE GANJ NAM AH INSCRIPTIONS 171 

suffered most. A crack runs through it from top to bottom on 
the left, or southern, side and broadens out as it nears the lower 
ledge till it becomes a fissure almost five inches wide and thus 
destroys several letters. Moss was growing in this cleft when 
I saw it and water was trickling through the groove, tending 
to expand the crack still more. A small crack also mars the 
upper part of the third column of this tablet, and there is a 
defacement in the lower part of the middle column. The 
socket holes noticeable about the framework of both recesses 
are apparently ancient and were probably intended to hold the 
supports for the scaffold while the sculptor was at work. 

The Darius inscription, like its companion-piece, is arranged 
in three columns, written respectively in Old Persian, Susian 
(Neo-Elamitic), and Babylonian. The columns are clearly sepa- 
rated from each other by a narrow space that looks like a per- 
pendicular band, and the lines of the inscription are sharply 
marked off by narrow grooves about four inches (10 cm.) apart. 
Each column contains twenty lines of text. The height of the 
cuneiform letters themselves is between 2-| in. and 2J in. (be- 
tween 6 cm. and 7 cm.). The width of the first column is 44^ 
in. (113 cm.), of the second, 29 in. (77.5 cm.), and of the third 
26| in. (68 cm.). 1 The letters of the inscription may perhaps 
best be likened in shape to horseshoe nails ; the carving is 
in general clear and bold, and the words are perfectly legible, 
with the few exceptions caused by the cracks to which allusion 
has been made. The first word in the Persian column, baga, 
the word for ; God,' looks nearly as in my drawing reproduced 
on the accompanying page. 

More interesting than such technical details and measure- 
ments is a statement regarding the contents of this inscrip- 
tion, which begins Baga vazraka Auramazda, hya imam bumim 
add, hya avam asmdnam add, etc., and may be translated as 
follows : 

1 For assistance in making the meas- inscription, I am indebted to Rev. 
urements while I was examining the N. L. Euwer of Hamadan. 



172 HAMADAN, THE ANCIENT ECBATANA 

t( A great god is Auramazda, who created this earth, who created 
yonder heaven, who created man, who made Peace for man, who made 
Darius king, the one king of many, the one ruler of many. I am 
Darius, the Great King, the King of Kings, King of the countries 
which have many peoples, King of the great earth even to afar, the 
son of Hystaspes, the Achsemenian.' l 

The inscription of Xerxes by the side of this is precisely 
identical with it in contents except that the name of Xerxes 
is substituted for that of Darius throughout the tablet. The 
first, or Persian, column in this table is slightly marred near 
the top, and a crack, beginning in the eighth line of the second 
column and running nearly to the eighteenth line of the third, 
slightly damages several words. In width the first column is 
46 in. (116.5 cm.), the second, 27J in. (70 cm.), the third, 25 
in. (65.5 cm.), and, as in the other inscription, each column has 
twenty lines of text, the letters being nearly of the same size as 
in the Darius inscription. 

In addition to their historic value there is a special interest 
attached to these two tablets of Darius and Xerxes, from the 
standpoint of importance in deciphering cuneiform records. 
They are the two tablets that gave the key to the English 
decipherer, Sir Henry Rawlinson, whose work contributed most 
to placing cuneiform studies in general on a firm foundation. 
In the modern Persian tongue G-anj Ndmah means 4 Treasure- 
Story,' and the natives call the sculptured tablets so because 
they imagine that a secret of some hidden treasure is concealed 
in them and that this will be revealed to the one who shall 
be able to decipher them. This fiction has proved to be a 
fact, though in a different way. The stone has been forced 
to tell its story of hidden riches in the realm of history and 
has handed over the key which the great decipherer used when 

1 Dar. Alv. 1-20. For translitera- 1893 (containing also a bibliography, 

tion of the Persian text of this, see p. 6) ; Spiegel, Die, Altpersischen Keil- 

Weissbach and Bang, Die Altpersischen inschriften, pp. 46, 64, Leipzig, 1881. 

Keilinschriften, pp. 36, 42, Leipzig, See also pp. 175-185, below. 



RECORDS OF DARIUS AND XERXES 173 

he translated the rock records of King Darius at Behistan and 
thus unlocked many treasures of the past. 

Scarcely a hundred paces distant from the tablets of Darius 
and Xerxes, but around the spur to the right of the moun- 
tain, and overlooking the gully through which the stream flows, 
are two small niches. They are blank, without the trace of a 
single letter, but they appear to have been prepared to receive 
short inscriptions. One of these recesses is cut at a consider- 
able slant, shaped like a rhomboid, the other is a narrow paral- 
lelogram carved vertically so that it looks almost like a small 
window in the rock. How these panels were to have been 
filled fancy alone can surmise. 1 

On the road homeward from the Ganj Namah I stopped to 
examine the site of some ruins on a high ridge called that of 
the Nakdrah Khdnah, or ' Band Tower,' about two miles distant 
from the city. The outlines of a structure which the people 
name the Bury, 'Tower,' can be clearly made out in spite of 
its razed condition. Fragments of bricks mark the walls of 
a building which the guide termed the ' Governor's House,' a 
title commonly given by the Persians to the ruins of any large 
edifice that looks as if it might have been used officially. The 
remnants of a circular tank or reservoir (lioz), whose thick 
walls of red brick were set in a very durable mortar, could also 
be traced, and there were evidences, besides, of a terraced 
approach to the brow of the hill, and no doubt the pick and 
spade, if used, would reveal more and tell something perhaps of 
the history of these crumbled remains. 

The view of Hamadan from this height, overlooking the 
plain of many colors, repaid me for the ascent, if nothing else 
did, for it presented the distant city to me in still another light. 
The Musallah, once crowned with walls, stood out clear to the 

1 There is also a cave in the moun- believed to create wind for threshing 

tains in this same vicinity, about two grain (this information I owe to Mr. 

miles south of Hamadan, and likewise H. L. Rabino of Kermanshah), but I 

a ' windstone,' which,when shaken, is did not see either of these. 



174 HAMAD AN, THE ANCIENT ECBATANA 

view, but looked barren and desolate. Fine gardens, which 
afford a delightful resort for an afternoon in the suburbs, 
especially in the vicinity of the Armenian village of Sheverin, 
border the town. The domes of the shrines of the Imdmzadahs, 
or Moslem saints, tell how the newer religion has supplanted 
the old religion of Zoroastrianism, and the minaret of the 
'Friday Mosque,' Masjid-i Jum^ah, points to the fact that 
the older worship has given place to a newer. As I 
rode back through the busy streets, all astir with life and 
activity, and fully alert to the interests of the present, I could 
not help thinking that, despite its three thousand years, Hama- 
dan is still youthful in spirit as in appearance, even though 
reft of the magnificence which once made the city the boasted 
pride of Media. 

' Where's the wisdom-hoary sage 

Shall unriddle us this page ? 

Temples toppled from their base, 

Victor race o'errunning race, 

Yet, within the ancient place, 

Mirth, and love of maid and man, 

Round the walls of Hamadan.' 



CHAPTER XIII 
THE ROCK INSCRIPTIONS OF THE GREAT PERSIAN KINGS 

* Sermons in stones/ 

SHAKSPERE, As You Like It, 2. 1. 17. 

THE Bible refers to 4 the book of chronicles of the kings of 
Media and Persia,' 1 but those documents written on perishable 
parchment were not the only records which the Persian mon- 
archs, Cyrus, Darius, Xerxes, and Art'axerxes caused to be 
made of their deeds and of memorable events in their reigns. 
I refer to the cuneiform records, covering a period of nearly 
two centuries (B.C. 541-340), which the Achsemenian kings 
inscribed upon the living rocks. These documents in stone 
have defied the ravages of time, in part at least, and preserved 
for the present and the future an account of events long past, 
many of which would otherwise have been buried in oblivion. 

By far the most important of these inscriptions is the great 
inscription of Darius, carved far up on the mountain side of the 
Behistan rock. Next in interest and value are those which 
the same monarch caused to be chiselled on the palace walls 
and platform at Persepolis, as well as around his tomb at 
Naksh-i Rustam. The portals and pillared halls of Xerxes 
and Artaxerxes at Persepolis, though in ruins, contain also 
short inscriptions left by these kings, and a sculptured monolith 
near the tomb of Cyrus the Great at Pasargadse was inscribed 
with four words from the lips of the great king himself. Be- 
sides these, there are tablets of Darius and Xerxes on Mount 
Alvand near Hamadan, which I have already described, and 
additional inscriptions of the Achsemenians are found at Susa 

i Esther 10. 2, cf. 6. 1. 
175 



176 ROCK INSCRIPTIONS OF THE GREAT PERSIAN KINGS 

in southwestern Persia, at Kerman in the southeast, at Van in 
Armenia, and even at Suez in Egypt. 1 

We owe to travellers our earliest direct knowledge of the 
Acheemenian inscriptions and of the places where they are 
found. The list begins with the Venetian envoy Josafa 
Barbaro in the fifteenth century and ends with the scholarly 
Niebuhr in the eighteenth, whose more accurate copies of some 
of the inscriptions at Persepolis gave a basis for students to 
work upon. 2 But up to the year 1802 no Daniel had been 
found to interpret the mysterious handwriting on the wall, 
although scholars were generally agreed that the inscriptions 
owed their origin to the Achsemenian kings. It was the 
German schoolmaster and philologist Grotefend who first 
solved the mystery of the cuneiform writing, and to him be- 
longs the honor of being the first to decipher the Old Persian 
inscriptions. Attracted to the subject by reason of his classical 
interests, he devoted himself to the cuneiform problem with 
the enthusiasm that marks a scholar, and the story of how he 
deciphered the characters, letter by letter, reads like a chapter 
in a novel. Placing side by side two of the shorter Persepoli- 
tan tablets, which he assumed to be Achsemenian records, he 
made the shrewd conjecture that one of the words which was 
most often repeated was the name for king and that the king's 
name preceded it. In this manner, by means of a number of 
careful comparisons and scholarly deductions, he was able to 
spell out the name of Darius, of his father Hystaspes, and of 
his son Xerxes. The results of these investigations he laid 

1 With regard to the location of the bach and Bang, Die altpers. Ke.il- 

inscriptions, the history of their de- inschr., Leipzig, 1893 ; Spiegel, Die alt- 

cipherment, editions of texts and pers. Keilinschr., 2d ed., Leipzig, 1881; 

translations, see Rogers, History of Tolman, Old Persian Inscriptions, 

Babylonia and Assyria, 1. 1-83, New New York, 1893. 

York, 1901 ; Booth, Discovery and 2 For a detailed account of the 

Decipherment of the Trilingual In- earlier travellers and investigators of 

scriptions, London, 1902; Weissbach, the cuneiform inscriptions see the books 

Die altpersischen Keilinschriften, in by Rogers and by Booth referred to 

Grundr. iran. Philol. 2. 64-74 ; Weiss- in the preceding note. 




**% 



THE ROCK OF BEHISTAN FROM THE SOUTH 
(From Rawlinson) 




VIEW OF THE BEHISTAN ROCK FROM THE UPPER END 
(Sketched from a photograph by the author) 



GROTEFEND THE FIRST DECIPHERER 177 

before the Academy of Sciences at Gottingen, September 4, 
1802, and thus founded the science of cuneiform decipherment. 
The key to the riddle having at last been discovered, other 
scholars continued the work begun by Grotefend, so that we 
are able to-day to read all the Persian inscriptions and also to 
translate the parallel versions of them in Elamitic and Baby- 
lonian. 1 But foremost among the contributors to the science 
of cuneiform interpretation, whether German, French, or Dan- 
ish, was the noted Englishman Rawlinson, who was a soldier 
as well as a scholar. To Major Henry Creswicke Rawlinson, 
afterwards Sir Henry Rawlinson, Privy Councillor, belongs the 
honor of deciphering the Ganj Namah tablets at Hamadan, and 
the glory of being the first to ascend the Behistan rock and 
copy the inscription of Darius. 2 

The Behistan Mountain, l&ayio-Tavov "Opo?, or Bisitun, as the 
natives call it, had been known from times of antiquity, but no 
one understood the meaning of the inscribed tablets or could 
identify the dozen or more human figures sculptured above 
them on the inaccessible side of the rock. Diodorus Siculus 
in the first century B.C. attributed them to Semiramis. 3 Ya- 
kut, twelve centuries later, has little to say about Behistan, 
except to allude to the equestrian statue at the base, which is 
now known, even in its mutilated condition, to be of Parthian 
origin 4 (about A.D. 50). The first European to call attention 
to Bisitun seems to have been the French traveller Otter, 
about the year 1734 ; Olivier noticed it also some sixty years 

1 It is appropriate to mention the account of his work in Rogers, lifts- 
names of the earlier workers in cunei- tory, pp. 63-73, and Booth, Discovery, 
form philology, such as de Sacy, Saint- pp. 102-114. 

Martin, Rask, Burnouf, Lassen, Beer, 3 See p. 189, below. 

Jacquet, Rich, Westergaard, Holtz- * See p. 209, below. Yakut, pp. 124- 

mann, Oppert, Me"nant, and Spiegel, 125, speaks of the finely carved horse, 

without including names of the present which shows that he gave his attention 

time. only to the Gotarzes sculpture below, 

2 Consult the interesting biography which is now destroyed. He calls the 
by his brother, Canon George Rawl in- horse 'Shabdiz,' another statue of 
son, Memoir of Sir Henry Rawlin- which he describes at Tak-i Bostan, 
son, London, 1898, and compare the cf. p. 224, below. 

N 



178 ROCK INSCRIPTIONS OF THE GREAT PERSIAN KINGS 

later ; Jaubert visited it ; and Gardanne fancied that the sculp- 
tures might represent the twelve apostles. 1 Ker Porter, in 
1818, conjectured that the bas-reliefs were portraits of King 
Shalmaneser and two of his generals, together with the ten cap- 
tive tribes of Israel. 2 But no one had examined the carvings 
in detail or copied the inscriptions below and beside them. 
The danger of climbing the rock proved too great a barrier. 
Kinneir, in 1810, did not attempt the task ; Ker Porter climbed 
halfway up and sketched the sculptures, but did not reach the 
ledge to copy the inscriptions, and in speaking of the danger of 
the ascent he says that 4 at no time can it be attempted without 
great personal risk.' 3 It remained for Rawlinson to accom- 
plish the feat, in the year 1835. 

Rawlinson was at that time a young military officer, twenty- 
five years old, and employed in training native recruits for the 
army of the Shah. While stationed at Hamadan he had learned 
to decipher the cuneiform characters, 4 and he was now given an 
excellent opportunity to examine the Behistan rock by receiv- 
ing an appointment from the Shah, in 1835, which transferred 
him to Kermanshah, some twenty miles distant from the in- 
scription. He made the best of his opportunity, and at risk of 
life and limb made several ascents of the rock in the next two 
years, 1835-1837, and was able to transcribe the first column 
of the Persian text with remarkable exactness and prepare it 
for publication. Nearly ten years later, in 1844, after active 
service in the Afghan war, he transcribed the rest and took a 
copy also of the Elamitic version (variously called Median, 
Scythian, Neo-Susian). He supplemented this, in 1847, by a 
revision of the entire text and a reproduction of the Babylonian 

1 See Ker Porter, Travels, 2. 154, before his time and that scholars were 
and Booth, Discovery, pp. 82, 105. at work on the texts, but it is clear 

2 Ker Porter, Travels, 2. 159-162. from his later correspondence with 
8 Ker Porter, Travels, 2. 158. European savants how much he had 
4 See p. 172, above. While working been able to accomplish independently. 

on the Ganj Namah inscription Raw- See his account in The Athenaeum, no. 
linson knew that the key to the 2976, p. 693, Nov. 8, 1884. 
cuneiform had been discovered long 



RAWLINSON'S GREAT ACHIEVEMENT 



179 



version, which he secured through the help of a wild Kurdish 
lad, who performed the perilous feat of taking a paper squeeze 
of that almost inaccessible inscription. 1 

This is not the place to discuss the great importance of 
Rawlinson's work and the value of later contributions based 
upon it, as they are well known to scholars ; 2 but Rawlinson's 
copy was made more than half a century ago, and there had 
been no opportunity to test its accuracy, as he was the only one 



1 Rawlinson's account of the lad's 
dangerous climb may be found in 
Archceologia, 34 (1850), pp. 73-75, re- 
printed in George Rawlinson, Memoir, 
p. 156, n. 1. Rawlinson's own devo- 
tion to the work on the Behistan rock 
may be judged from what he wrote more 
than forty years later in The Athe- 
nceum, no. 2976, p. 593, Nov. 8, 1884, 
from which I make a selection: ' Dur- 
ing my service as a military officer in 
Persia, from 1833 to 1839, my visits to 
the rock of Behistun were few and 
hurried. On these occasions I worked 
hard, but the difficulties were so great 
that I had only succeeded in copying 
one half of the Persian text of the in- 
scription (the Median and Babylonian 
texts being entirely untouched) when 
I was compelled to leave the country 
in order to take part in the first Afghan 
war. At the close of that war in 1843 I 
was offered, as a reward for my services, 
the highest political employment and 
an assured career in India ; but I had 
not forgotten Behistun. It had become 
the ambition of my Iif6 to carry on my 
cuneiform researches, and especially 
to work out the Babylonian puzzle ; 
and accordingly, to the astonishment 
of my friends, I deliberately declined 
the brilliant prospect opened out to me 
in India, and elected to return to what 
was called " exile " at Baghdad, where 
for twelve weary years broken by 



only one brief visit to England I re- 
sided, in an exhausting climate, cut off 
from all society, sparingly supplied 
with the comforts of civilization, and, 
in fact, doing penance in order to attain 
a great literary object. During this 
period of probation, on two occasions 
in 1844 and 18471 again visited 
the rock of Behistun, riding 1000 miles 
for the purpose and disbursing above 
1000 I. from my own funds for the ex- 
penses of the expeditions. I will not 
say much as to the danger or difficulty 
of ascending the rock and reaching the 
upper part of the sculptures, which are 
some 500 feet above the plain. I did 
not think much at the time of the risk 
to life and limb, but it must be remem- 
bered that Messrs. Coste and Flandin, 
having been deputed to the spot by the 
French Government with express in- 
structions to copy the inscriptions, re- 
turned re infectd,, declaring the sculp- 
tures to be absolutely inaccessible ; 
and I may further add that although 
there is still something to be copied 
and much to be verified, I have never 
heard but of one traveller accomplish- 
ing the ascent since the period of my 
last visit. 1 

2 Rawlinson's epoch-making work 
was published in the Journal of the 
Eoyal Asiatic Society, 10. 1-349, Lon- 
don, 1847. 



180 ROCK INSCRIPTIONS OF THE GREAT PERSIAN KINGS 

who had studied the rock itself. Accordingly, one purpose of 
my journey was to make the ascent, if possible, and examine 
some of the mooted passages in the cuneiform text. How far 
I was able to accomplish this, in the limited time at my disposal, 
will be recorded in the next chapter. 

Before giving an account of the contents of the inscriptions, 
it may be well to give some idea of the bas-reliefs that are 
carved above them on a surface over twenty feet in length and 
more than ten feet in height, and to tell whom the sculptures 
are intended to represent. 

The king, who is the principal figure in the group, is Darius 
himself, and the image is majestic in its bearing and carved in 
bold relief. In his left hand Darius holds a bow, and he 
raises his right hand as he pronounces sentence of doom upon 
nine captives standing before him, each with the hands tied 
behind the back and a rope about the neck. Above the head of 
each of the captives, but below the prostrate figure and on the 
skirt of the tunic of the third prisoner, is the name of the rebel 
king whose effigy is given, and in each case the nature of the 
rebellion and the place where it started is recorded in a short 
tablet and a word is added to the effect that the pretender 
4 lied ' in making his claims to the throne. 1 

The names of some of these rebels, like Nadintabaira, or 
Nadintu-Bel of Babylon, we know also from other sources, and 
the last in the line is ' Skunka the Scythian,' who is marked by 
his high pointed cap. Beneath the feet of the king lies a tenth 
foe, imploring mercy with upstretched hands, but trampled 
upon by the stern monarch. This fallen enemy is Gaumata 
the Magian, otherwise known as the False Smerdis, who 
usurped the crown upon the death of Cambyses, under pretense 
of being the king's brother, but was unmasked and slain by 
Darius and his six trusty followers, as described in the inscrip- 
tion itself and narrated also by Herodotus. 

1 In the main body of the inscrip- of each revolt, and the battles which 
tion below the bas-reliefs an account followed, is given. 



V OF THE 

UNIVERSITY 





THE KEBMAN INSCRIPTION OF KING DARIUS 
1. The Persian Text 



THE GRAND INSCRIPTION OF BEHISTAN 181 

Directly back of the king stands his quiver-bearer and cham- 
berlain, Aspachanah, and behind the latter is the king's spear- 
bearer, Gaubruva, if we may assign to these the names attached 
to the similar figures at Naksh-i Rustam names known also 
from the Greek sources as Gobryas and Aspathines. 1 Above 
the head of the king there floats a winged figure of the god 
Auramazda, who presents to him a ring, the symbol of sover- 
eignty, and adds his blessing in sanction of the king's deeds. 2 
This figure, like the rest of the sculptures, shows the influence 
of Assyro-Babylonian art, and all the figures are very robust 
and stocky in build, like some of those at Persepolis. The 
king is represented as heroic in size in contrast to the captives, 
who appear dwarfed beside him. 

The position of the inscriptions with reference to the bas- 
reliefs may next be noted. The Persian tablets are directly 
below the sculptured group ; the Elamitic version is carved 
around the ledge on the lower left-hand side and is very diffi- 
cult to reach ; the Babylonian projects over this, on the face of 
the scarped rock ; and supplementary inscriptions and trans- 
lations are carved around the bas-reliefs and on tablets to the 
right. After giving these details we may turn to the inscrip- 
tions themselves and their contents. 

The five columns of the Old Persian give a brief history of 
the main events that occurred in the early part of the reign 
of Darius. They are carved in about four hundred lines of 
beautiful cuneiform writing, on polished tablets, and are sup- 
plemented by versions in Elamitic and Babylonian, which 
comprise together as many lines more. Darius tells in brief 
form how he became king 4 by the grace of Auramazda ' 
vashnd Auramazddha and by his own prowess, and he re- 
counts the battles he fought and the victories he won and 
how many revolts he had to crush in organizing and adminis- 

1 For references, see Justi, Ira- also Andreas, in Verhandl. 13. Internat. 
nisches Namenbuch, pp. 46, 111 ; Bar- Orientalisten-Kongr.ip. 97, Leiden, 1904. 
tholomae, Air. Wb. pp. 217, 482. Cf. 2 See Grundr. iran. Philol. 2. 631. 



182 ROCK INSCRIPTIONS OF THE GREAT PERSIAN KINGS 

tering his great empire. The language is nearly allied to that 
of the A vesta, and the style, which shows the influence of the 
Assyro-Babylonian inscriptions, 1 is marked by a dignity and 
simplicity that is suited to such a record, despite the inevitable 
baldness of an official document and the tendency to repetition 
which characterizes an Oriental communication. 'I am Daraya- 
vaush (Darius), the Great King, the King of Kings, the King 
of Nations, the son of Vishtaspa, the Achsemenian ' such are 
the opening words; and after tracing his right to the throne 
through a double line of descent and enumerating his tribu- 
tary countries, he gives a notable account of the usurpation 
and overthrow of Gaumata the Magian, already referred to. 
Throughout the entire five columns each paragraph that deals 
with a new subject is introduced by 'Thus saith Darius the 
King ' thdtiy Ddrayavaush khshdyathiya which lends a cer- 
tain formal dignity to the style ; and a religious tone is im- 
parted to the edict by the fervor with which Darius again and 
again attributes his successes to Auramazda. This is particu- 
larly noticeable in the fourth column, where the style rises to 
some degree of literary merit. To illustrate what I mean I 
shall translate a portion of that column, as. I devoted especial 
attention to it during the time which I spent upon the rock. 
I shall follow the original closely in my rendering, so as to 
convey an idea of the inscriptional style, making only trifling 
modifications for the sake of greater intelligibility, and one or 
two slight omissions for the sake of brevity. 

Bh. 4. 33-36. 'Thus saith Darius the King: Those countries 
which became rebellious, the Lie made them rebellious, so that they 
deceived the people. 2 Auramazda deli vered them into my hand.' 

1 See Gray, Stylistic Parallels be- 2 The word drauga, 'Lie, False- 

tween the Assyro-Bab. and OP. Inscr. hood, Deceit,' is personified in the 

in Am. Journ. Semit. Lang. 17. 151- inscriptions and used almost as an 

159, and compare my article, Persian equivalent of Satan, Fiend ; see my 

Literature, in Progress, 2. 35-55, Chi- article in JAOS. 21. 170. 
cago, 1896. 



OF THE" 

UNIVERSITY 

OF 





THE KEEMAN INSCRIPTION OF KING DARIUS 
1. The Persian Text 



CONTENTS OF THE BEHISTAN INSCRIPTIONS 183 

36-40. ' Thus saith Darius the King : Thou who shalt be king 
hereafter, be constantly on thy guard against the Lie. The man 
who is a liar, punish him well with punishment, if thou thinkest 
" my country must be firmly established." ' 

40-45. ' Thus saith Darius the King : That which I have done 
I have done altogether by the grace of Auramazda. Thou who shalt 
hereafter read this inscription, let that which hath been done by me 
appear to thee true ; hold it not for a lie. Thus saith Darius the 
King : May Auramazda be witness that it is true, not false ; I did it 
altogether.' 

45-50. 'Thus saith Darius the King: By the grace of Aura- 
mazda there is much else, besides, done by me, which is not written 
in this inscription ; on this account is it not written, lest that which 
I have done may seem exaggerated to him who shall hereafter read 
this inscription, and may not appear to him true and may seem to be 
a lie. . . . J1 

52-59. 'Thus saith Darius the King: Let that which I have 
done appear unto thee true, as it is ; therefore conceal it not. If thou 
shalt not conceal this edict, but shalt publish it to the people, may 
Auramazda be a friend unto thee, and may thy seed be multiplied, 
and mayest thou live long. Thus saith Darius the King : If thou 
shalt conceal this edict and shalt not publish it to the people, may 
Auramazda be thy slayer, and may thy seed be cut off.' 

59-64. ' Thus saith Darius the King : That which I have done I 
have done altogether by the grace of Auramazda. Auramazda, and the 
other gods that be, brought aid unto me. For this reason did Aura- 
mazda, and the other gods that be, bring aid unto me, because I was 
not hostile, nor a liar, nor a wrong-doer, neither I nor my family, 
but according to Rectitude have I ruled.' 2 

The inscriptions of Xerxes and of the later Artaxerxes are 
hardly more than reproductions of the minor tablets of Darius, 
formulaic in their contents and mechanical in their structure ; 
but they have a historical and philological value, and are inter- 
esting also from the standpoint of religion, because the inscrip- 
tions of Artaxerxes II and III recognize Mithra and Anahita as 
divinities by the side of Auramazda, whereas they are not men- 

1 The frank simplicity of this state- 2 For the difficulties in the last two 
ment is delightful. lines, see pp. 203-207, below. 



184 ROCK INSCRIPTIONS OF THE GREAT PERSIAN KINGS 

tioned in the inscriptions of Darius, although perhaps included 
in 4 the other gods that be' (aniyd bagdha tyaiy hantiy). Au- 
ramazda, however, is still the supreme head of the world 
according to all the tablets : 4 A great god is Auramazda, the 
greatest of the gods, he who created this earth, who created 
yonder heaven, who created man, who created Peace for man, 
who made Xerxes (or Artaxerxes) king.' 1 But in general the 
ring of the metal seems less true, and in these later inscriptions 
the language, like the style, shows signs of decadence. 

Before I close this chapter I wish to call attention to one 
small inscription of Darius, previously known but not easily 
accessible in the cuneiform characters. It is preserved at the 
village of Maghan, near Kerman, in the shrine of Nimat-ullah 
Vali, founder of the order of Nimat-ullah dervishes, but its 
previous history is unknown. The characters are carved on 
three faces of a small tetragonal pyramid of dark stone, which 
is about 4 inches high and 3J inches square at the base, 2 and 
therefore a little larger than the photographs that I reproduce. 
The photographs were kindly sent me by the late J. C. Van 
Roosbroeck, Director of Persian Customs and Post, and for- 
warded to him by the courtesy of Mr. A. Miller, Russian Consul 
at Kerman. The contents of the eight lines are familiar from 
other inscriptions of Darius, and they read : 4 1 (am) Darius, 
the Great King, the King of Kings, the King of Nations, King 
of this Earth, son of Vishtaspa, the Achsemenian.' The same 
lines are repeated in an Elamitic and a Babylonian version. 3 

It may be said in conclusion that, viewed as a whole, the 
story of the deciphering of the cuneiform records is one of the 
most instructive chapters in the history of philological research, 
and the achievements of Grotefend and his successors are to be 

1 With regard to the expression Bang, Altpers. Keilinschr. pp. 7, 38, 

Peace (Tkiyati) cf. Isaiah 45. 7, and see and Gobineau, Traite, 1. 323 seq.; also 

JAOS. 21. 166. Bartold, Historico-geographical Ac- 

2 Approximately 10 cm. by 9 cm. count of Iran, pp. 94-95 (in Russian), 

8 For the text of this inscription St. Petersburg, 1903. 
in transliteration, see Weissbach and 




THE KERMAN INSCRIPTION OF KING DARIUS 
3. The Babylonian Text 



OTHER CUNEIFORM RECORDS 185 

reckoned among the memorable accomplishments of the nine- 
teenth century. Let us hope that those blessings may always 
come true which Darius invokes in the inscription itself upon 
those who preserve the inscription and make it known to the 
people, for Auramazda shall be their friend. 



CHAPTER XIV 

THE GREAT BEHISTAN ROCK AND AN ASCENT TO READ 
THE CUNEIFORM RECORDS OF DARIUS 1 

' I wol yow all thys shap devyse 
and site, and all the wyse 
How I gan to this place aproche 
That stood upon so high a roche 
Hyer stant there noon in Spayne, 
But up I clomb with alle payne.' 

CHAUCER, House of Fame, 3. 23-28. 

EASTER MONDAY, April 13, 1903, will remain for me a mem- 
orable date in the calendar, for on that morning, after four days 
on horseback from Hamadan, I caught my first glimpse of the 
mountain of Behistan and the great inscription of Darius. For 
miles before one reaches it the huge mass of rock is constantly 
in sight, lifting its giant head seventeen hundred feet above the 
plain; and several times in the distance my eager eyes were mis- 
taken in fancying I could see from afar the smoothed surface 
where the Great King's edict is inscribed. This was an error, for 
in approaching by the Hamadan road one must round the north- 
east corner of the mountain before the inscription can be seen. 
It was shortly before noon, or, to be more accurate, 11.25 A.M., 
when my caravan halted at the base of Bisitun, as the Persians 
call it, and far above I could see the inscription and the sculp- 
tured figures which the natives term 4 the Nine Dervishes.' 

1 Reprinted with some additions and harahya ; the notice of the Gotarzes 
minor corrections from my report in sculpture ; and the account of the 
JAOS. 24. 77-95. The additions are : monolith at the close of the chapter. 
1. 61, patiyavahyaiy ; 2. 61, Qauravd- 

186 



FIRST VIEW OF THE GREAT ROCK 187 

With all I had read about Behistan, with all I had heard about 
it, and with all I had thought about it beforehand, I had not 
the faintest conception of the Gibraltar-like impressiveness of 
this rugged crag until I came into its Titan presence and felt 
the grandeur of its sombre shadow and towering frame. Snow 
and clouds capped its peaks at the time, and birds innumerable 
were soaring around it aloft or hovering near the place where 
the inscriptions are hewn into the rock. There as I looked 
upward, I could see, more than three hundred feet above the 
ground, the bas-relief of the great king, Darius. Prone at 
his feet lay Gaumata, the Magian usurper, who had seized the 
throne on the death of Cambyses. In front of Darius stood 
the row of captive kings, and above the head of each I could 
discern a faint trace of the tablet with the ' lie ' which each 
had uttered in his false claims to the throne, although the 
letters were not legible at such a distance. My memory 
recalled the story of each of these rebel lords, and I could 
picture the torture and agonized death that each suffered at 
the hands of the king. 

From the descriptions I had read, or perhaps from the mental 
picture I had previously formed of the scene, I had always 
fancied that the inscriptions and the sculptures were carved 
nearer the middle of the mountain, whose general contour on 
this side runs from northeast to southwest. Not so. They 
are cut high up in the side of a steep gorge or craggy gully 
that makes a deep gash in the face of the rock and extends 
three hundred feet downward to the plain beneath. But 
before proceeding further with the description, it may be well 
to turn to the middle part of the mountain front itself and 
examine its appearance. 

As one faces the great Behistan rock, the striking feature 
that catches the eye is a huge space carved near the middle 
of the base, but left entirely bare of an inscription. Even Ker 
Porter in his description seems to have given less attention 
than it deserves to this magnificent tabula rasa, the more 



188 THE GREAT BEHISTAN ROCK 

conspicuous because of its vacant, wall-like stare. 1 It must have 
been prepared with an especial design of recording some historic 
event, as I felt certain after devoting part of an afternoon to a 
study of it. A space of nearly five hundred feet in length 
I paced it off and over a hundred feet in height has been 
cut out of the mountain front to form a rocky canvas for com- 
memorating some record of importance. The idea that it is 
due to mere quarrying vanishes at once when one studies the 
appearance of it and observes the evident design. Two rocky 
ledges, one somewhat higher than the other, are cut on either 
side to furnish a means of nearer access to the mammoth screen, 
while the overhanging canopy of rock forms a framework above, 
and a terrace of earth and stones offers an approach to the place 
from below. Such is the general scheme of arrangement. 2 

The question naturally arises, and is always asked by those 
who have seen the great blank space: 4 When and by whom was 
it cut, what was its purpose, and why is it without a trace of 
the cuneiform chronicler's chisel ? ' To this inquiry the natives 
respond by saying 4 it is the work of Farhad.' The sentiment 
of such an explanation will appeal to every reader of Nizami's 
romantic epopee ; he will recall the tragic story of the enam- 
ored sculptor and the lovely Shirin, and he will trace in fancy 
the marks of the ambitious wooer's steel or hear the ring of the 
inallet as the rock yielded to his herculean blows. 3 But the 

1 See Ker Porter, Travels, 2. 149- cit. p. 287). Mr. E. L. Mitford (From 
162. England to Ceylon Forty Years Ago, 

2 When I gained access to my libra- London, 1884) believed that it was de- 
ry I found that M. de Morgan (Mis- signed ' apparently for the back wall of 
sion Scientifique, 4. 286-289) has given some extensive building,' and he adds 
an elaborate description of the probable that * the only sculpture on the scarp 
manner in which the vast surface was was a single female mask. 1 If this 
prepared by the stone-cutters, and he still exists, I failed to see it, and I am 
shows how the markings on the stones inclined to think that the signs in the 
which have been thrown down may have fallen stones which are scattered 
been made. He is of the opinion that about are mason's marks rather than 
the surface was prepared to receive an characters of an alphabet. But I may 
inscription, * qui, peut-e"tre, devait rela- be wrong. 

ter tous les faits de 1'histoire perse ' (op. 8 See p. 226, below. 



A HUGE SPACE LEFT BLANK 189 

classicist at the same moment will remember a passage in Dio- 
dorus Siculus, which tells how Semiramis visited c Bagistanon,' 
encamped near by, built a 4 paradise ' on the spot, and com- 
memorated the occasion by an inscription on the mountain. 
The quotation from this Greek author, who lived in the first 
century before the Christian era, is worth repeating for the 
sake of comparison : 

< When Semiramis had brought to an end the works upon which 
she was engaged, she set out for Media with a large military force 
and, halting near the mountain called Bagistan, pitched her camp 
there. She made a park, twelve furlongs in circumference, in the 
plain, which has a great fountain that waters all the cultivated area 
round about. The mountain of Bagistan is sacred to Zeus, and on 
the side toward the garden it has steep rocks extending upward to 
the height of seventeen furlongs. On the lower part of this she 
caused her own image to be carved, with a hundred lance-bearers 
standing round about her. She inscribed likewise in Syriac charac- 
ters (Svptbis ypa/A/xao-i) on the rock, that " Semiramis had ascended 
from the plain to the top of the height by laying, one upon the other, 
the packs of the beasts of burden that followed her." ' 1 

That we have in this passage a direct allusion to our rock is 
undoubted. The only question is whether the story which 
Diodorus gives is to be applied to the unlettered space or to the 
familiar sculptures and inscriptions of Darius. The difficulty 
with the former application is the fact that a careful examina- 
tion of the huge central table does not reveal the slightest trace 
of its ever having been inscribed. I studied it with great atten- 
tion, having in mind the Diodorus passage, and I asked also the 
judgment of my native servant, who was very intelligent in 
such matters ; but I could not convince myself that this portion 
of the rock had ever been engraved, or that an inscription had 

1 Diodorus Siculus, Hist. Lib. 2. 13, Five Great Monarchies, 3. 31, n. 18. 

tr. Booth, 1. 110. The statement Yakut, p. 125, says that Behistan is 

'seventeen furlongs, or stadia,' i.e. so high that 'its summit cannot be 

above 10,000 feet, is greatly exagger- reached,' which is also exaggerated, 

ated, as is noticed also by G. Rawlinson, although the peak is very lofty. 



190 THE GREAT BE HIS TAN ROCK 

been obliterated. If one were inclined to theorize and to build 
up a fanciful hypothesis on flimsy foundations, it would be easy 
to suggest that King Darius, after completing the well-known 
record and sculptures, had directed the present vacant space to 
be prepared for a memorial of his later deeds, especially the 
campaign against Greece. The misfortunes at Athos and Mara- 
thon, the uprising in Egypt, and the hand of death, frustrated 
his plan, changed the course of history, and left the blank page 
on the rock to bear witness e silentio to the triumph of Hellas 
and the beginning of the downfall of Iran. But this is mere 
guesswork, idle fantasy, especially when one asks why Darius 
should have reserved the central and best position on the moun- 
tain for the last. All that we can say is that the general appear- 
ance of the place and the nature of its surroundings gives the 
impression of great antiquity. So much for the blank and 
unfinished wall-space. We may now turn to the well-known 
tablet and sculptures that form the special subject of discussion. 
As stated before, the great record of Darius is situated far- 
ther to the northeast, some four or five hundred paces removed 
from the central point. As one stands beneath and looks 
three hundred feet upward within the rocky ravine, the general 
outline of the inscription and the figures of Darius, the two 
vizirs, and the ten captive kings, come clearly into view. It is 
easy to understand why the natives regard the latter as 4 the 
Nine Dervishes,' because the prostrate figure of Gaumata, with 
his upstretched hands, is not so easy to distinguish in the dis- 
tance. As to Skunka with his high Scythian cap, I am inclined 
to agree with the view that his figure was added some time 
after the others were carved. On viewing the smoothed spaces 
where the inscriptions are cut, I could make out, as I knew it 
beforehand, the general arrangement of the Old Persian in the 
centre below the sculptures, the Neo-Elamitic to the left of the 
ledge, and the Babylonian above this and also above on the right. 
The familiar broad bands that indicated by their peculiar 
grayish color where the water had streamed down and washed 



A DANGEROUS CLIMB 191 

away portions of the inscription, were all too plain. Even at 
the moment, water was oozing out from the upper part of one 
of the tablets and trickling over its face. It was evident at a 
glance that a telescope would be of no service in copying the 
lower part of the Old Persian text, because the projecting ledge 
cut off a portion of the inscription from below. It was neces- 
sary to get nearer. Climbing past huge boulders and fragments 
of fallen crags, which make the ascent of the gorge not easy, it 
was possible to get closer to where the tablets and sculptures 
are. The precipitous sides of the gorge form an angle ; the 
Darius record is on the side that faces almost directly toward 
the east. The opposite wall or other face of the shaft-like 
ravine is so steep and rugged as to defy the climber's attempts 
to ascend it for the purpose of photographing the inscriptions 
from their own level. The natives assert that it is practically 
impossible to mount that side of the rocky couloir. The ques- 
tion now arose, how best to ascend to the inscriptions. 

Having heard from a Persian friend that it would probably 
be best to be let down from above, I had previously studied 
some of the methods employed by the bird-nesters in the 
Hebrides in being lowered by ropes over craggy cliffs. A brief 
examination of the situation, however, showed that the only 
feasible approach was by climbing and being drawn up by cords. 
In less than an hour the preparations for the task were begun. 
Meshad Ali, the owner of the caravansarai nearest to the rock, 
found five men who were ready to undertake the ascent. A 
sixth, Kuli, the guide and best of them all, was added later ; 
and the procession with ropes and a ladder was soon under way 
toward the beetling precipice. Whatever may be said against 
the ladder, which proved of little use, nothing can be main- 
tained against the Persian goat-hair ropes, for their quality is 
excellent. The cords that bound the luggage on the caravan 
pack-horse, supplemented by ropes furnished by the Bisitun 
guides, and firmly fastened about my chest with knots that only 
a Persian knows how to tie, were a precaution against the 



192 THE GREAT BEHISTAN ROCK 

danger of slipping and gave confidence for the climb. The 
stout protest of the guides against my riding-boots was well 
founded, as the risks of the first day proved ; but a happy sub- 
stitute for these was later found in the native glvahs, resembling 
rough tennis shoes, which were loaned by one of the Per- 
sian bystanders and firmly sewed upon the feet with a heavy 
pack-thread needle. All was then ready. The exciting task 
began. 

The ascent of the first huge fissure in the side of the couloir, 
the clamber with torn hands and clothes along the brink of a 
precipitous crag, the tugging ropes that helped up the steep 
incline of the second rock, the scramble past the thorn bush 
that barred the way farther up, and the final tug and spring 
that brought to the edge of the ledge, together with khaill khiib, 
4 very good,' and the encouraging word of the guides, 4 no fear 
now, the danger is over' will not readily be forgotten. Only 
when one has stood on the narrow ledge by the side of the in- 
scriptions and looked out over the magnificent plain far beneath, 
and listened to the dull murmur of the stream below, as it bursts 
from the mountain's base, does one know how to appreciate Raw- 
linson's work. It may interest others, as it did me, to learn that 
he has carved his name in the stone, a few inches below the very 
inscriptions which he first made known to the modern world. 
This he was entitled to do, and one is almost inclined to append 
after his simple 4 H. C. Rawlinson, 1844' the words of ancient 
India's homage namo namah. 

In the words of Rawlinson, ' the climbing of the rock to arrive 
at the inscriptions, if not positively dangerous, is a feat at any 
rate which an antiquary alone could be expected to undertake.' 
On the first day it took a while to get somewhat used to the 
giddy height, so I devoted my attention to examining the gen- 
eral condition of the rock, making notes, observing the sculp- 
tures, which one can study better, however, from below than from 
the ledge, and to getting the size of the cuneiform letters and 
of the tablets themselves. The four columns of the Old Persian 




THE BEHISTAN SCULPTURES AND INSCRIPTIONS 
(From Rawlinson) 




KULI, THE HEAD GUIDE IN THE ASCENT 



REACHING THE LOFTY LEDGE 193 

record are each about six feet broad. The exact measure- 
ments in meters, if one cares to have them, are : 1st col. = 
1.90; 2dcol. = 1.94; 3d col. = 1.95 (approx.) ; 4th col. = 1.94 
(approx.). The 5th column I did not measure, owing to the 
difficulty of access to it. The place occupied by the Neo- 
Elamitic (Scythian or Median) inscription is around a crag to 
the left of the Old Persian, as one faces the inscription, and for 
taking me there my most reliable guide wished to quintuple 
his price, while the Babylonian tablet on the overhanging ledge 
above to the left and to the right is the most inaccessible of 
all, as Rawlinson himself discovered, when his guides failed 
him and he found only one Kurdish shepherd lad who would 
venture to undertake to reach it and accomplished with diffi- 
culty the risky task of taking the squeezes of that inscription. 1 

On looking at the mass of scarped rock one wonders how 
the daring boy ever accomplished the perilous feat. Perhaps 
he still lives and can tell, but, as regards Rawlinson, I could 
not find the slightest recollection of him among the inhabitants 
of whom I inquired ; although I learned afterwards that his 
special Persian guide died a few years ago at Hamadan. 

A study of the Old Persian tablets soon revealed the fact 
that the inscription has suffered much since the days of Rawlin- 
son. Mention has already been made of the water that was 
oozing from the upper part of the inscription when first I saw 
it, so that it was wet in places for the space of several feet. 
Some photographs, which I succeeded in taking on the second 
day upon the ledge, make clearer what we have lost and are 
losing, and I fear that other and fresher proofs of this will be 
found when the rock is examined with more detail than was 
possible in my short week's stay. But I wish to call attention 
to one point. I found that after the eye had become accus- 
tomed and had some practice, it was possible to restore lost 
letters and words by a careful examination of the indentures 

1 See p. 179, above, and cf . H. C. Raw- Rawlinson, Memoir of Sir Henry Eaw- 
linson, Archceologia, 34. 73-75, and G. linson, pp. 156-157, London, 1898. 



194 THE GREAT BE HIS TAN ROCK 

which the heavy stroke of the engraver's chisel had left in carv- 
ing the character. The head of the nail-shaped letters (for the 
Behistan letters look perhaps more like horse-shoe nails than like 
wedges) can still be discerned as a dot or hole in the washed-away 
stone ; and a knowledge of the cuneiform writing enables one to 
combine these indentures into skeleton letters that often remove 
all doubt as to the true reading. I understood how Rawlinson 
must sometimes have done this, and more easily, because the 
stone had suffered less in his time, I believe, than to-day. 

In contrast to the disintegrated parts stand those portions 
of the inscribed tablets where the flow of the water has not 
mutilated and defaced the rock. Here, instead of the peculiar 
dull steel-gray bands, we have the beautiful brown color of 
the inscription as perfect as when the stone-cutter of Darius 
laid his mallet aside. No granite tablet in Central Park or 
Trafalgar Square could be more perfect. It was interesting, 
moreover, to compare the style of the Behistan characters with 
the somewhat larger letters of the Ganj Namah inscriptions 
(Dar. Alv. and Xerx. Alv.), which I had been examining at 
Hamadan the week before. On Alvand the space between 
the lines is 4 inches and each letter averages nearly 3 inches 
in height. At Behistan, where economy of space was neces- 
sary owing to the length of the inscription, the sharply drawn 
lines are about If inches (42 mm.) apart, and the clear-cut letters 
each approximately 1^ inches (32 mm.) high. The brown 
shellac or varnish makes them stand out in bolder relief and 
gives a fine finish to the whole, although I could see no traces 
of the 'flakes' of the cement, which Rawlinson speaks of as 
having sifted down upon the narrow ledge ; nor, again, did I 
observe any evidence to show that letters had been preserved 
by reason of this shellac withstanding the water when the rock 
itself had disintegrated beneath its surface. 1 But this may be 
still another proof that the rock has suffered since Rawlinson's 
time, and it is to be hoped that M. de Morgan will make casts 
1 So Rawlinson in JRAS. 10. 193. 



COPYING AND PHOTOGRAPHING 195 

of the entire inscription, as I learned in Persia it is the inten- 
tion of his Mission to do. 

My attempt to take squeezes of certain words was a failure ; 
this was due partly, among other causes, to the wind that 
prevailed during the four days when I was up on the ledge, 
and was the stronger owing to the height and the peculiar 
formation of the rocky cut. This made me wish for more 
time, in order to wait for better conditions. Owing to the 
physical strain of the ascent, for it requires some athletic 
prowess, and owing to the exciting interest of the work, which 
is somewhat of a tax on the nerves, the element of time is 
necessary for accomplishing what one would wish to do. To 
this I may add that money is likewise an indispensable factor in 
the equation. But above all one must not be hurried. On the 
last day of my stay, for example, after I had finished all I could 
reach or clearly see, I begged the guides to let me use the 
ladder in order to examine some of the less certain readings in 
the upper part of the inscription. This they stoutly refused to 
do on account of the extreme danger from the high wind blow- 
ing at the time. And that afternoon I was obliged to start 
back to Hamadan. But although some points like these had to 
be left, I was glad to find I had been able to examine most of 
the doubtful passages and to prove in general the wonderful 
accuracy of Rawlinson's transcript. To this I shall revert also 
below when I speak of the two or three photographs I took, the 
first, I believe, ever taken on the ledge ; they were 'snapped' 
as I leaned out over the precipice, held by the guides, while 
focusing the camera and hastily taking the picture. Most of 
my time, however, was spent in copying, collating, or verifying 
the readings on the rock itself without resorting to my photo- 
graphic apparatus. I may add, regarding the means of ascent 
in ancient times, that there is not the slightest trace of any- 
thing of the kind to-day. If ever there was any, it cannot 
have been of a permanent material. 

Regarding the descent I may say that after I had been for 



196 THE GREAT BEHISTAN ROCK 

hours on the cramped and narrow ledge, the going down seemed 
much more difficult than the ascent, and it was a joy each time 
to hear my faithful Persian servant, Safar, who had remained 
below, call out, 4 Now you are safe,' when I passed the last 
dangerous place and could jump to the ground. The unloosen- 
ing of the tight-bound ropes quickly followed with his aid. 
All had gone well. 

So much for the incidental side. I now turn to the far more 
interesting and important matter of what I was able to note, 
verify, or restore. 

Bh. 1. 47, dy a 8 a td: the reading of each letter is quite clear. 

Bh. 1. 51, paranam : reading absolutely certain. 

Bh. 1. 55, patiydvahyaiy : so my photograph (%), cf. also 
Skt. avasya, 4 flee for refuge.' 

Bh. 1. 65, vWWis". a : barring this troublesome word, the 
entire line from -caris gaiOdmcd mdniyamca to tyddis gaumdta 
hya is quite as given in Spiegel and in Weissbach and Bang, 
except that the g and u of Gaumata's name are defaced, and 
the y of hya is illegible, owing to the weathering of the rock. 
The question arises with regard to the much-discussed word 
beginning with v i in the middle of the line. The latter part of 
the word is mutilated, but my memoranda show that we must 
accept two letters after -bis. The latter I have marked as 
4 apparently a,' the former I noted at first as 4 illegible,' but 
added afterwards, 4 probably right as .' This shows the bearing 
of Rawlinson's 4 extremely doubtful ' as regards the c at least. 
A photograph which I took of the first part of the word 
vWb . d is interesting as showing that there is no i inserted 
either before or after the 0. This is a matter of importance for 
future reference. I am not unmindful of the various dis- 
cussions of this word in the journals the most recent being 
that of Gray, JAGS. 23. 56-60. Regarding abi- or abd- 
of abicari$, I unfortunately find on returning to America that I 
had made no special memorandum, but my inference from the 
absence of a note is that the text stands as first given by Raw- 




PHOTOGRAPH OF A SECTION OF THE INSCRIPTION OF DARIUS, BEHISTAN 1. 61-71 
(Taken up on the ledge) 



COLLATING THE CUNEIFORM TEXT 197 

linson {obi-) and also by Weissbach and Bang, because I had 
the latter volume with me on the rock and should probably 
have recorded a variation if there had been one. 

Bh. 1. 66 : my memoranda and ' snapshot ' photographs of 
portions of 1. 66 show that this line stands as given in the 
accepted text, excepting the c in pdrsamcd and the final mcd 
of mdda\mca]. But this is a matter of minor importance. 
My 4 snapshots ' also help to assure the accuracy of several 
other words in 11. 6570 ; I only wish I had taken more photo- 
graphs, despite the great difficulty in using a camera on the 
ledge. 

Bh. 1. 86, m a -y a kduvd (?) etc. : the first letter of this word is 
very uncertain, but the notes which I made upon it on two 
different days seem to confirm the accuracy of the initial m. 
My notes on the last occasion remark that the first part of the 
word looks more like m a y a , and I twice sketched the remnants 
of the cuneiform characters, with a special comment on the very 
scanty space between the m (?) and the y. My second drawing 
in pencil indicates more especially the illegibility of the m> 
which can be inferred, however, from the dots that are faintly 
visible, but can be made out only with difficulty. The same 
sketch seems to emphasize again the small space between it 
and the y. On returning to America and gaining access to 
my books, I find that whereas in JRAS. 10. p. xlv, Rawlinson 
first said, 4 There would appear to be a sufficient space for two 
letters between m and &,' he afterwards corrects this statement 
(JRAS. 12. p. ii, appendix ; cf. Bartholomae, IF. 12. 132 
note) by noting l there is only one character wanting in the 
word ma-kd'uwdS This later remark would agree precisely 
with my own independent observations. As to the correctness 
of ?/, which is not given in any of the editions except in that 
of Weissbach and Bang and is marked as doubtful by them, I 
have no hesitation. Both my pencil sketches of the cuneiform 
characters present a y, and so do my memoranda. The last 
part, -kduvd, of the word under consideration is perfectly clear, 



198 THE GREAT BEHISTAN ROCK 

as my notes on each letter show. The only Avestan word that 
I can recall that is at all like this dubious m a y a k a auv a a is Av. 
maekaintls, Ys. 38. 3 ; or is it 'beasts of burden,' cf. Skt. 2 
mdyaf 1 As to the reading of the two words adam Jcdram, 
which stand before this provoking word, there is no uncertainty. 

Bh. 1. 86, aniyam usabdrim : an examination and re-examin- 
ation of the rock proves the certainty of this reading. The 
word-divider precedes the u and is all right. The u itself, 
while not clear, can be made out sufficiently well, for I exam- 
ined it on two different occasions in order to be perfectly sure. 
The s I found to be beyond question, and that without notic- 
ing the loss of any sign after it in the margin, as Foy, KZ. 35. 
36, would assume. The chiselling of -bdrim in the following 
line shows that portion of the compound to be perfectly clear. 
Oppert's original conjecture usabdrim would therefore be sub- 
stantiated. 2 I refrain here from entering into a discussion of 
the derivation of this much mooted word. 

Bh. 1. 87, aniyahyd asam (sic) [..^dnayam : the form aniyahyd, 
as given, is accurate, though the word is damaged. The read- 
ing asm .. of Spiegel and Kossowicz, or tasma\_kam\ of Friedrich 
Miiller, WZKM. 1. 222; 11. 253, and as [pd] of Weissbach 
and Bang, though the latter were on the right track (cf . Gray, 
AJP. 21. 21), must be abandoned. The word is simply asam, 
4 horse ' (ace. sg.). The m at the end is very distinct ; the 8 is 
very plain ; and the initial a is quite clear. To these comments 
my note-book further adds, 4 the word-divider after it is quite 
clear.' Further conjecture is therefore unnecessary, and I find 
that my observation as to the m receives additional corroboration 

1 [I should now be inclined to read 2 Foy writes me, however, (Dec. 20, 

ma$kd A uva,i accepting the suggestion 1903) that he still believes that w[r] a 

made to me by Professor Ferdinand should be read and that the stone-cutter 

Justi, in a letter dated Nov. 25, 1903, made a mistake as the word stands at 

and in his published notes in IF. An- the end of a line ' ein steinmetzver- 

zeiger, 17. 125. Similarly also Foy, sehen (usa statt ustra)." 1 I should 

KZ. 37. 553, and Hiising, as mentioned feel considerable hesitation about 

by Justi, IF. Anzeiger, 17. 125.] accepting such a view. 



READINGS IN THE FIRST COLUMN 199 

from Rawlinson, who read 4 asm . . . dnayam^ but he did not 
notice the word-divider after asam and consequently vocalized 
the word erroneously. The old difficulty, however, with regard 
to the obliterated prefix of [. .~\dnayam, still remains. Over each 
of the cuneiform letters of the legible part, -a n a y a m a , I have 
written 'O.K.,' i.e. 'all right,' in my note-book. But on ex- 
amining the conjectural pati as prefix in Weissbach and Bang 
I have added a memorandum ' pati extremely questionable ; 
the initial letter can hardly be p at all.' In fact, as my notes 
continue, ' it is hardly possible to read the prefix,' because the 
rock is so damaged. I appended a further note that the 
appearance of the word suggested rather \up\dnayam or 
\uz\ dnay am. As to form and composition neither of these 
prefixes would be impossible, as I have since found on being 
able to consult my Sanskrit dictionary on the verbal prefixes 
under ym. But such a restoration is quite uncertain, though 
I tried my best to assure it by examining the weathered stone 
again and again. Whatever the prefix may be, the sense seems 
clear when combined with the new reading asam (ace. sg. for 
plur., special for general), and the sentence aniyahyd asam 
updnayam may be rendered, 4 1 brought up horse (s) for the 
rest (of the army).' This interpretation is apparently also in 
accord with the Elamitic version, cf. Weissbach, Achameniden- 
inschriften Zweiter Art, pp. 63, 64, and Foy, KZ. 37. 554. 1 

Bh. 1. 88, exit avadd, re-enter avam : the reading of Spiegel, 
Kossowicz, and Weissbach and Bang is wrong as far as avadd 
is concerned, and that too despite the fact that Spiegel, 
Keilintchriften, 2d ed., p. 11, n. 88, is following the authority of 
Rawlinson's revision in JRAS. 12. p. ii, appendix. The 
rock plainly gives avam at this particular point, even though 
avadd occurs often elsewhere in the inscriptions. The m of 
avam is clear, as is shown by my notes and my sketch of the 

1 [Messrs. King and Thompson, of asam, 'horse(s),' and they suggest 
the British Museum, now write me franayam, 'I led forward,' for the 
(June 13, 1905) that they also read verb.] 



200 THE GREAT BEH1STAN ROCK 

cuneiform characters. The v is not quite plain, but can be 
made out. Regarding the initial a there is absolutely no 
doubt. In JRAS. 10. 211, etc., Rawlinson originally read 
quite correctly 4 awam kdramj ' that army,' just as in Bh. 2. 
20, 25, 41, 46, etc. There was no occasion for his departing 
from that. It may be added by way of supplement that the 
letters -dva of the adjacent word [pas]dva are right, though 
the first part of the word is broken. 

Bh. 1. 92-96 : the t in nadHabaira (1. 92) is legible, and may 
reasonably be removed from italics in our transcribed editions 
of the text. I made an incidental note also that disa hadd 
(1. 93) is accurately recorded in our texts, and that akumd 
(1. 96), though defaced, is still legible. There were evidences 
also that the rock has suffered since the days of Rawlinson. 

Bh. 2. 59-61 : these three lines are precisely as given in 
Rawlinson and Spiegel. The lacuna indicated by Weissbach 
and Bang in 4 nd . . avam Tcdram ' (2. 61) is wrong and is evi- 
dently due to a misprint (which misled Bartholomae, IF. 12. 
135). In printing, the two points . . have accidentally slipped 
in by mistake from the fragmentary -iyamanam in the next line. 
The reading is therefore mand avam kdram, as in Spiegel. 
This note applies consequently to the Weissbach and Bang 
edition alone. 

Bh. 2. 61, \@ a u']r a v a ah a r a 'h a y a a : the first two letters are much 
defaced, but the third letter (r) may be deduced from the three 
dots or holes left in the rock by the deeper cutting at the head 
of the wedge, although the less sharply chiselled parts of the 
letter have been marred by the disintegration of the rock. By 
practice on the rock the eye becomes accustomed to distinguish- 
ing such details, and the restoration can be made with com- 
parative certainty. 

Bh. 2. 75 (cf. 2. 89), utd8ai[y~\ [ca8ma~\ avajam, etc. : at the 
beginning of this line the y of utdsa\iy, though faint, is never- 
theless to be inferred from the indentures or dots that are still 
quite distinguishable. Recall what was said above on such 



THE SECOND AND THIRD COLUMNS 201 

dots as means of restoration. The obliterated word, read as 
casma or caxsma, yielded no new results and is equally illegible 
in both 2. 75 and 2. 89. 1 At 2. 75 I have merely noted regard- 
ing the fragments of an internal letter that it 4 looks more like 
an h than it does like an *,' but the likeness between the two 
letters in the cuneiform character leads easily to misappre- 
hension, and certainly casma suits the sense, for the loss of an 
eye or both eyes, inflicted as a punishment in addition to other 
mutilations, is precisely what one notices or hears of in remote 
parts of Persia to-day, as in the days of King Darius. The 
sight is destroyed by means of a red-hot iron brought near 
to the ball as noted, for example, by Landor, Across Coveted 
Lands, 2. 191. This latter observation may throw some addi- 
tional light on the meaning of avajam (2. 75). The reading 
of this word is beyond question. So also -is duvaraydmaiy ; 
but the words basta addriy are now illegible still another 
proof of the damage done by the water since Rawlinson exam- 
ined the rock some sixty years ago. Simply by way of record 
it may be worth adding that the last two words of this line 
(2. 75), haruvasim k\dra, are in perfect condition at present. 

Bh. 3. 87-91 : some time was spent in trying to see if any- 
thing new might possibly be got out of the closing lines of the 
third column, but the action of the water had so completely 
obliterated the words that even the last two at the bottom of 
the tablet were less clear than they were to Rawlinson. My 
notes show that dp of \uzmay\dpatiy may be inferred from the 
faint remnants of these two letters ; the last part of the word 
is all right. The k of dkariya n tdm is apparently rightly read ; 
I have added ' k is best,' but have repeated that it is much 
damaged. 

Bh. 4. 46 : so far as the first three words xsdya\6iya vasnd 

1 [Messrs. King and Thompson write a possibility, comparing Av. a- in 

me they read ' utsam avajam, with Bartholomae, Air. Wb. p. 229, although 

u8a, "eye," cf. Skt. aksa." 1 There a dual would rather be expected. I 

seem to be phonetic reasons against have not overlooked the remarks by 

this, but I could understand asam as Foy, KZ. 37. 554-555J 



202 THE GREAT BE HIST AN ROCK 

aura[mazddTia] are concerned, the reading is as in the texts 
already cited ; but I was able to make out the faint remnants of 
m a z a d a in the divine name. In the fourth word, like Rawlin- 
son, I could only read the latter portion, or -maiy, so there is 
still an opportunity for conjecture, and the suggestion of Gray, 
JAOS. 23. 62, to read avdmaiy for avdt-maiy, is as satisfactory 
a way of filling the lacuna as any. The word aniyasciy, fur- 
thermore, I have marked with an 'O.K.' ('all right') in my 
memoranda. 

Bh. 4. 49, avdh\yd paruv 6ada . . . : the first two words are 
clear, but the verb is in bad condition. Its introductory part, 
however, can be made out, and I have marked the d as ' O.K.,' 
but with the latter part of the word I could do nothing. 

Bh. 4. 50, maniy \_ataiy~]'. despite the syntactical grounds 
favoring a subjunctive, we must accept a short a, judging 
from the rock. On the margin of my text I have distinctly 
recorded, 'no space for long a; what remains of the t comes 
directly after y." 1 1 

Bh. 4. 51, paruvd xsdyad[iyd'] . did dha, etc.: the first two 
words are quite clear on the rock; the remnants of the third 
one I read as . did without looking at the printed text of Spiegel 
or of Weissbach and Bang, which I had with me. Rawlinson, 
followed by the later editors, gives only the final a, whereas I 
distinctly made out . did independently, as stated. Weissbach 
and Bang conjecture [ydt~\d, and compare Elam. kus. With 
regard to the next word, I have noted : ' dha can be made out 
on the stone without question.' The reading, therefore, is 
assured, and I have marked avaisdm and astiy as all right ; but 
the naiy between them is no longer clear, although it may be 
inferred from the appearance of the stone. Accordingly I have 
marked it 'O.K., inferable.' 

Bh. 4. 58, ddraya[vaus xsdya]6iya nuram : the name of 
King Darius is apparently somewhat more damaged now than in 
Rawlinson's time ; but that is a minor matter. More important 
1 Cf. also Foy, KZ. 37. 488, note. 



PROBLEMS IN THE FOURTH COLUMN 203 

is a memorandum regarding the absence of any break before 
nuram* In my notes I have drawn a circle around the dots in 
the Weissbach and Bang edition, to indicate that the lacuna is 
to be struck out, and have added a definite memorandum, 4 omit 
the space ; the word nilram comes after the word-divider that 
followed [xsdya] Qiya? Foy's conjecture of add, KZ. 35. 34, n. 1, 
is therefore needless. 

Bh. 4. 64, na\iy zurakara dha[m~\: the last part of naiy is 
much broken, but the reading appears to be all right. On 
examining zurakara I first noted 4 not wholly clear, as the stone 
is somewhat marred, but still z a u r a k a r a does seem all right.' 
On re-examining it the following day in a better light, I added 
that the reading is confirmed. On looking up Rawlinson I 
find that he gives the cuneiform quite clearly, which again bears 
out the idea that the rock has suffered since his time. As to 
dha[m~\ I have marked 'all right' over dh a in my copy and 
have added 'probably right' with reference to what can be 
made out of the final ra. The printed editions have the same. 

Bh. 4. 64, [naiy adam na]imaiy taumd : I was not able to 
make anything out of the missing letters that are indicated here 
by being enclosed in brackets, and I wrote ' absolutely illegible 
to me ' over [adam] . But on the last day in the strong sun- 
light I corrected this by a supplementary remark that ' I fan- 
cied I could discern the a quite clearly, and remnants of the 
crossbar and upper parts of a d, together with a fragment of 
the horizontal wedge and possible traces of the indenture caused 
by the deep strokes in the nail-heads of an ra.' This, therefore, 
is something towards assuring the accuracy of [adam]. Re- 
garding the first i in [no] imaiy I have noted that it is 4 right, 
but broken badly.' As to taumd there is no doubt. The read- 
ing is quite accurate, as in all our texts. 

The next three words, and those following them, called for 
much study, the results of which will now be given. 

Bh. 4. 64, upariy abstain upariy : I spent much of my last 
two days in examining this passage, which was one of the 



204 THE GREAT BEHISTAN ROCK 

incentives for my going to Behistan, because of the bearing of 
the whole sentence on the question of the religion of Darius as 
a Zoroastrian and the faith of the Achsemenian kings, which I 
have discussed in JAOS. 21. 169, 172-175. I returned to the 
line again and again, studied each word under different lights, 
sketched it, and made rubbings, so far as I could without injur- 
ing the stone, which I was fearful might in some way become 
defaced. The first of the two upariy' s is much damaged and 
difficult to read ; but on the last day I was fortunate in having 
bright sunlight, so that I could examine it well and compare it 
again and again with the similar word at the end of the line. 
I found distinctly that it is upariy (with it), not apariy as has 
been suggested, for example, by Foy, KZ. 35. 45, n. 1; 37. 502, 
where the Elamitic version is also discussed. This first upariy 
therefore remains unchanged, as in Rawlinson. I came to like 
results as to the second upariy, which is more distinct. Over 
its u in my text and over a part of p and over the y I have 
written 'all right.' Below the part pari I have marked ' much 
defaced.' But on the following day when the sun was bright 
the word came out quite clearly, and I appended the note, ' it is 
all right.' Therefore the second upariy must likewise stand, as 
in Rawlinson. 

I was most anxious, however, to examine the word between 
these two and to find whether it is abistdm^R^^ abastdm^R^), 
or drstdm (Foy's conjecture) see my remarks in the article 
already referred to, JAOS. 21. 169, 172-175. Great care and 
attention were given to the examination, and after working on 
the individual letters I made it a point to turn to something 
else and then to return again and again to verify my memo- 
randa and my sketches. In the first place, there is no i in the 
word ; any such reading as dbistdm must therefore be dropped. 
We have, accordingly, to do either with abastdm or with Foy's 
conjecture drstdm, the point being merely whether the text has 
a b or an r, because each of the other letters a.stdm is perfectly 
clear, as my memoranda again and again show. The whole 



UPARI ARSHTlM 205 

question between the two mooted letters is whether we have 
the slight horizontal mid-bar of the cuneiform ^| r or simply 
the two parallel wedges of the y b. I must mention at once 
that the shape of the cuneiform character for r on the Behistan 
rock does not exaggerate the middle of the three horizontal 
strokes so much as does the type in our printed editions. I 
noticed this particularly, and my photographs taken on the ledge 
also bring it out. Therefore the middle wedge is naturally 
less prominent, and when defaced by the action of the water, as 
this word has been, it becomes very faint. A feeling of un- 
certainty arises as to whether the mark be an intended indenture 
or an accidental dot, because of the peculiar brownish mottled 
appearance of the somewhat porous stone when it is exposed 
to the disintegrating water. But each time I returned to the 
word I became surer that Foy is right and that r, not 6, is to 
be read. I examined the letter in comparison with the other r's 
in the vicinity, when these had suffered from the water, and 
always with the same result. I believe therefore that ar a 8 a t a am a , 
i.e. drstdm for drstdtam, ' Arshtat, Uprightness, Rectitude,' is to 
be read with Foy, and he is to be heartily congratulated on his 
shrewd conjecture. 1 At the same time, I would suggest the 
need of caution in making further conjectures. The days spent 
up on the ledge at Behistan have made me more conservative 
than ever, and in cases of doubt I should generally rely on the 
faithful Rawlinson until the rock itself be examined. 

As to the beginning of the next line, which is variously given 
as mam naiy, as ay am naiy, etc., I confess that when I first read 
the fragmentary second element I marked it as agreeing with 
naiy in the transcribed text which I had with me. But on 
re-examining the damaged fragment I recorded in my note-book 
that 'instead of ay am naiy, it looks more like a long word 
ending in -h a iy or -} a iy the former, -h a iy, is, however, better, 
and it seems so to be clear.' Later I added again, 'it does 

1 On the grammatical formation of [King and Thompson also read arstdm 
the word, see also Foy, KZ. 37. 503. letter dated June 15, 1905.] 



206 THE GREAT BE HI STAN ROCK 

not look like naiy.' This makes the question of the reading 
naiy for a moment uncertain. On returning to America and 
gaining access to Rawlinson's draft of the cuneiform characters, 
I was interested in finding that he also has -h a iy, and he must 
likewise have had the impression of a longer word, as his 
y a . . . t a h a iy a seems to prove. Yet in his later revision (JRAS. 
12. p. viii, appendix) his cuneiform text runs up a r a iy a m a am a 
n a iy a , with the word-divider before mam, and with this reading 
Spiegel and the other editors have operated. Whatever we 
may say on that matter, the Elamitic version is certainly in 
favor of reading the second word as naiy, for it has the 
usual negative particle inne, 'not,' as I find on consulting 
the text. 

Bh. 4. 65, ... sakaurim, etc.: the text in Spiegel and 
Kossowicz, partly following Rawlinson, gives upariy \ mam naiy 

sakaurim huvatam zura akunavam ; Oppert, Le Peuple 

et la Langue des Medes, p. 183, writes 4 upariy ay am naiy uvdrim 
naiy druvaptam zaura akunavam* ; Fr. Miiller, WZKM. 1. 60, 
reads, 4 aparijd\jdma naij sakaurim [naij a~]huwatam zaura 
akunawam' ; Weissbach and Bang present upariy ay am naiy 
sakaurim [naiy] .... huvatam zura akunavam ; Foy, KZ. 
35. 45, first suggested a correction of the text, i.e. 'in dpari- 
ydyam zu verbessern,' and he altered sakaurim into h ukdrim 
and huvatam into d u uskaram (on which see Bang, IF. 8. 292); 
then Bartholomae, IF. 12. 130, made the radical conjecture 
naiy d h urim naiy duruvd"tam ; finally Foy, KZ. 37. 557, shifted 
his ground and made a new guess, dasurim . . . \ai\na h uva n tam. 
I can only add that with regard to sakaurim suggestions for 
altering the text may be practically abandoned. The stone 
plainly gives s a k a u r a i (?) m a . Regarding the first three 
letters, s a k a u, there is no doubt, as a repeated examination of 
the word proved. The r a , however, is very unclear, but the 
holes or dots of the defaced wedges would allow an r. The 
same is true of i, which looks somewhat like an a, but the dots 
favor an i. The final letter m is marked in my notes as 4 even 



THE END OF THE FOURTH COLUMN 207 

less clear, but the dots would not be against m ' ; and on a third 
examination I became still surer of the m. I added a remark 
to the effect that the passage must have suffered since Rawlin- 
son's time. 

Bh. 4. 65, n[aiy~\ : regarding the word after sakaurim, I have 
noted 4 the n of naiy(?) after sakaurim can be made out by the 
dots.' This reading is supported by the Elamitic, which again 
has inne, 'not,' just as it has before the word corresponding to 
sakaurim. 

Bh. 4. 65, [..]ttt?ota?(?) : for this much debated word, formerly 
read . . huvatam, etc., I have now some new material to offer. 
The text is indeed much mutilated, but each of the letters 
u v a t a m a is legible, although the final m is in bad condition. 
On studying the first part of the word, I noticed that 
instead of an A, as is commonly supposed, we have another letter, 
a character that looks more like s. On looking closer, it became 
perfectly clear to me that the supposed s was not s at all, but 
apparently n u , although this might possibly be a mistaken read- 
ing for r", the uncertainty being due to the resemblance between 
the characters in the original cuneiform if the horizontal bars 
are somewhat marred. The sketch made in my notes, however, 
looks precisely like n u . A further examination of the damaged 
part revealed an apparent m standing before this, so that we 
may assume that the word began with m. 1 

Bh. 4. 65, zura akunavam, etc. : each letter of these two 
words is legible, and the same is true with regard to each of 
the words that follow in this line. 

Bh. 4. 66, v l \i~]6iyd: this is rightly read, although the v* is 
'broken, but O.K.,' and it is 'hard to be sure of the final a,' 
although it is ' probably correct,' as I have noted in my text. 

Bh. 4. 68, hy a aparam ahy, etc. : these first three words of 

1 [Even if my reading be correct, I I am not unmindful that the Elamitic 

have no etymology to offer in explana- version is read as m istukra ; nor have 

tion of the word, not even to compare I overlooked the various conjectures 

Skt. manuvat as I doubtfully sug- by the scholars I have mentioned in 

gested at a hazard in JAOS. 24. 24. my paragraph on sakaurim, above.] 



208 THE GREAT BEHISTAN ROCK 

the line are weathered and defaced, but they are rightly given 
in the texts. The same may be said of ' italic ' martiya of the 
editions, but the word is to be inferred from the stone. With 
regard to draufana I have recorded ' weathered, but O.K.' The 
subjunctive ahatiy stands as in the editions; so does hyavd, but it 
is weathered. Regarding [a]tar[], the whole word, except the 
internal letters, is 'so weathered as to be practically illegible.' 

Bh. 4. 69, ahat\iy avaiy md daustd avaiy, etc.: the first and 
third words are all right ; so also is the first avaiy, although 
it is hard to read. The last part of daustd is scarcely legible, 
though it may be inferred from the appearance of the stone. 
The second avaiy is illegible. The long word ahifrastddiy is 
'all right, but in part difficult to read.' The imperative parsd 
is 'almost illegible.' 

Bh. 4. 71, 73, vikandhy : 'so best with the letter k, not s.' 

Bh. 4. 76, avataiy auramazdd : the first word is ' apparently 
all right, but almost illegible ' ; the second, or divine name, is 
' inferable ' ; as to mazdnam(f) of Weissbach and Bang, for 
which Oppert conjectured vazrakam (see Foy, KZ. 35. 47 ; 
37. 558), I have written ' illegible' above it. 

Bh. 4. 77, vikandh[i~\dis : the k is 'fairly clear' and 'best 
read so.' The [i'] is 'omitted on the stone.' 

By this time the westering sun for one learns in Persia to 
live by the sun warned me that I must descend for the last 
time from the rocky height in order to start once again for 
Hamadan and begin my journey to Southern Persia. I was 
loath to leave, but leave I had to at last if I were to carry out 
my plans for seeing Isfahan, Persepolis, and Shiraz, and for vis- 
iting the Zoroastrians at Yezd before going to the capital and 
journeying thence to Merv, Bokhara, and Samarkand. On 
reaching the plain once more there was an opportunity to urge 
the inhabitants of Bisitun charily to guard their inscription and 
to tell them of the divine blessings which King Darius invoked 
upon all such as do so, and of the curses that were assured 
if it were injured. 



THE SCULPTURE OF THE PARTHIAN KING GOTARZES 209 

I did not leave Behistan, however, without making an 
examination of the mutilated Parthian sculpture on the panel 
at the base of the hill, upon the right of the approach to the 
great inscription. The bas-relief has been nearly destroyed, 
but we know that it represents the triumph of the Parthian 
king Gotarzes (A.D. 46-51) over his rival Meherdates, who was 
likewise a Parthian, but was brought up at the court of the 
Roman emperor Claudius, and made an unsuccessful attempt 
to gain possession of the Persian crown. An inscription, 
written in Greek, after the manner of the Philhellenic Parthi- 
ans, records, or once recorded, the names of ' Mithra[da]tes 
the Persian . . . Gotarzes, Satrap of Satraps (i.e. king of 
kings), the son of Gev ' ; but now the names are almost 
obliterated with the exception of TOTAPZHC, which can still 
be read distinctly. Most of the sculpture was destroyed about 
a hundred years ago by an act of vandalism of a curious kind. 
A Persian overlord, Sheikh Ali Khan Zanganah, cut an arch- 
shaped panel directly in the middle of the sculpture, in order 
to record, in an Arabic inscription, a gift of the income of two 
villages, which he donated for keeping up a caravansarai he 
had built at Bisitun. We can but regret that his act of gener- 
osity led him to mutilate a valuable historic sculpture. 1 The 
elements have likewise contributed much to the destruction of 
the bas-relief, which was but rudely carved at best, so that the 
carvings are greatly weathered, but it is still possible to dis- 
tinguish two figures of heroic size on the left as one faces the 
panel, and vestiges of still another on the right. Above the 
latter there once were two small figures of mounted warriors, 
each armed with a spear, one pursuing the other; over the 
latter's head, however, floats an angel holding a garland of 
victory. 2 These effigies symbolized the struggle between 

1 The arched panel is seen already 2 These figures are to be seen in 

in the sketch by Ker Porter, Travels, the drawings of Flandin and Coste, 

2. 151 ; for a note on Sheikh Ali Voyage en Perse, Ancienne, 1. plates 

of Zanganah, cf. also 2. 85-86. 16, 19. 



210 THE GREAT BEHISTAN ROCK 

Meherdates and Gotarzes, but the figures are now nearly 
obliterated, and I could make little out of the sculpture, although 
much depends upon favorable light at the time when it is 
examined. 1 

As we stand at the base of the panel and look out over the 
plain in the direction of the village of Bisitun, the eye can 
plainly see the site of what must have been an ancient building, 
as indicated by the outlines of the walls. The natives call this 
4 Khosru's Place ' (G-ah-i Kei Khosru), and possibly this Sasa- 
nian king once had a villa there, as the country round about 
was a favorite territory of his and his name is preserved in the 
locality in various ways. 

There is still one more relic of antiquity at Bisitun, which 
I shall describe, because it seems to have escaped detailed 
notice in the books that I have consulted, although the monu- 
ment is well known to the natives. My guides, when asked 
whether there were any carvings or inscriptions besides those 
which I had examined, told me that there was a sculpture on 
a large boulder around the right flank of the mountain, not far 
from the base of the cliff where the Gotarzes panel is cut. 
Pointing out the direction, they led me around the spur and 
then up a slight incline to a place where stood a huge boulder 
of stone, some twenty feet in circumference and ten in height, 
carved on three sides with life-size figures in low relief. As I 
could not take a photograph, because it was too late in the 
afternoon, I made careful notes of the sculptures and shall 
describe them with more detail than might otherwise be 
necessary. 

The middle figure, facing the natural approach up the 
mound, is a bearded personage, with mustache and hair 
distinctly marked, and wearing upon the head a roundish cap. 

1 For the history of Gotarzes, and linson, Sixth Oriental Monarchy, pp. 

for bibliographical references concern- 249-261. Mr. Rabino kindly sent me 

ing the inscription, see Justi, Grundr. notes regarding the Greek inscription 

iran. Philol. 2. 504-505 ; cf. also Raw- to supplement my own memoranda. 




FLANDIN'S SKETCH OF THE GOTARZES SCULPTURE 




THE PARTHIAN SCULPTURE OF GOTARZES, MUTILATED BY A MODERN INSCRIPTION 



A SCULPTURED BOULDER 211 

The close-fitting upper garment and tunic have no decoration, 
but a girdle is worn about the waist and a necklace encircles 
the throat, as five rings of the band can be distinctly seen. 
The left hand holds a bowl ; the right hand is extended over 
a low column that may be a fire-altar, and it holds some object 
which I could not make out, but which is apparently connected 
in some way with the oblation. The legs of the figure are very 
fat and are spread apart, as in walking, and they appear to be 
encased in greaves, or in coverings marked with grooved lines, 
which make them look like the buckskin leggings of a crick- 
eter. In appearance the figure looks as if it might possibly be 
intended to represent a Magian priest, although it would be 
difficult to affirm this; there is certainly nothing either mili- 
tary or kingly in the bearing of the sculpture. 

The second figure is carved on the right side of the boulder. 
The face is fat and round and apparently without a beard, so 
that the head looks almost like that of a woman or a youth. 
There is a necklace about the throat, and a bracelet on the left 
arm, which is clearly marked, although the right arm is not 
distinct. The body is so poised as to throw the weight on the 
right foot, which is close to the left, as if making a step, and 
the fatness of the legs is as noticeable as in the case of the other 
effigy. 

The third figure occupies the left side of the boulder and 
appears to be approaching the central figure. Like that of the 
central figure, the face seems to have a scraggly beard, but 
this detail is not clear. The left arm is not visible, but the 
right is clearly carved and in the hand something is held, which 
again is indistinct. The legs look as if they were covered 
by a low boot or cothurnus, somewhat after the manner of the 
sculptures at Persepolis. In pose the figure is quite lifelike. 

From the standpoint of art the figures are crude and heavy, 
although not wholly lacking in effectiveness. The plumpness 
of the legs in each case is striking, and it reminded me of the 
bas-reliefs over the inscriptions of Darius which I had been 



212 THE GREAT BE HI STAN ROCK 

examining, and also of some of the Achsemenian sculptures 
at Persepolis. There was no evidence of the flowing trousers 
or the balloon-shaped hats that belong to the Sasanian period. 
On the whole I believe that the sculptures on this monolith 
near the Behistan rock are to be ascribed to the Achsemenian 
era rather than to any other. 

Time was now up, and I left the scene of the great inscrip- 
tion on Friday afternoon, April 17, feeling painfully aware 
that I might have accomplished more if my time had been 
longer and my means greater, but happy in heart at the thought 
of having possibly contributed something toward our better 
knowledge of the Behistan text of the inscription of Darius, 
and inspired by the hope that an opportunity may in some way 
be offered me to go again and complete such parts as had to be 
left undone at the moment. 



CHAPTER XV 
TAK-I BOSTAN AND KERMANSHAH 

' Near Kermanshah, and between Hamadan and Hoi van, is the Castle of 
Shirin. The name Shirin in Persian signifies " sweet," and she was the favorite 
slave of King Parviz.' YAKUT, Majma^ al-Buldan, s.v. Kirmasm. 

THE crags of Mount Bisitun looked black and gloomy 
through a mass of fog in the early dawn as I rode out of the 
manzil, after my first two days at the scene of the inscription, 
and directed my way toward the city of Kermanshah and the 
beautiful villa of Tak-i Bostan in its environs. This latter 
place was a famous park in Sasanian times, thirteen centuries 
ago, and it is still renowned for its ancient sculptures. As 
it was necessary to make only a slight detour on my way to 
reach it, I decided to go to Tak-i Bostan before proceeding to 
Kermanshah. 1 

My road lay through a rolling plain bounded on the north 
by the mountain ridge of Kuh Paro, which runs from Bisitun 
to Tak-i Bostan, and on the south by spurs of the great range 
of hills that extend past Kermanshah into Luristan. This 
tract of country is one of the richest pasture-lands in Persia 
and is renowned for the good horses which it produces and for 
which Persia has ever been famous. 2 It is well watered by the 
Gamasiab River, which runs past Bisitun, and the river Karasu, 
which joins with the latter not far from Kermanshah and con- 
verts the region into a succession of meadows. The nomad 
tribes of the Iliats pitch their black tents for a longer stay with 

1 In point of time this chapter falls 2 Darius calls Persia the land ' of 

between my two different sojourns good horses ' (uvaspa) in his inscrip- 
at Bisitun, which I have combined tion, Dar. Pers. d. 8. 
in the preceding chapter. 

213 



214 TAK-I BOS TAN AND KERMANSHAH 

their flocks when they reach this fertile belt, and they dot the 
green plains for miles with their portable villages. My own 
small caravan quickened its pace as the sun burned the mist 
away, and we traversed the twenty miles quickly, reaching 
Tak-i Bostan before ten, having passed in that time to scenes 
associated with events a thousand years later than Achserne- 
nian Bisitun. The sculptures of Darius and the carvings of 
the Parthian Gotarzes gave place to bas-reliefs cut by Sasanian 
kings, and the cuneiform characters of the Achsemenian inscrip- 
tions were transformed into the cursive ligatures of the Pahlavi 
script of the Middle Persian period. 

Tak-i Bostan lies about four miles northeast of Kermanshah 
and now forms part of the estate of a wealthy Persian land- 
owner, the late Haji Agha Hasan, Vakil ad-Daulah, 4 Deputy 
of State,' and British agent at Kermanshah. The people 
look upon the enclosure as a pleasure ground, and many 
of them have forgotten that it was once a chosen seat of the 
Sasanian kings. Its situation is certainly a delightful one. 
Couched at the base of the mountain ridge that runs from 
Bisitun, and richly supplied with water that springs from the 
mountain's feet and converts the enclosure into a garden, the 
place when I saw it showed traces of the veritable paradise 
that it must have been in the palmy days of the Sasanidse. 

The name Tak-i Bostdn, which often sounds like Tdgh-i Bos- 
ton, or is even turned into ' Taw-ou-stdn ' by a slovenly labial 
pronunciation among the peasants, means 'Arch of the Garden' 
and is given to the enclosure because of the arched recesses 
carved in the rocky base of the mountain. The place is often 
called also Takht-i Bostdn, 'Throne of the Garden,' appar- 
ently from a stone ledge cut high in the rock above the arches 
and approached by a hundred or more steps in the face of the 
cliff. * In earlier times it was sometimes popularly known as 

1 The Arab geographer Ibn Rostah Kitdb al-A'ldk an-Naflsah, ed. De 
(about A. D. 950) speaks of these steps Goeje, Bibl. Geog. Arab. 7. 166. 
as numbering * about 250,' in his See p. 224, below. 




THE SCULPTURED PANEL AT TAK-I BOSTAN 
(From de Morgan, Mission Scientiftque en Perse) 



A GARDEN OF THE SASANIAN KINGS 215 

Shabdiz, from the statue of Khosru's horse, and sometimes 
spoken of as Kasr-i Shirin, ' Castle of Shirin,' from the beauti- 
ful favorite of Khosru Parviz. The former name we know 
from Ibn Rostah and Al-Hamadhani, a thousand years ago, 1 
and the latter, 4 Shirin's Castle,' from Yakut, two centuries 
later, who mentions it as a place containing the remains of 
'many porticoes, halls, pavilions, great vaulted arches, lofty 
terraces, gardens, and a park, which surpassed in magnificence 
even the splendor of its royal founder.' 2 The name Kasr-i 
Shirin, however, belongs more strictly to the great mass of 
ruins known by that title some eighty miles westward from 
Kermanshah, near the Turkish border. 

A wall surrounds the park of Tak-i Bostan, and as we pass 
through the crumbling gateway we find ourselves directly on 
the edge of a miniature lake, a reservoir about one hundred 
and twenty yards square, fed by the waters which burst from 
beneath the massive rock that overlooks its crystal surface, 
and find outlet in the reservoir below, whence they stream 
through a number of channels used for irrigation. A path- 
way shaded by willows leads around the pond, and by the 
water's edge and near its channels there are to be seen sev- 
eral fragments of white marble columns, capitals of pillars, 
and some broken pieces of statuary dating from Sasanian 
times. 3 A comparatively modern two-story building belong- 
ing to the Vakil ad-Daulah stands as a pavilion on the 
edge of the pond at the point where the water rushes into 
it. 4 The whole scene when I saw it, clothed in the rich 
green of springtime and set off by a background of blue sky 

1 Ibn Rostah, ed. De Goeje, 7. 166 4 This villa does not appear in the 
(see preceding note), and Al-Hama- sketch made nearly a century ago by 
dhani, ed. De Goeje, 5. 214-215 Ker Porter, Travels, 2. 169, pi. 61, nor 
(Shabdiz) and 5. 211 (Kasr-i Shirin). were the reservoirs full, judging from 

2 Yakut, p. 448 ; cf. also pp. 438, his drawing and description of the 
345. place, but the water is to be seen in 

3 Drawings of these may be seen in Flandin's plate (Voyage en Perse, 
Flandin and Coste, Voyage en Perse, Ancienne, 1. pi. 3), which I have 
Ancienne, 1. plates 1-14, 17, 17 2 . reproduced below. 



216 TAK-I BOSTAN AND KERMANSHAH 

that brought the rugged outline of the bare hills into bolder 
relief, was picturesque in the extreme and reminded me of 
some Italian villa ; but I can imagine that much of its beauty 
may be lost when the water of the reservoirs dries up or is 
reduced to a running stream, which happens in the drought 
of summer. 

To the visitor at Tak-i Bostan the most interesting feature is 
the collection of Sasanian bas-reliefs sculptured in two large 
grottos that are hewn in the bosom of the rock and carved 
upon a panel on its lower surface. These sculptures are per- 
haps the best extant examples of Sasanian art, and they show 
Roman influence of the Byzantine period, with a possible trace 
of Greek art due to Alexander's conquest. 1 There are three 
separate groups of bas-reliefs, and I shall discuss them in 
the order of their arrangement, and probable sequence of 
time. I shall begin with the one on the panel at the base 
of the rock, then proceed to the two statues in the smaller 
grotto, and conclude with the elaborate sculptures in the 
larger vaulted recess. 

The first set of bas-reliefs and presumably the oldest, although 
this is not certain, consists of four figures carved on a smoothed 
surface of the rock next to the villa and covering an expanse of 
about eighteen feet by ten. A special interest attaches to the 
group, as I have stated in my Zoroaster, because it has been 
thought that the figure to the extreme left may represent the 
Prophet of ancient Iran. 2 For that reason I devoted particu- 
lar attention to it on the occasion of my visit to Tak-i 
Bostan. I shall briefly describe the composition of the sculp- 
tured group. 3 

The two figures to the right represent personages of royal 
rank ; the central one is a king who stands with a triumphant 
air, his left hand on his sword and his right hand grasping a rib- 

1 See Justi, Empire of the Persians, 2 See my Zoroaster, pp. 288-294. 

p. 258; de Morgan, Mission Scienti- 3 Consult in this connection the ac- 

ftque, 4. 309, 333, and cf. p. 221, below. companying photograph. 




THE NIMBUS-CROWNED FIGURE 



SASANIAN BAS-RELIEFS 217 

bon-decked coronet, which he receives from or bestows upon a 
second personage of lofty bearing. The latter stands before 
him with the right hand on the chaplet and the left resting 
gracefully upon the hip. Both figures wear crowns, the central 
one a helmet-shaped cap surmounted by the globular adorn- 
ment commonly seen in the sculptures of the Sasanian kings, 
but conventionalized here into a bulb, the other wearing a 
mural crown, beneath which thick, flowing hair is seen. Both 
figures have the characteristic Sasanian head-decoration (sar- 
push) of streamers and veil hanging down behind, and the 
remainder of their ornaments and dress belong also to that 
period. So much for the two royal personages. 

Directly behind the central figure stands a third, differing 
from the other two in that he has his head encircled by a halo 
of rays and his feet resting upon a heavily carved sunflower, 
while he raises before him in both hands a long fluted staff. 
This image the Parsis of India, as well as the Gabars of Persia, 
have taken to represent Zoroaster, and they have made it the 
basis of all the pictures of their Prophet, a matter which I dis- 
cussed with them in Bombay and likewise at Yezd. For this 
reason I shall give further details regarding the sculpture, 
making use largely of my memoranda as I jotted them down 
on the spot, and then I shall briefly touch upon the question of 
the identification of the image. 

The outlines of the figure are good, and in general the pose 
is excellent, the weight being thrown slightly upon the left 
foot. The body is clothed in a tunic-like robe, belted at the 
waist and richly set off at the neck by an embroidered border 
with tassels. The elaborateness of this decoration makes it 
difficult to decide whether a necklace is worn (as appears in 
the case of the central figure) or whether the band at the 
throat is a part of the decorative edging of the garment. The 
tunic itself falls into natural folds, the effect of which is particu- 
larly good in respect to the wrinkles at the elbow; and the 
tunic is decorated quite as much as those of the other two figures, 



218 TAK-l BOSTAN AND KERMANSHAH 

although the belt is simpler than the ornate girdle of the central 
figure and is more like the belt of the figure to the right. 
The nimbus about the head is sharply cut, and a cap of some 
sort, though not surmounted by the globular ornament, covers 
rich hair which can still be seen despite the damaged condition 
of the statue. It looks as if earrings might have been worn, 
but this is uncertain because we have no evidence of such a 
decoration in the case of the other figures. The upper lip is 
covered with a short straight mustache, and the chin is bearded 
with a curly beard. The expression of the face is hard to 
catch because the eyes, nose, and forehead have been destroyed, 
probably at the time of the Mohammedan conquest, by some 
iconoclast that carried out the commandment of the Koran 
against graven images. The staff in the hand has not been 
broken, however, and the grooved lines which run parallel with 
its entire length are plainly visible and resemble the flutings on 
the scabbard of the middle figure. As in the case of the 
others, the peculiar head-dress with crinkled streamers, and the 
wavy scarf flowing down from the shoulders to a point below 
the waist are particularly noticeable. 1 The trousers are flowing 
and heavily fringed, as in the case of the other two figures, and 
the footgear, which appears to include spurs, is quite the same 
as theirs. The sunflower beneath the feet of the image, an 
early symbol of sun-worship, is a triple flower, and the stem 
from which it rises is clearly marked. The size of the entire 
statue is more than that of an ordinary man, being seven feet 
from crown to foot. 2 

1 A crack in the rock has cut the Bostan : Height of the image, 7 ft. 
sar-push in two, and in this fissure a (213 cm.) ; breadth across the shoul- 
plant had taken root, but I had it dug ders, 2 ft. (62 cm.) ; height of head, 
out so that it should not enlarge the including cap, 1 ft. 7 in. (48 cm.) ; 
split and damage the sculpture any length from waist to top of head, 
more than it has done. 3 ft. $ in. (93 cm.) ; length from waist 

2 I add here the more important to sole of foot, 3 ft. 11} in. (120 cm.) ; 
measurements of this figure, which diameter of sunflower, 2 ft. 4f in. 
I took with the assistance of the (72 cm.) ; length of staff, nearly 4 ft. 
Rev. N. L. Euwer, of Hamadan, who (119 c.m). 

accompanied me on my visit to Tak-i 




THE SCULPTURED FIGURE OF THE FALLEN FOE AT TAK-I BOSTAN 



THl 

^ 
Of 

* ' ' :- 



A GROUP OF FOUR FIGURES 219 

A fourth figure completes the group. It is that of a prostrate 
warrior who is trampled beneath the feet of the two victorious 
kings, the one of whom to the right tramples upon his head, the 
other upon his feet. The head of the fallen foeman rests limply 
upon the left arm in a manner intended, apparently, to portray 
death. The helmet, which differs from the others, is decorated 
by a band that seems to be jewelled and is set off by large 
bosses, which might be thought to be pearls in a massive setting, 
but are more probably the studding of the iron-nailed crown of 
the Parthian kings. 1 The face is bearded, although the whiskers 
are not curly like the others. In the case of this figure also 
iconoclasts have perpetrated their vandalistic work. The entire 
trunk of the body has been shattered to such a degree that much 
of the effect of the sculpture is lost; but the remains of the 
sword, although much damaged,- can be traced beneath the 
body, and the scabbard is marked with sharp grooves, which 
show up sufficiently to be noticed and compared with the 
furrows on the sheath of the central figure, which they re- 
semble. The folds of the mantle, moreover, can be clearly 
seen at the arms, neck, and shoulders, and there is a jewelled 
collar or necklace worn about the throat. The legs of the 
fallen enemy have a different sort of buskin from the others, 
but in size the image is not less heroic than they, since it meas- 
ures 7 ft. 2J in. (220 cm.) in length from head to foot. 

The question of the identity of this and the other three 
figures of the group, as I have stated in the book already 
referred to, is still open to discussion. 2 The trend of present 
opinion is against identifying the nimbus-crowned figure with 
Zoroaster, even though opinions may vary as to the other 
three effigies. The latest and best authority on the subject, 
Professor Ferdinand Justi of Marburg, maintains the view which 
he formerly expressed to the effect that it represents the sun- 

1 For the nail-studded helmet of Jasti, Grundr. iran. Philol. 2. 
the Parthian kings Vologases and 515-516. 
Artaban at Naksh-i Rustam, see 2 See my Zoroaster, pp. 289-293. 



220 TAK-I BOSTAN AND KERMANSHAH 

god Mithra, that the central figure is Ardashir I, and that the 
figure to the right is Ormazd, who is bestowing upon him a chap- 
let of victory on the occasion of his overthrow of the Parthian 
dynasty represented by Artaban V, the fallen foe. 1 I must con- 
fess that, although the radiated figure may not portray Zoroaster 
personally, it represents to me an embodiment of the religion, the 
authority of church and state combined in some Magian vizir, 
or priestly chancellor, who blesses the occasion by his presence, 
while the other figures seem to me to represent Ardashir and 
his son Sapor (Shahpur), triumphing over the fallen fortunes 
of the Parthian dynasty represented by Artaban at their feet. 2 
The second group of bas-reliefs is carved in the smaller of the 
vaulted chambers, a few yards beyond the sculptures just 
described, and the identity of the personages is fortunately 
given by -inscriptions adjoining the images. The two effigies 
are carved in the tympanum of the vaulted recess, and the 
vault itself is nearly twenty feet wide by seventeen high and 
twelve deep. The sculptures represent two kings, larger than 
life size, standing side by side. Each monarch holds perpen- 
dicularly before him a sword, blade downward, the point of 
which rests upon the ground, while the hilt is clasped in both 
hands which rest over it. The style of dress, including the 
balloon-shaped headgear, is characteristically Sasanian and the 
identity of each king is made known by the inscription beside 
it. Sapor II, son of Hormizd, stands to the right, and his son 
Sapor III to the left. The former reigned for seventy years 
(A.D. 309-379) during a period which was one of the brightest 
in Sasanian history, though stained by cruel persecution of the 

1 See Justi, Life and Legend of come and trodden under foot. They 

Zarathushtra in Avesta Studies, etc., regard the figure with the halo as prob- 

in Honour of Peshotanji Sanjana, ably Zoroaster, and seem to associate 

p. 157, Strassburg, 1904. The Zoro- it in some way with a sculpture at 

astrians of Yezd gave me a ineta- Balkh, but they did not appear to be 

physical explanation of the sculpture, quite positive on the subject, 
interpreting the prostrate figure as 2 See also my Zoroaster, p. 291. 

representing one's evil nature over- 




THE LARGER AND THE SMALLER ARCH 




TAK-I BOSTAN, OR THE ARCHED GARDEN 



STATUES IN THE SMALLER ARCH 221 

Christians in Persia ; the latter, Sapor III, ruled only five 
years (A.D. 383-388) and was then assassinated in an uprising 
of the soldiers and succeeded by his brother, Bahrain IV, the 
founder of Kermanshah. 1 

The larger arch directly adjoins the smaller recess and is 
nearly double its size, measuring twenty-four feet in width, 
twenty-two in depth, and more than thirty in height. It is far 
more elaborate in its arrangement and more ornate in its deco- 
ration, and contains several sets of carvings instead of a single 
group. Around the outer edge of the arch there is carved in 
alto-relievo a border that resembles a heavy cording faced with 
a notched beadwork of a lotus pattern, which lends a graceful 
finish to the sweep. The base of the arch is balanced on either 
side by heavily carved panels filled with conventional floral 
designs in scroll-form. The framework above the vault is cut 
in such a manner as to resemble miniature battlements, and a 
streamered crescent in high relief crowns the point which would 
correspond to a keystone in mason-work. On either side of 
this emblem hover two winged Victories, like angels, 2 bearing 
coronets and cups, and they are sculptured in Roman style. 
The design throughout shows the influence of Byzantine art 
and is thought to have been executed by Grecian artists from 
Constantinople an inference which would be borne out by 
the statement of Al-Hamadhani, that 4 the sculptor was Fatus 
(or Katus) ibn Sinimmar Rumi (i.e. of Rum, the Byzantine 
Empire), who was the architect also of Khvarnak in Kufah.' 3 

1 See Justi, Grundr. Iran. Philol 2. 2 See p. 142, above. 

525, and for reproductions of the sculp- 3 Al-Hamadhani, ed. De Goeje, 

tures and inscriptions, consult Flandin, Bibl. Geog. Arab. 5. 214. For a more 

Voyage en Perse, Ancienne, 1. pi. 3 detailed description of this arch and 

(from which my photograph is taken) ; the other, see Ker Porter, Travels, 2. 

Ker Porter, Travels, 2. 188; Dieulafoy, 169-195; Curzon, Persia, 1. 560-563 ; 

L'Art Antique, 5. 115, 120, 122 ; and de Morgan, Mission Scientifique, 4. 

(especially for the inscription) consult 304-335. See also Justi, Grundr. Iran. 

de Morgan, Mission Scientijique, 2. Philol. 2. 527. For the name Fatus or 

104-105, pi. xxxi ; 4. 310-318, pi. Katus (perhaps a disguised form of 

xxxvi, and fig. 185. Farhad), see p. 226, n. 2, below. 



222 TAK-I BO STAN AND KEEMANSHAH 

On the outer wall, adjoining the steps to the right, there has 
been cut in recent times a panel, which looks like a church 
window, and in it is carved a modern Persian inscription corn- 
memoratiDg a visit of the late Shah Nasr ad-Din. This, how- 
ever, has nothing to do with the grotto or its sculptures, which 
antedate it by thirteen centuries. These carvings (although 
two centuries later than the bas-reliefs which have previously 
been described, and showing a corresponding advance beyond 
them in the matter of artistic design and execution) are to be 
assigned to the time of the Sasanian king Khosru Parviz. 

Khosru Parviz, with whom we have already become acquainted 
as Chosroes II in the chapter on Takht-i Suleiman, had a reign 
of nearly forty years (A.D. 590-628) and enjoyed great pros- 
perity during a part of it, although his later years were marred 
with misfortune and darkened by defeat. The present sculp- 
tures were made when he was at the zenith of his fortune, and 
they portray happy scenes in his life. As we enter the vaulted 
chamber we see carved on the right a hunting-scene in which 
the king is engaged in pursuit of the deer a favorite sport 
with the Persian kings from the time of Cyrus the Great 
and on the left is chiselled a panel showing Khosru hunting 
the wild boar. Each of the groups is spirited in conception, 
although imperfect in execution, especially in the manner of 
representing perspective ; the panel of the boar hunt, more- 
over, appears not to have been completely finished and polished. 
Nevertheless, both bas-reliefs have a historic interest and an 
archaeological value. Still more important are the two 
great bas-reliefs on the rear wall, and I shall describe these 
with some detail. 

The wall is divided by a carved ledge into two parts, an 
upper and a lower. The lower portion is devoted to a huge 
equestrian statue of Khosru, representing him as mounted upon 
a war-horse of gigantic size. This is Khosru's favorite charger 
Shabdiz, c Black as Night,' a steed praised by Oriental writers 
as the nonpareil of horses. He is caparisoned in heavy war- 



SCULPTURES IN THE LARGER ARCH 223 

trappings, the poitrel, or metal flounce which protects his chest, 
being richly ornamented with tassels artistically chiselled, but 
here again iconoclasts have marred the effect of the figure by 
mutilating the animal's head and legs. The royal rider is of 
superhuman size and is clad in massive armor. A coat of mail 
covers his body, a casque with chain hangings protects his 
head, and a round shield held in the left hand guards his 
breast, while the right hand poises an enormous spear. All 
the details of the king's accoutrements, including the quiver 
at his side, are executed with great care and will bear the 
closest scrutiny of the archaeologist. 

The upper part of the rear wall is occupied by a threefold 
group resting upon the carved ledge directly above the eques- 
trian statue. In the middle of the group stands the king in 
full regalia, holding his sword before him, with the point rest- 
ing upon the ground. With his right hand he receives a chap- 
let decked with streamers, which is presented by a bearded figure 
clad in a long mantle belted at the waist, a tunic, and heavy 
trousers that do not reach quite to the ankle. The third figure, 
directly at the king's right, is that of a woman, who presents 
to the king a garland of victory and pours upon the ground at 
the same time a libation from a vessel held in her left hand. 1 
There is no doubt that in this threefold group, as in the case 
of the equestrian statue, the king is Khosru, but before I dis- 
cuss the interpretation of the scene that is represented I shall 
quote from some of the Oriental writers who described this 
bas-relief and the equestrian statue a thousand years ago. 

The earliest notice of the sculptures that I have been able 'to 
find is the one quoted above from Al-Hamadhani (A.D. 903) who 
devotes to ' Shabdiz and its wonders ' a section in his geographi- 
cal work written about three centuries after Khosru's time. 



1 The photograph which I have of thousand years ago, although some 

the group no longer leaves any doubt later writers have been in doubt on 

that the figure is that of a woman, as the subject, 
stated also by the Oriental writers a 



224 TAK-I BOSTAN AND KERMANSHAH 

1 It is one of the wonders of the world, and its sculptor was Fatus 
(or Katus) ibn Sinimmar Rumi, who was the builder of Khvarnak in 
Kufah.' (He then devotes a paragraph to the praises of the horse 
Shabdiz, and says that if two men came from the farthest ends of 
Iran, Ferghana, and Susa, to see the sculpture they would not have 
any occasion to regret the journey.) 1 

To the same period belongs the more detailed notice by Abu 
Dulaf Misar (A.D. 940), quoted by Yakut. 

'The monument of Shabdiz is one farsakh from the town of 
Kirmasin (Kermanshah). Carved there in the rock is the figure of 
a warrior whose head is surmounted by a helmet and whose body is 
protected by a coat of mail. The workmanship of the armor is so 
perfect that you would declare that the joints of the mail-coat were 
movable and twisted as you examined them. The statue is that of 
[Khosru] Par viz mounted upon his horse Shabdiz, and as a sculpture 
there is nothing comparable with it in the world. In the same arch 
there are also several carved figures of men, women, footsoldiers, and 
horsemen. Before the king 2 there stands a man who looks like a 
workman. Upon his head he wears a round cap and about his waist 
a girdle. In his hand he holds a cup, out of which he pours water 
upon the earth, and it seems to run under his feet. . . . 3 [The third 
figure] is the image of Shirin, the favorite slave of Parviz. ' 4 

A third writer belonging to the same epoch, Ibn Rostah 
(about A.D. 900-950), gives a similar account. 

'It is a distance of three farsakhs from Kirmasin to Shabdiz. 
The latter is an arched recess hewn in the mountain, and in it there 
are pictures of various birds and other representations. In the bosom 
of the vault there is a sculpture of a man wearing a coat of mail, and 
in front of him the effigy of a woman, who, they say, is Shirin ; and 
on the side of the arch is a figure of a man, from under whose feet 

1 Al-Hamadhani, ed. De Goeje, BiU. * Abu Dulaf Misar, cited by Yakut, 
Geog. Arab. 6. 214-216. pp. 345-347. The attribute of the 

2 Here begins the description of the vessel of water would seem to refer 
threefold group above the equestrian to the figure at the king's right rather 
statue. than to the one at his left, as seen in 

8 Here a digression is made by the the photograph of the sculpture and 

author, who quotes at considerable in the published drawings of the bas- 

length Al-Hamadhani's account of relief. 
Shabdiz. 



THE EQUESTRIAN STATUE OF KHOSRU PARVIZ 225 

there flows a stream of water large enough to turn the two stones 
of a mill. On the other side there are about two hundred and fifty 
steps, which are hewn in the rock from below to above the arch.' l 

Contemporaneous with these accounts is the allusion by a 
fourth writer, the well-known Masudi (A.D. 944), who speaks 
of these sculptures of Khosru and Shabdiz. 

' He (i.e. the horse Shabdiz) is carved in the mountain in the dis- 
trict of Kirmasin in the region of Dinavar and Mah al-Kufah. In 
these sciilptures Khosru and the horse are represented beside others. 
Because of the marvellous pictures sculptured in the rock, this place 
is one of the curiosities and wonders of the world. 7 2 

Scholarly opinion is unanimous in agreeing that the central 
figure of the group represents Khosru, although the views differ 
with regard to the interpretation of the other two figures, some 
authorities maintaining that the feminine figure to the right of 
the king is Anahita, goddess of the streams, and that the effigy 
to the left is the god Ormazd. 3 The most probable identifica- 
tion, in part at least, was given more than a thousand years ago 
by the anonymous poet from whom Yakut quotes a verse which 
gives the names of the figures as 4 Khosru, Shirin, and the High 
Priest of the Magi.' The natives to-day, as of old, all regard 
the feminine figure as Shirin. According to some accounts she 
was the daughter of the Byzantine Emperor Mauricius, and a 
Christian, and Khosru fell in love with her at the time when 
he was an exile from Persia at her father's court. 4 The scene, 

1 Ibn Rostah, ed. De Goeje, Bibl. attributed, by implication, to Khosru 
Geog. Arab. 7. 166. I am indebted to Parviz. G. Rawlinson, Seventh Ori- 
Dr. A. Yohannan for this version from ental Monarchy, pi. facing p. 613, calls 
the Arabic. the two figures by the side of the king 

2 Masudi, Les Prairies cT Or, ch. 24, ' emblematic.' 

ed. Barbier de Meynard, 2. 215. For 4 For the view that Shirin, or Sira, 

help with the Arabic I am again in- was a Christian, see Rawlinson, Seventh 

debted to Dr. Yohannan. Oriental Monarchy, p. 497. Accord- 

3 Such a view is held by Justi, Em- ing to Justi, Iranisches Namenbuch, 
pire of the Persians, p. 275 (although p. 302, she was an Aramaean from 
the central figure is there called Sapor Khozistan and is said to have been the 
II) and Grundr. iran. Philol 2. 540, daughter of Mahin-Banu, although the 
where the origin of the sculpture is latter point has been questioned. 

Q 



226 TAK-I BOSTAN AND KERMANSHAH 

it has been thought, represents Khosru receiving back his crown 
at the hands of Mauricius, who had espoused his cause, and 
obtaining at the same time in marriage the Princess Shirin, 
who bestows upon him as a dower a share in her crown. 1 

Popular tradition repeats many legends of Shirin and Khosru, 
and of her admirer Farhad, the royal sculptor, who was attracted 
by her beauty and is said to have executed the group at the 
king's command. 2 The story of the enamored artist's passion 
was a favorite theme with the older Persian poets, but Nizam i 
has carried off the palm by his romantic treatment of it in his 
poem entitled Khosru and Shirin? According to this version of 
the story, Khosru discovered the admiration of Farhad for Shirin 
and took advantage of it to evoke new miracles in marble from 
the inspired sculptor's chisel under promise of receiving Shirin's 
favors as a reward, and to this fact we owe the sculptures at 
Tak-i Bostan, like many others in Persia, according to the 
popular view. Farhad's chef (Tceuvre, however, was to be the 
accomplishment of the herculean task of cutting a channel 
through Bisitun and leading a stream from the other side of 
the valley, after which the longed-for boon was to be bestowed. 
Nizami's verses immortalize the achievement of the task and its 
fatal consequences ; I quote them from a free poetic version in 
English. 4 

' On lofty Beysitoun the lingering sun 
Looks down on ceaseless labours, long begun : 
The mountain trembles to the echoing sound 
Of falling rocks, that from her sides rebound. 
Each day all respite, all repose denied, 
No truce, no pause the thundering strokes are plied ; 
The mist of night around her summit coils, 
But still Ferhad, the lover-artist, toils, 

1 Thus Ker Porter, Travels, 2. 186, disguised form as Fatiis or Katus in 
and after him Curzon, Persia, 1. 561- Al-Hamadhani, quoted above, p. 224. 
662 ; Kiash, Ancient Persian Inscrip- 8 See also pp. 5, 84, 188, above. 
tions, p. 198, Bombay, 1889. * Quoted by Costello, Rose Garden 

2 It might possibly be suggested that of Persia, pp. 84-92, London, 1845, 
the name of Farhad is preserved in a 1887; new ed., pp. 91-97, London, 181)9. 



FARHAD'S TRAGIC LOVE FOR SHIRIN 227 

And still the flashes of his axe between 
He sighs to ev'ry wind, " Alas ! Shireen ! 
Alas ! Shireen ! my task is well-nigh done, 
The goal in view for which I strive alone. 
Love grants me powers that Nature might deny ; 
And, whatsoe'er my doom, the world shall tell, 
Thy lover gave to immortality 
Her name he loved so fatally so well ! " 

The piles give way, the rocky peaks divide, 
The stream comes gushing on a foaming tide ! 
A mighty work, for ages to remain, 
The token of his passion and his pain. 

As flows the milky flood from Allah's throne, 
Rushes the torrent from the yielding stone ; 
And sculptured there, amazed, stern Khosru stands, 
And sees, with frowns, obeyed his harsh commands : 
While she, the fair beloved, with being rife, 
Awakes the glowing marble into life. . . . 

Around the pair, lo ! groups of courtiers wait, 
And slaves and pages crowd in solemn state ; 
Prom columns imaged wreaths their garlands throw, 
And fretted roofs with stars appear to glow ; 
Fresh leaves and blossoms seem around to spring, 
And feathered songs their loves are murmuring ; 
The hands of Peris might have wrought those stems, 
Where dewdrops hang their fragile diadems ; 
And springs of pearl and sharp-cut diamonds shine, 
New from the wave, or recent from the mine. 

" Alas, Shireen ! " at every stroke he cries ; 
At every stroke fresh miracles arise. 
" For thee these glories and these wonders all, 
For thee I triumph, or for thee I fall ; 
For thee my life one ceaseless toil has been, 
Inspire my soul anew Alas, Shireen ! " 

The sequel of this rapturous devotion is a tragic one; the 
keynote is struck in two lines : 

1 Ah, hapless youth ! Ah, toil repaid with woe ! 
A king thy rival, and the world thy foe. ? 



228 TAK-I BOSTAN AND KERMANSHAH 

Khosru, in his anxiety to be relieved from fulfilling his promise 
to the sculptor, resorted to the counsel of an old hag, who 
engaged, for a rich reward, to free the monarch from the neces- 
sity of redeeming his pledge. Intrusted with the task, this 
messenger of evil portent proceeds to the lofty rock where 
Farhad is employed, and hoarsely whispers her fatal falsehood. 

' " Cease, idle youth, to waste thy days," she said, 
" By empty hopes a visionary made ; 
Why in vain toil thy fleeting life consume 
To frame a palace ? Bather hew a toinb. 
Even like sere leaves that autumn winds have shed, 
Perish thy labours, for Shireen is dead ! " 

He heard the fatal news no word, no groan ; 
He spoke not, moved not, stood transfixed to stone. 
Then, with a frenzied start, he raised on high 
His arms, and wildly tossed them towards the sky ; 
Far in the wide expanse his axe he flung, 
And from the precipice at once he sprung. 
The rocks, the sculptured caves, the valleys green, 
Sent back his dying cry " Alas ! Shireen." ' 

Even when robbed of its poetic garb the story seems to have 
an element of truth as its basis, and the legend lends a romantic 
interest to the ' Villa of Shir in ' and its sculptured halls. 

Another monument of Khosru, a platform on which royal 
assemblies were held, was located in the vicinity of Kerman- 
shah, according to Yakut, and probably not far from Tak-i 
Bostan. I shall quote Yakut's own words describing it : 

' Near Kermanshah is situated the famous platform where Khosru 
Parviz received homage in royal assembly from the kings of China, 
Turan, India, and Byzantium. It is a quadrilateral, a hundred 
cubits long and a hundred cubits broad, built of dressed blocks 
of stone, skilfully matched and joined together by iron clamps so 
closely that they look like a single piece.' l 

This description should be sufficient to allow an identification 
of the location of the platform to be made, as some remains of 

i Yakut, p. 438. 



THE ROAD TO KERMANSHAH 22$ 

the structure must exist, but I am unable to give any precise 
information on the subject myself, although there are ruins 
near Tak-i Bostan, 1 and the natives point out some mounds -as 
the site of Khosru's palace. 2 I noticed also some marble walls, 
and columns on the road between Kermanshah and Bisitun. 3 

I have made my description of Tak-i Bostan somewhat more 
detailed than otherwise might be expected, because of the his- 
toric associations connected with the place, and we may now 
resume our journey to Kermanshah, which lies about four miles 
to the south-southwest on rising ground that commands a con- 
siderable view over the plain. 

To reach the city I found it was necessary to take a some- 
what circuitous route, as the Karasu River, though ordinarily 
narrow, was now swollen by floods and could be forded only at 
a particular point. Leaving the miniature lake and its associa- 
tions with the name of Shirin, our cavalcade proceeded on its 
way, and we soon were approaching a modern three-story build- 
ing, near the river, which looked in the distance like a European 
apartment house suddenly transported to Persia, where three- 
story houses are a rarity. I found it was a ducal palace belong- 
ing to the family of one of the former governors of Kermanshah, 
Imam Guli Mirza, known as Imad ad-Daulah, and hence the 
building was called the Imadiah. It is now going to rack and 
ruin, because the heirs of the original owner have failed to keep 
up this villa and its grounds, in order to avoid the expense it 
would entail to entertain in it with lavish hospitality each new 
governor that assumes the administration of Kermanshah. 

Hardly had we crossed the Karasu and ridden a mile farther 
toward the city than we were completely enveloped in a snow 

1 See the plan of Tak-i Bostan in Khosru ' was pointed out to me near 
Flandin and Coste, Voyage en Perse, Bisitun, but this was too far from Ker- 
Ancienne, 1. pi. 1. manshahto answer to Yakut's descrip- 

2 Information from Mr. H. L. Ka- tion. I am not unmindful that there- 
bino. are ruins at Kasr-i Shirin, Sar-i Pul r 

8 See likewise Wilson, Handbook and Takht-i Girrah, cf. de Morgan, 
of Asia Minor, p. 327. A 'Takht-i Mission Scientifique, 4. 335-357. 



230 TAK-I BOSTAN AND KERMANSHAH 

squall which was severe enough to blot out the recollection 
that we had been lunching only a short time before in a garden 
of "springtime, and to leave only a memory of the snowstorms 
produced by the black art of the sorcerers in the Shah Namah. 
But it soon passed over, and in another half hour the tin-cov- 
ered minarets of the mosques of Kermanshah and the lofty 
towers of the Governor's Palace shone out in the afternoon 
sun. 

Kermanshah, or Kermanshahan, as it was more generally called 
in former times, is a place of considerable antiquity as well as the 
modern capital of the district of the same name. 1 Tradition 
ascribes the founding or the re-establishing of the town to the 
Sasanian king Bahrain IV (A.D. 388-399), who had been ruler 
of Kerinan (^Karmdn-shdJi) before he came to the throne, and for 
this reason gave that name to Kermanshah, although Yakut 
assigns the founding of the city to a date a century later, 
ascribing it to Kobad I, son of Piruz (A.D. 488-531). 2 It is 
probable that the city is much older than either of these dates 
and that it may occupy the site of the ancient Kambadene, 
mentioned in the itinerary of Isidor of Charax. 3 Little is 
known in detail regarding the history of the city for a long 
time after the Sasanian period. Although mentioned by the 
Arab geographers Al-Hamadhani, Ibn Rostah, and Yakut, 
already referred to, no allusion is made to it by Pietro della 
Valle, who must have passed near it on his journey through 
this plain three hundred years ago (1617). 4 There are native 
records, however, including poetical accounts in Kurdish, which 
give the history of some of its wars in the eighteenth century, 
and since the nineteenth century the town has become better 



1 The Arab pronunciation of the 8 Gk. Kayu/foS^v^, see Isidorus Chara- 

name of the town varies between cenus, Mansiones Parthicae, 6, ed. 

Karmasin and Kirmlsln. Miiller, Paris, 1855, 1882 ; and cf. de 

a Yakut, p. 438. For the tradition Morgan, Mission Scientifique, 2. 100. 
about Bahram IV, see Justi, Grandr. 4 For the itinerary of Pietro della 

iran. Philol. 2. 626-526, and for the Valle's journey in this region, see 

reign of Kobad (Kavadh), 2. 631. the edition of Pinkerton, 0. 16 seq. 



THE CITY OF KERMANSHAH 231 

known to the West through trade and travel, although as 
regards familiarity its name can bear no comparison with Tehe- 
ran and Isfahan. Many persons, in fact, know of Kerman- 
shah only through the rugs for which the city and its vicinity 
once were famous, even if the rug manufacture is almost a lost 
industry now in Kermanshah itself and the carpets which are 
exported through its customs to-day come mostly from other 
parts of Persia and are merely shipped by way of this dis- 
tributing-centre. 1 

Commercially the city of Kermanshah is favorably situated, 
as it lies on the main caravan route between Persia and Meso- 
potamia, being nearly equidistant from Teheran and Baghdad, 
two hundred and twenty miles from the latter and two hundred 
and fifty miles from the former. The town enjoys the advan- 
tages of a busy trade, especially on commission, and its popu- 
lation is now reported at fully sixty thousand, the inhabitants 
being largely of Kurdish blood, besides Persians, Turks, 
some Jews, and a few Christians. 2 The municipal adminis- 
tration is in the hands of three magistrates (kadkhuda), each 
of whom presides over one of the three wards into which the 
city is divided, and is accountable, through a number of higher 
officials, to the Governor of Kermanshah appointed by the 
Shah. 

In area the extent of the town is considerable, as it measures 
about four miles in circumference, and its circuit was formerly 
enclosed by walls, although these have now disappeared, except 
that one or two of the towers have been built into the walls of 
dwellings, and traces of the moat are visible where it has not been 
completely filled up. The five city gates have been preserved 
in name at least, as their names are still employed to designate 
the several quarters where the main roads enter the town. 

1 Just as ' Hamburg ' grapes and kindness of Mr. H. L. Rabino, Agent 

' Astrakhan' furs are so named from of the Imperial Bank at Kermanshah, 

the place of shipment. and to the British Consular Reports, 

2 For valuable information regard- Persia, no. 590, miscell. ser. ; no. 3189, 

ing Kermanshah I am indebted to the annual ser., London, 1903, 1904. 



232 TAK-I BOS TAN AND KERMANSHAH 

Architecturally the city has little of interest to attract the 
traveller, and most of its buildings are of comparatively recent 
date. There are a number of public squares and buildings, 
but they are of minor importance. Among them may be men- 
tioned the Governor's Palace, whose high towers overlook the 
Top Meiddn, or 4 Artillery Square.' In the midst of this square 
is a reservoir, and around the plaza are shops adjoining the 
bazaars. The arsenal itself is behind the palace, and to the 
south is another square called Meiddn-i Sarldz Khdnah, or 
'Barrack Square,' because the soldiers' quarters are built 
around it and it serves as a parade-ground. There are several 
mosques in different parts of the city, but none of them are 
ancient or especially renowned. A bank, custom-house, post 
and telegraph office, and about thirty baths make up the rest 
of the quota of public buildings ; there are some private houses 
of the finer sort and a number of handsome gardens and villas 
in the environs, the property of wealthy owners. 

The city is well supplied with caravansarais, and they are 
usually crowded with merchants or with pilgrims on their way 
to and from Kerbela. The principal hostelry was full on the 
afternoon when I arrived, so that I had to seek elsewhere for 
lodgings. I met with the same experience at the next, and 
again at the next, riding for more than an hour through 
crowded streets and bazaars, jostling against camels and pack- 
mules, whose load nearly tore one of my riding-leggings to 
pieces, stopping only long enough to repair damages and 
sample a baker's supply of sugar cakes, which were really 
excellent, then proceeding once more. At last I found shel- 
ter in an unpretentious manzil, the owner of which was an 
obliging person, and he succeeded in making me fairly com- 
fortable. My enforced peregrination had at least this in its 
favor, that it gave me an opportunity to gain some idea of the 
commercial activity of Kermanshah. The bazaars seemed to be 
carrying on a prosperous trade in the products of the district, 
grain, wheat, barley, fruit, gum, and also opium, besides doing 



POINTS OF INTEREST IN KERMANSHAH 233 

a fair business in goods which are transported through the city, 
a traffic which includes also numerous imports from abroad, so 
that I was able to stock up again with several articles of foreign 
manufacture which I found I needed on the journey. 

I had expected to remain an extra day at Kermanshah, espe- 
cially as I had met with kind hospitality at the Imperial Bank, 
but I changed my plans when I learned that the two other 
scholars whom I thought I might meet and have join in the 
work upon the Behistan rock had been prevented from com- 
ing. I felt, therefore, that I must return to the rock at once 
and accomplish as much more as I could in the limited time at 
my disposal before leaving for the south of Persia. Accord- 
ingly I started shortly after sunrise the next morning and 
galloped my horse, Rakhsh, most of the twenty-one miles back 
again to Bisitun. After taking a hasty meal I proceeded 
immediately to the height, and within an hour after noon I had 
again ascended the cliff and was busy at the work, which I 
continued that day and the next with all the intensity of 
application of which I was capable, and with the results that I 
have already described. 



CHAPTER XVI 

THE GREAT RUINED TEMPLE OF THE PERSIAN DIANA 

AT KANGAVAR 

4 1 shall offer unto the holy Ardvi Sura Anahita, goddess of the heavenly 
streams, pure and undefiled, a goodly sacrifice accompanied by an oblation. ' 

AVESTA, Yasht 5. 9. 

KANGAVAR is a small town of great antiquity, lying directly 
on the route between Bisitun and Hamadan, and it is the site of 
some important ruins which I shall describe, as they are those 
of a temple of the Ancient Persian Diana. On my journey 
outward to Bisitun I knew that there were some ancient 
remains to be seen at the place, but in my anxiety to reach 
the inscription of Darius I had no time to visit them, and I 
waited till I should be able to inspect them on my return 
journey to Hamadan. Accordingly, I mounted my horse and 
started with Safar, Shahbas, and the rest of my caravan on 
the road to Hamadan by way of Kangavar, the same route 
I had traversed eight days before. 

As I rode along, my attention was attracted by a large land- 
tortoise, the first I had seen in Persia; it had been tempted 
out of its winter quarters by the warm spring sun and was 
slowly crawling along by the side of the trail. The Persians 
call the tortoise 'stony-back' (sangl-pmht), but the ancient 
Zoroastrians named it zairimyanura or nicknamed it zairi- 
mydka (a word of uncertain meaning) and looked upon the 
tortoise as one of Ahriman's creatures and therefore to be 
destroyed. 1 Happily in this respect the harmless creature is 

1 Avesta, Vd. 13. 6 ; see also (kasy- dure'; Bartholomae, Air. Wb. p. 1682, 
apa) Vd. 14. 5. Darraesteter, Le ZA. as ' des Glieder (oder Zehen) in einem 
2. 195, n. 8, gives the meaning of festen Gehaus stecken. 1 
zairimyahura as 'qui de"vore la ver- 

234 




A PERSIAN SHEPHERD 
(Near Sahnah) 




A TYPICAL CARAVANSARAI 



OF THE 

UNIVERSITY 

Of 




ON THE ROAD TO SAHNAH 235 

no longer under a ban, and my Mohammedan attendants 
allowed it to pass unmolested. Sunset found me entering 
the hamlet of Sahnah, riding through lanes lined with rows 
of plum and apple trees in full blossom. I directed my cara- 
van to proceed to the same manzil at which we had stayed on 
the outward journey, and hardly an hour had passed before I 
was asleep in the bdld-khdnah, or upper room above the entrance. 
If I had known at the time that a tomb, fabled to be that of 
the legendary king of Iran, Kei Kaus, is located in a gorge back 
of Sahnah, I should have engaged some of the natives to take 
me, with torches, to visit the chamber and draw me up to 
inspect it, or should have waited an extra day to examine it. 1 
But Sahnah meant for me, on my two visits, only a halting-place, 
as it does for the Kerbela pilgrims whose passing to and fro is 
a source of revenue to its thousand or more inhabitants, so I 
arranged to continue my march early next morning. 2 

The hands of my watch indicated precisely 6.00 A.M. when 
I gave the signal to start, and our slow-moving procession filed 
out beneath the mud portal of the caravansarai and headed 
eastward again toward Hamadan. 

In an hour and a half we reached a pretty village which I 
remembered noticing particularly, on the outward journey, 
because of its green groves and orchards, its rich grass, and 
abundant water. The frogs in the pools were croaking lustily, 
no longer in fear of the old Zoroastrian law (long since passed 
into oblivion) which accounted them noxious animals and 
regarded it as a meritorious act to kill thousands of this brood 
of Ahriman. 3 The merry chorus of their voices reminded me 
of the Frog Hymn in the Rig Veda, 4 as each tried to out- 
croak the other, but the inharmonious music was soon lost 

1 Sahnah is mentioned by Ibn see Flandin and Coste, Voyage en 
Haukal, p. 167, Yakut, p. 305, and Perse, Moderne, pi. 75 b, and Texte, 
Pietro della Valle, Viaggi, 1. 440 ; p. 11. 

Travels, ed. Pinkerton, 9. 17 ; but none 3 Vd. 14. 5 ; 18. 65, 73. 

of them allude to the tomb of Kaus. 4 See Rig Veda, 7. 103. 1-10. 

2 For a sketch of the ' Tomb of Kaus, 1 



236 THE RUINED TEMPLE AT KANGAVAR 

in the distance, and before long we were ascending a ridge of 
rolling hills and after that descending into a plain flooded with 
sunlight. 

I halted for a few minutes here to speak to a Persian shep- 
herd who was grazing his flocks along the rich plain. His 
short-sleeved sheepskin coat, shoes of rough hide, heavily 
rolled turban, and shepherd's staff gave me an impression of 
what the pastoral life must have been in early Media. It was, 
perhaps, to such a shepherd, in these very mountains, that King 
Astyages committed the infant Cyrus with orders that the 
child should be exposed to die, if we are to believe the ancient 
legend, and possibly this very shepherd's wife resembled the 
fictitious Spaka who reared the foundling to become a shep- 
herd lad and king. 1 

We rode forward through a fertile plain and reached Kanga- 
var before noon. The town is about thirty-two miles distant 
from Bisitun, or four farsakhs from Sahnah, and is now a settle- 
ment of about eighteen hundred houses, 2 the lodging of some 
eight thousand inhabitants. The main street, which winds 
up and down the elevation on which the town lies, is lined with 
shops and booths of merchants trading especially in wheat, 
barley, cotton, and pears, and is bordered on either side with 
halting-places to offer a manzil to the heavily laden trains of 
camels and asses on their way to Kermanshah and Baghdad. 

Kangavar is a place of genuine antiquity ; it was known, 
under the name of Konkobar, to the Greek geographer Isidor 
of Charax in the first century of our era ; 3 and my attention 

1 Herodotus, History, 1. 107-114. D'Anville, Compendium of Ancient 

2 Information from Mr. H. L. Rabino. Geography, pt. 2, p. 460, London, 
The Italian traveller Pietro della Valle, 1791. Since that time much has been 
who stayed in the town Jan. 20, 1617, written about the site ; see Rawlinson, 
described Kangavar as a large place JUGS. 9. 112; Buckingham, Travels 
('grossa terra chiamata Chienghieuer,' in Assyria, Media, and Persia, pp. 
Viaggi, 1. 440 ; Travels, ed. Pinkerton, 150-154, London, 1829; Masson, Illus- 
9. 17). trations of Isidorus of Charax, in 

8 The first scholar to identify Kanga- JRAS. 12 (1850), pp. 97-124; Ker 
var with 'Concobar,' I believe, was Porter, Travels, 2. 139-144; Texier, 



THE TOWN OF K AN G AVAR 237 

was at once attracted by certain peculiarities of its site. The 
low hills and mounds which surround it, some of them capped 
with buildings erected on the foundations of older structures, 
immediately suggested to me an etymology for Kangavar. The 
name Kangavar, or Kankivar, Gk. Kory/coftdp, may be derived 
from a hypothetical Avestan form *Kahha-vara, c enclosure of 
Kanha,' the first element of the compound being probably 
a proper name, Kahha, as in Avestan Kahha-daeza and 
Firdausi's Kang-diz, and the second element, vara, ' enclosure,' 
being cognate with the designation of the Vara of Jamshid. 1 
The mounds and elevations which surround the acclivity on 
which Kangavar is built, make the attribute a fitting one and 
the suggested etymology probable. 

Among the buildings which attract notice at Kangavar are 
one or two mosques and a brick citadel, but they are of minor 
importance when compared with the ruins of what must once 
have been a magnificent edifice in the heart of the town. Here, 
on the principal thoroughfare and near a large caravansarai, are 
the remains of a wall of white marble blocks of mammoth size, 
hewn with precision, and crowned with broken columns and 
pilasters which show the outline of a grand enclosure of build- 
ings. The style of the architecture is noticeably Greek, 
and I made a series of memoranda concerning the ruins, 
although I did not know at the time that they had been sub- 
jected to a detailed examination by others. I shall give my 
notes as I took them down, and supplement or correct them 
from other sources when occasion demands. 

So far as my observations allowed me to judge, the ruins 
seemed to represent the remains of two large buildings, the 
one to the northwest, lying directly on the main street, and 

Description deVArmenie, la Perse, etc., 207 ; Curzon, Persia, 1. 51, n. 1 ; Mar- 

1. 160 seq., pis. 62-64; Flandin and quart, Erdnsahr, p. 24; and K. A. 

Coste, Voyage en Perse, Ancienne, Floyer, Unexplored Baluchistan, pp. 

1. pis. 20-23, Moderne, pis. 72-74, 424-425, London, 1882. 
Texte, pp. 11-14; Dieulafoy, i: Art l See p. 126, above. Kahha was 

Antique de la Perse, pt. 5, pp. 7-8, 11, located somewhere in eastern Iran. 



238 THE RUINED TEMPLE AT RANG AVAR 

the other at some distance to the southeast, situated on the 
edge of a declivity. I find that authorities such as Ker Porter, 
Texier, Flandin and Coste, and Dieulafoy, to whose judg- 
ment I should defer in such a matter, see in the granite 
blocks which compose the structure the remains of a single 
great platform with a peristyle, in the midst of which stood 
the chief edifice. 1 Ker Porter, for example, who visited the 
ruins in 1818, regarded the whole enclosure as a temple pre- 
cinct whose walls formed a huge rectangular terrace, three hun- 
dred yards square and crowned with a colonnade. 2 Flandin's 
drawings and Texier's sketches, which apparently served as 
the chief source of Dieulafoy's information, give a plan of the 
possible arrangement of the platform and the temple precinct. 
To speak with positive certainty as to details, however, would 
only be possible after a careful re-examination of the ground, 
and it would also be necessary to make clearings and excava- 
tions, since a mass of modern buildings and dilapidated struc- 
tures has been built over the ruins, or crowded in among 
them, in order to take advantage of some of the ancient 
material still standing. 

The supposition that the original structure was an extensive 
platform, on which various superstructures were raised, is borne 
out by an allusion to Kangavar in Yakut, about A.D. 1220. 
This Arab writer says that the place was called also Kasr-i 
Shirln, 'Castle of Shirin,' after Khosru's favorite, 3 but more 
often Kasr al-Lasus, 'the Robber Castle,' because the Arab 
army which invaded Persia after the battle of Nahavand lost 
some pack-animals here, which were stolen by robbers; it is 

1 Compare Ker Porter, Travels Flandin and Coste, Texte, p. 13, who 

(1822), 1. 141 ; Texier, Description de give 217.93 m. by 229 m. ; Texier, op. 

VArmenie, la Perse, etc. (1842), 1. 160- cit. 1. 161, gives 202 m. by 172 m. 

162, and pis. 62-68 ; Flandin and Coste, 2 The Arab geographer Al-Hama- 

Voyage en Perse, Ancienne, 1. pi. 21 dhani (ed. De Goeje, 5. 267) says that 

(reproduced below), Texte, p. 13, * in all the world there are no columns 

Moderne, pis. 72-73 ; Dieulafoy, V Art more wonderful than those at Kasr al- 

Antique, pt. 5, pp. 8, 207. For meas- Lasus (i.e. Kangavar).' 

urements of the platform, compare 8 See p. 225, n. 4, pp. 226-228, above. 



APPEARANCE OF THE RUINS 239 

more likely, however, as Mustau.fi says, that the name was given 
it because the place was later infested by bandits, a reputation 
which it long retained. 1 Yakut describes the place as follows : 

' The Kobber Castle is a very remarkable monument, and there is 
a platform some twenty cubits above the ground and on it there are 
vast portals, palaces, and pavilions, remarkable for their solidity and 
their beauty.' 2 

The part of the foundation which is best preserved stands 
at the northwest corner. A solid retaining-wall can there be 
seen, twelve or fifteen feet high, running north and south for 
more than seventy feet and forming what I took to be the 
base of a single building. The northern wall, which extends 
at right angles eastward from this point, is equally massive, 
being composed of granite blocks, some of them more than 
seven feet long and four feet high, 3 and broad in proportion, 
although it was not easy to measure the stones, as the wall is 
partly buried beneath earth and debris. The rampart is capped 
with a heavy coping which forms a stylobate to sustain what 
once must have been an imposing colonnade. 4 Three of these 
columns, each nearly six feet in diameter, were still standing 
on the cornice of the northwestern wall when I saw it, having 
been preserved by being built into the side wall of a modern 
building, as shown also in Flandin's picture (which I have repro- 
duced), while a fourth truncated shaft at the extreme upper 
angle of the stylobate, where the wall turns eastward, was 
standing and could be easily measured because it was quite 
detached. 5 Several of the pillars have fallen since the time 

1 Yakut, p. 451 ; cf. also pp. 450, 2 Yakut, p. 451. 

495; for the robbers cf. Mustaufi, cited 3 210 cin. by 130 cm. 

by Barbier de Meynard, Diet. geog. p. 4 For a sketch of a section of this 

451, n. 1, and Le Strange, JRAS. 1902, cornice, see de Morgan, Mission Scien- 

p. 511 ; furthermore, Ibn Haukal, p. tifique, 2. 139. 

166 (Kasr-i Duzdan ' Robber Castle '), 6 The height of this drum was over 

and Masson, JRAS. 12 (1850), p. 116, six feet, and its diameter nearly six 

which bears out the idea that the place feet also, or, more exactly, 170 cm. 

was infested by freebooters. by 160 cm. 



240 THE RUINED TEMPLE AT K AN G AVAR 

of Ker Porter (1818), who speaks of pedestals of eight col- 
umns as still surmounted by the chief part of their shafts in 
good preservation, thus evidently not including the base of 
the rectangular shaft adjoining the column at the upper end, 
which is shown in the drawings and plans of the Frenchmen 
Texier (1839-1840) and Flandin and Coste (1839-1841), and 
of the Englishman Masson (1845). 1 

Following my guide eastward along the elevation and then 
turning southward among the hovels and rubbish, I found a 
large collection of massive stones near the southeast corner. 
A mass of huge blocks was tossed about in confusion, as if a 
building had collapsed, but I was able to trace the general out- 
line of a wall running for about a hundred feet north and 
south, as shown in my photograph. The stones were of the 
same large size as at the northwest corner, and one granite 
drum, which had fallen down the slope, measured nearly five 
feet by eight. 2 The base of another column was still to be 
seen in its place in the line of the wall, and near it a third 
pedestal rising two feet above the ground. 3 Within the enclos- 
ure where the main edifice must have stood was a part of a 
column (shown in the middle of my photograph), which had 
apparently been set partly on a foundation of stone and mortar, 
so that it looked like an altar for libations, as it was slightly 
hollowed out, but it may have been simply an overturned capi- 
tal. I was inclined to the former view because I believed, 
when examining the ruins, that they were the remains of a 
temple of Anahita, the great Persian goddess of the heavenly 
streams, whom the Greeks identified with Artemis, or Diana, 

Compare Ker Porter, Travels, 2. era building between them still intact). 

141 ('pedestals of eight'); Flandin Dieulafoy's plans (I? Art Antique de la 

and Coste, Voyage en Perse, Ancienne, Perse, 5. 8-9) are after Flandin. 

1. pis. 21, 22, Moderne, pis. 72, 73 (8 + 1 * More exactly, 230 cm. by 144 cm. 

columns, including the pilaster) ; Tex- For a sketch of the mass of 

ier, Description de VArmenie, la Perse, columns as they lay about half a 

etc. 1. pis. 64, 65, [66] (8+1 columns) ; century ago, see Flandin and Coste, 

Masson, JJRAS. 12.pl. p.117 (8 + lpil- Voyage en Perse, Moderne, pi. 74 b, 

lars, with most of the wall of the mod- and Texier, Description, 1. pi. 68. 



A SHRINE OF ANAHITA 241 

and whose worship was widespread throughout Iran in the 
time of Artaxerxes Mnemon, in the fourth century before 
Christ. 

On gaining access to my books I was able, by positive evi- 
dence from the classics, to substantiate the view that I held 
regarding the identity of the temple, but I find that I have been 
partly anticipated by others. 1 The Greek geographer Isidor 
of Charax, who entered Media by this route in the first cen- 
tury A.D. and kept detailed notes of his itinerary, which he 
published under the name of 'Parthian Stations,' mentions 
Kangavar as Konkobar, and alludes to its temple, sacred to 
Artemis. His laconic note reads : 

' Three schoeni (eight or nine hiiles) from the frontier of Upper 
Media is the city of Konkobar, where there is a temple of Artemis. 
Three sclioeni beyond this is Bazigraban, which is the place of receipt 
of customs. Four sclioeni thence to Adrapanan, a palace in the ter- 
ritory of Batana (i.e. Ecbatana), 2 destroyed by Tigranes the Arme- 
nian. Twelve schoeni thence is Batana (Ecbatana) the capital of 
Media, the treasury, and the temple where they sacrifice constantly 
to Anai'tis. Then after that there are three villages, in each of which 
is a station/ 3 



1 See the names referred to above, te/>6, <rxoivoi y'. Elra Bafrypdpav, 8 
p. 236, n. 3. t<rri Tt\6viov, exoivoi y'. Efra efs 

2 Lit. 'of those in Batana' (i.e. 'Adpatrdvav rd pa<rl\eia ruv v Eard- 
Ecbatana). vots, & Ttypdvijs 6 'Apptvios icadetXe, 

3 Isidor of Charax, Mansiones Par- <rxoivot 5'. Erro BdVaw, p.-ijTpbrro\t<i 
thicae, 6. For convenience I add the Madias Kal 0i]<ravpo<pv\dK(,ot> teal iepbv, 
Greek text of Isidor's entire itinerary 8irep 'Avatrtdos del 66ov<rn> - <rxoit>oi if?. 
from Cambadene (Kermanshah ?) and Elra e%rjs rpeTs KW/XCU, tv als ffra.dp.6s. 
Behistan (misvvritten as Baptana for 7. 'Evrevdev ['PcryiafTj] M^S^a, ffxoTvot 
Bagistana Greek 7TT for riCT) to ["?'] "Ev y KU/J.O.I i', 7r6Xeis 5^ e'. 'Airb 
Konkobar and Ecbatana, thence to ffxolvwv f ' 'Pdya Kal Xdpa, $>v p*yiffTi) 
Rhaga (mod. Rei) and the Caspian T&V xard T^V M?;5^aj' 17 'Pdya. Ets 5^ 
Gates. The passage reads : 5. 'Evrev6ev TT)V x^P aKa irpwros f3a<ri\et>s Qpadrys 
K.afj.fiad'rjvr), TJTLS Kar^et ffxolvovs Xa', iv robs MdpSovs <j}Ki<rev earn? virb rb 6/)0$, 
rj K&fMi e', tv ah crra^s, ?r6Xts 5^ Ed- 8 KaXctrat Rao-Trios, d<p' o5 Kdo-mai, 
irrava eir' 6povs Kei^vrj fvda 2e/j.Lpd/j.i- irtXai. (See Isidorus Characenus, 
5o$ aya\p.a Kal <TT^\rj. 6. 'EvrevBev ij Mansiones Parthicae, in Geographi 

17 di/w, (?xoTvoi X?j' Kal apx?rat. Graeci Minores, ed. C. Miiller, Paris, 

1855,1882.) 



242 THE RUINED TEMPLE AT KANGAVAR 

The ruined temple, therefore, was one dedicated to Ardvi 
Sura Anahita, as goddess of the streams. The situation at 
Kangavar must have been a suitable one for a sanctuary 
devoted to her worship, for within the town itself is a cas- 
cade which pours its waters down into the plain to be lost 
ultimately in the Garnasiab, and may therefore be described 
in the words of the Avesta itself as 4 a holy stream enriching 
life, enriching the herds, enriching property, enriching wealth, 
enriching the whole country.' l 

My idea that the temple was erected in Achaemenian times 
and may have been founded by Artaxerxes II, is not in agree- 
ment with Dieulafoy, I find, who argues that the remains show 
a confused Greek style, that the columns are bastard Doric, 2 
that the architecture has nothing in common with the Per- 
sepolitan, which shows Egyptian characteristics, and that we 
must conclude to assign the temple to the Parthian period, 
ascribing it to some one of the Arsacid kings, all of whom 
were strongly under Hellenic influence. 3 If this assumption 
be correct, I should regard it as an addition to our knowledge 
of the Zoroastrian religion during the Parthian period, regard- 
ing which our information is in many respects scanty. In any 
event Kangavar offers a good field for archaeological research, 
and I believe that scientific excavations in the vicinity, as at 
Tak-i Bostan and throughout the valley, would yield important 
results. 

Continuing the journey toward Hamadan, I stayed over 
night, for the second time, at the small walled village of Asad- 
abad, or Saidabad, four and a half farsakhs (about twenty 
miles) distant from Kangavar. 4 This settlement lies at the 

1 Yt. 5. 1. Dieulafoy, UArl Antique de la 

2 For specimens of the bases and Perse, pt. 5, pp. 7, 8, 207. 

capitals of the columns, see Ker Porter, 4 The distance is given by Masson, 

Travels, 1. pi. 43 c ; and compare also JRAS. 12. 99 (after Webb) as twenty- 

the drawings of Flandin and Coste, two miles ; Curzon, Persia, 1. 57, says 

Voyage en Perse, Texte, p. 13, and the (approximately) twenty-three miles, 
allusions to the Doric order. 







THE RUINED TEMPLE AT KANGAVAB 

(Flandin's drawing of the northwest wall. The modern buildings to the left of the 
picture have disappeared) 




RUINS OF THE TEMPLE AT KANGAVAR 
(Part of the southwestern wall with a modern building in the background) 



SITE OF AN ANCIENT CUSTOM HOUSE 243 

base of the great mountain ridge whose steep and rugged 
heights make a formidable barrier to surmount before reach- 
ing the city of Hamadan. Asadabad is the regular halting- 
place for all caravans that go by this ancient route, and I believe 
that this not inconsiderable village represents, in location at 
least, the Bazigraban, or Custom House, mentioned by Isidorus 
in the passage already quoted. The etymology of the name 
Bazi-grabdn (Gk. ~Bafyypd/3av) immediately becomes clear when 
we restore the word to its probable form in Ancient Persian, 
*Bdji-graband, ' tribute-taking, toll-collecting,' indicating the 
place where the customs dues were levied, somewhat like the 
Modern Persian bdj-gdh, 'toll place.' 1 

Asadabad, being situated in a plain at the base of the spurs 
of Mount Alvand and watered by the streams that descend from 
the great ridge, has a fertile soil and a temperate climate, and 
it was once a place of some renown. Yakut, writing seven 
hundred years ago, says that it was formerly the residence of 
the son of the Sasanian king Khosru Parviz (A.D. 590-628), 
although the monarch himself resided for the most part at 
Kangavar. The Arab geographer narrates an amusing legend, 
which he characterizes, however, as poetic fiction and a 'lie,' 
to the effect that ' Khosru's Kitchen ' was located midway 
between Asadabad and Kangavar; and whenever the king 
dined, a long line of pages 'passed the dishes from hand to 
hand ' over the entire distance. His son observed the same 
custom when living at Asadabad. The viands, Yakut says, 
must have been cold when they reached the king, even if borne 
on eagle's wings, but he adds that we are, perhaps, to under- 
stand that ' Khosru's Kitchen ' (maibakh) was merely the 
place which served as headquarters from which the royal 
cuisine was stocked. 2 

My journey over the desolate pass from Asadabad was made 

1 Cf . also Spiegel, Altpers. Keilin- 2 See Yakut, pp. 34, 536, whose 
schriften, 2d ed., p. 233. Cf. likewise description and comments are worth 
OP. Pati~\grabana, Bh. 3. 4. consulting. 



244 THE RUINED TEMPLE AT K AN G AVAR 

on the following day, April 20, through fierce storms of sleet 
and snow that swept pitilessly from the north during most of 
the day ; and it was not until after five o'clock in the after- 
noon that I again reached Hamadan, having taken twelve 
hours to accomplish a distance of less than thirty miles. 



CHAPTER XVII 

FROM HAMADAN TO THE RUINED FIRE-TEMPLE NEAR 

ISFAHAN 

4 Unto Fire, the son of the God Ormazd ! Unto thee, Fire, thou son of 
Ormazd, be grace, for thy worship, praise, propitiation, and glorification.' 

AVESTA, Introduction, 2. 

AFTER remaining two nights at Hamadan upon the occasion 
of my second visit, I started late on the morning of the third 
day to continue my journey southward toward Isfahan, 
especially to visit the ruined fire-temple near that city. 
Weather and road alike were favorable, and we reached Nanaj 
at sunset, having travelled some thirty miles, which, owing to 
the late start, was less than my usual march, for I sometimes 
accomplished fifty miles, and occasionally even seventy. But 
I felt fatigued enough to be glad when my camp-bed was 
stretched for the night on the floor of the chdpdr-khdnah, after 
the servant of the post-house had swept the room a little more 
clean. There was much talk about bandits, as the post had been 
robbed on the previous night, but I paid little attention to the 
stories, fell asleep soon, and after a good night's rest was ready 
before daylight to mount Rakhsh and sit thirteen hours in the 
saddle. 

Our cavalcade halted for the second night at the small 
village of Hassar, and we rose with the lark again next morn- 
ing and proceeded along a well-watered plain that was fed by 
streams from the rocky hills on the right. The pace of our 
animals was good, and we easily overtook several caravans that 
had started an hour ahead of us, and all that day the condi- 
tions were favorable for rapid progress. It was Shakspere's 
birthday, a day memorable to me even in Persia, and the season 

246 



246 FROM HAM AD AN TO ISFAHAN 

of spring was well advanced, so that I had an opportunity to 
watch the progress made by the peasants as they tilled their 
farms, and to compare their way of working with the agricul- 
tural methods employed in Zoroaster's day, when the occupa- 
tion of the husbandman was synonymous with a religious 
pursuit. 1 

The Avesta alludes to farms, fields, and husbandry; it praises 
the work of the laborer who tills the earth ' with his right arm 
and his left, with his left arm and his right, ' and lauds the irri- 
gation of arid land and the production and harvesting of crops. 
All kinds of work connected with the soil were equivalent to 
acts of ' righteousness,' and the agriculturist ranked next to 
the priest and the warrior in the constitution of the Zoroastrian 
community. Farming is not a lost industry in Persia to-day, 
but it has made little progress since the days of the Avesta, 
more than two thousand years ago. 

The Persian farm is not fenced off, like ours, but has its 
boundaries marked by trenches and watercourses, which the 
Avesta describes as being ' the depth and breadth of a dog,' 2 
or has its limits indicated by a row of trees, which it well 
repays the laborer to plant, because of the scarcity of wood for 
fuel arid timber in many parts of Persia. The government to-day 
would do well to encourage arboriculture, as it apparently did in 
the time of Darius. 8 

The implements of the husbandman are still of the most 
primitive kind, and my notes regarding them will serve as a 
commentary upon a passage in the Avesta which describes the 
equipment of the peasant. 4 The first to be mentioned is the 
plow (Avestan aesha. Mod. Pers. khlsh) which I have already 
described as a rude affair, 6 consisting generally of the crotch of 
a tree cut so that one of the branches may serve as a plowshare 

1 See Vd. 3. 23-33. nische Kultur, pp. 373-387, Erlangen, 

2 Vd. 14. 12-14. 1882. 

3 For general references see Dar- 4 See Vd. 14. 10-11. 
mesteter, Le ZA. 2. 32 ; Jackson, 6 See pp. 85-86, above. 
JAOS. 21. 183 ; and Geiger, Ostlra- 



AGRICULTURE NOW AND OF OLD 247 

when shod with iron (cf. Av. ayazhdna paiti-darezdna) . l Such 
a plow, as contrasted with the heav}^ plow for deep furrowing, 
to which several yokes of oxen are hitched, 2 is drawn only by 
two cows or heifers (cf. Av. gavd azi)^ and only loosens the 
surface of the ground, as shown in my photograph, which gives 
an idea also of the kind of ox-goad (Av. gavdza) and yoke 
(Av. yuyo-semi) that are still in use in Persia. 4 Another stock 
article in the peasant's outfit, which has remained practically 
unchanged, I believe, since ancient times, if we may judge from 
an allusion in the Avesta, is the handmill. An essential part 
of this machine is the round hopper, or funnel, into which the 
grain is poured when about to be ground; and with this I 
would compare zgeresno-vaghdhana, one of the obscure Avestan 
terms applied to the handmill. 6 

A third article used by the Avestan husbandman was the 
spade (Av. kastra). The Persian spade has a long handle, to 
which is generally attached a wooden foot-rest to serve as a 
support for the foot when driving the spade into the ground, 
as shown in my photograph. 6 Instead of the wooden footpiece 
the upper part of the blade is sometimes bent over on either 
side so as to form a metal flange on which the foot can rest. 7 
When digging, two or three laborers, either barefoot or wear- 
ing the Persian glvahs, work side by side, striking the shovel 
into the earth at the same time and lifting it again at a given 
signal, working in unison as they cut the trench. 

1 This is at least a suggestion which Persia and its People, pp. 109-112, 
I offer in explanation of the difficult Rock Island, Illinois, 1899; Adams, 
Avestan words, Vd. 14. 10. Persia, pp. 153-155 ; and Ker Porter, 

2 See the illustration in the chapter Travels, 2. 533. 

on Urumiah, above. 5 So I explain Vd. 14. 10, asmana 

8 Ys. 46. 19 ; cf. Ys. 29. 5, etc. Havana, ydvardnam zgdrdsno-vaySa- 

4 For the technical terms in the ndm ; cf. also Knanishu, Persia, pp. 

Avesta, see Vd. 14. 10. The photo- 107-109. Query : cf. Nir. 94. 

graph which I took was snapped be- 6 This photograph I took between 

tween Hassar and Leilhahan, on the Hamadan and Asadabad. 

third day after leaving Hamadan. 7 This I noted particularly at Kurd 

For illustrations and descriptions of Balah, near Isfahan, and then more 

plows and plowing, see Knanishu, generally as I passed southward. 



248 FROM HAMADAN TO ISFAHAN 

In tilling the fields an ordinary large rake and harrow are 
employed, and sometimes a flat scraper, with spikes around the 
lower edge and with handles above to guide it, is drawn over 
the field by oxen. A mallet is likewise employed to break up 
the clods after plowing. The hoe, as far as my observations 
served me, has cords attached to it, which are pulled by one of 
the workmen while another guides the implement. 1 The sickle 
which is used in cutting the grain at harvest time I shall 
describe in a later chapter. The manner of threshing it is as 
primitive as in early Bible times. The wheat or barley is 
commonly trodden under the feet of muzzled oxen or mules, 
who drag over it a sort of sled ; but sometimes it is crushed 
beneath a spiked roller or flailed with a peculiar flail, after 
which the chaff is separated from the wheat by the winnowing 
process when a good wind is blowing. The threshing-floor 
itself is usually situated on the outskirts of the village, and 
lumbering carts Qardbah) carry the grain to the barn of the 
peasant, or more often to the granary of the landlord. 

Our halt on the night of the third day was at Leilhahan, a 
settlement of a thousand families, I should judge, the majority 
of them being Armenians. Here I was visited by a native 
preacher, Rev. Rabin Joseph, who was doing evangelistic work 
among the people of the place. He spoke English quite well, 
having acquired it at the Urumiah Mission, and was European- 
ized to the extent of collecting souvenir post-cards from 
foreign countries, and he asked me to add one from America to 
his album, a request which I fulfilled before the year was past. 
He gave me useful information regarding my route for the fol- 
lowing day, and advised me to take the longer road to Khomain 
because the shorter route was at the time infested by brigands, 
who had pillaged several caravans. 

An early start next morning brought me before eight o'clock 
to Khomain, and I was directed to the house of the chief man 
of the place. His courtyard was filled with servants, though 

1 For illustrations, see Knanishu, Persia, p. 110 ; Adams, Persia, p. 154. 



LEILHAHAN AND GULPAIGAN 249 

he himself had not risen, but he sent a message that he would 
be ready in a few minutes to receive me, which he did with a 
gracious welcome. In manners he was Eastern, but in costume 
more European than Persian, and the watch which he wore 
was carried so as to be appropriately conspicuous. His black 
wool hat was wholly Iranian in appearance, however, being a 
replica of the balloon-shaped head-covering worn in Sasanian 
days, and I noticed that this style of headgear was peculiarly 
characteristic of the people in this vicinity. He served tea as 
an act of hospitality and asked various questions, until it was 
time for me to take my leave, whereupon he sent two armed 
horsemen to accompany me over a part of the road, which was 
dangerous because of freebooters. The guards proved to be of 
little use beyond raising clouds of dust as they made their 
horses curvet in circles around our party and fired shots at im- 
aginary robbers supposed to be lurking in the hills. By the 
time we crossed the last mountain pass I was glad to dispense 
with these attendants, and .we descended without accompani- 
ment into the great plain in which lies the town of Gulpaigan. 
Gulpaigan is a town that was visited by the famous Italian 
traveller Pietro della Valle on his way to Isfahan three hundred 
years ago. He says that 4 it resembles Hamadan, but is smaller,' 
and that its name is composed of the three words gul, pdi, and gdn 
- which is merely a popular etymology. 1 As a matter of fact, 
the older name of Gulpaigan was G-arbddakan in Persian, or 
Jarbddakdn in the Arab geographers, and in still earlier times 
it is said to have been called Samroih, after a daughter of the 
Keianian queen Humai who is alleged to have founded it. 2 

1 Pietro della Valle, Viaggi, 1. 449 ; see likewise Sadik Isfahani, Geo- 
Travels, ed. Pinkerton, 9. 21. graphical Works, p. 86 ; Mokad- 

2 See Yakut, p. 152, for the former dasi, ed. De Goeje, Bibl. Geog. 
statement, and compare Barbier de Arab. 3. 257, 402 (mere mention); 
Meynard, Diet. geog. de la Perse, and Ibn Khordadhbah, ed. De Goeje, 
p. 152, note, for the latter. Yakut 6. 20, 155. Consult furthermore 
(p. 153) speaks of ' Jarapadakdn ' Tomaschek, Zur histor. Topog. von 
(Gulpaigan) as ' a large and celebrated Persien, in Sb. Akad. Wiss. zu Wien, 
place.' For other allusions to the town 102 (1883), pp. 168, 171. 



250 FROM HAMADAN TO ISFAHAN 

Two roads approach it from this direction across the wide 
plain. Since we had no guide to direct our choice I selected the 
one to the right, which was longer, but had the advantage of 
conducting us through the entire length of the town, as it led 
to the lower end of the main street. We were thus able to see 
something of the town, which I judged to be a place of consider- 
able activity. In the way of antiquities I noticed a number of 
large carved stones that looked as if they were ancient, and I 
observed several sculptured rams, with horns curved into a tight 
spiral, resembling the granite figures I had seen elsewhere in 
old Armenian settlements, especially in the burying-ground 
near Dilman. 1 From the impression which I received (an im- 
pression borne out also by the passages I have cited) I believe 
that Gulpaigan would be an interesting place in which to make 
archseological researches, although Ker Porter was against 
attributing any great antiquity to the town. 2 

I did not halt to make inquiries regarding ancient relics, for 
it was nearly three in the afternoon and I was anxious to reach 
the next station, so I proceeded at once to the principal caravan- 
sarai, which I found so crowded that many of the muleteers 
were obliged to lie on the ground by the side of their beasts of 
burden, for lack of other quarters. After an hour's stay we 
started once more, allowing sufficient time for making our 
second station before nightfall. 

Our route lay through a lonely plain girt on either side by 
barren hills which sometimes thrust their spurs almost across the 
track. Near the base of one of these jutting promontories the 
worthless guide whom I had hired at Gulpaigan to show us 
the way, fell into a violent altercation with the leader of my 
caravan and drew his carbine to shoot him. There was an up- 
roar at once. Quickly wheeling my horse about, I struck at 
the good-for-nothing rascal, diverted his aim, and separated the 
angry combatants. I did not wait to argue the respective 

1 See p. 79, above, and compare Ker 2 Ker Porter, Travels, 2. 67-68. 
Porter, Travels, 2. 614 (illustration). 



THE GARDENS OF KHONSAR 251 

merits of the claimants in the case, but immediately dismissed 
the fellow, with a threat to have him punished by the governor 
of the district if I saw his face again, and restored order once 
more in the excited caravan. We had no further trouble, but 
had to trust to luck in finding our way. This was not difficult, 
however, and we reached our destination before dark. 

Our halt for the night was at Banishun, or Wanishun, called 
4 Oniscion ' by Pietro della Valle. 1 This garden spot gave us 
a foretaste of the still greater attractions of Khonsar, where 
we arrived on the following morning. Khonsar left on my 
mind the impression of terraced hills, abundant streams, green 
foliage, and blossoming fruit trees. The town is picturesquely 
situated in a long and narrow gorge between high hills and 
through this passage it winds its way for five or six miles, now 
at the base of the mountains and now climbing over their 
spurs. It is a settlement of some twelve thousand inhabitants 
and is believed to date back to the time of Alexander, who is 
thought to have marched through it on his way to Ecbatana. 2 In 
Persian the name is written as Khudnsdr, and Yakut says that 
the town had the reputation of being the birthplace of several 
minor poets. 3 In addition to ats romantic site and historic 
associations Khonsar is a place that does a thriving business in 
cotton goods, so that its taxes, combined with those of Gul- 
paigan, with which it forms a single district for administrative 
purposes, yield a considerable revenue to the government. 4 
I could have wished that the town had a municipal ordinance 
to prevent butchering sheep in the public streets. 

Directly after we left Khonsar and entered the mountain 
passes of the Ashnar Kuh we encountered one of those sudden 
changes of weather which are characteristic of Persia, for we 
had to struggle for two or three hours through snow, hail, 

1 Pietro della Valle, Viaggi, 1. 450; 8 Yakut, p. 195; cf. Barbier de 
Travels, ed. Pinkerton, 9. 21. Meynard, Diet. geog. p. 195, note, and 

2 Ker Porter, Travels, 2. 70 ; Zolling, also Sadik Isfahan!, p. 94 ; Tomaschek, 
Alexanders des Grossen Feldzug in op. cit. p. 170. 

Central- Asien, p. 79, Leipzig, 1875. * See Curzon, Persia, 2. 480. 



252 FROM HAMADAN TO ISFAHAN 

sleet, torrents of water, slush, and mud, which transferred us 
from April 25 back into the heart of winter. About four miles 
this side of Dombanah I noticed the snow-covered outlines of a 
rectangular structure which looked as if it might have been 
one of those square-shaped dakhmahs, or Towers of Silence, 
pictured in the old volume of travels by Olearius three cen- 
turies ago. 1 I dismounted to examine the general contour and 
arrangement of the place, but the snow prevented me from 
making a careful investigation. There were many fragments 
of bricks and, in the middle of the enclosure, a depression which 
might have corresponded to the central pit, or grave, generally 
found in a dakhmah, but the ruin seemed to be too remote 
from any present settlement to have been an ancient ground 
for disposing of the dead. I am not unmindful, however, that 
in early times the entire region was occupied by Zoroastrians 
and that I was not many stages distant from the famous fire- 
temple near Isfahan, the ruined sanctuary which was my goal. 
Halting only for a night at Kurd Balah, and another at 
Hajiabad, I arrived early in the forenoon of the following day 
within sight of this ancient shrine of the Fire-Worshippers, 
which is commonly known as, the Atash Kadah or Atash Grdh 
near Isfahan. 

The sky had all the peculiar clearness which belongs to a 
true April morning in Persia, and a soft light quivered over 
the plain beyond Najafabad. The animals of my caravan 
moved slowly along the well-worn trail; Shahbas, the muleteer, 
was asleep in the saddle ; but my own eyes were busy watch- 
ing to catch the first glimpse of the Atash Grdh, or Kuh-i Atash 
Kadah, 'Hill of the Fire-Temple.' Far in the distance over 
the level horizon there arose unexpectedly before my view a 
lake bordered with delicate green cypress trees, and I fancied 
that the beautiful spot must be one of those Persian parks, 
or 'paradises,' which were so enchanting in ancient Iran. 
Instinctively I quickened my horse's pace ; but suddenly the 
1 Olearius (1600-1671), Persianische Reise-beschreibung, p. 296, Hamburg, 1696. 



THE AT ASH KADAH NEAR ISFAHAN 253 

picture vanished. I then became aware that I had been 
deceived by a mirage. This was the first of many experiences 
with that magic phenomenon which transforms rocks into 
ruined castles, bushes into troops of horsemen, puddles into 
sheets of clear water, and molehills into mountains. So vivid 
was this first impression that it took me some time to recover 
from the surprise, and I felt almost inclined to question my 
eyes when I actually caught sight of the ruined fire-temple. 

The deserted shrine stands on the top of a hill which rises 
about seven hundred feet sharply above the plain at a distance 
of three or four miles from Isfahan. The ascent is by a wind- 
ing path which starts from the southeastern slope of the eleva- 
tion and proceeds by a series of natural steps formed by 
ledges of the unhewn rock. Fragments of bricks and pottery 
of a yellowish clay strewed the rough pathway, and a few of 
these which I found were decorated by ornamental raised lines. 

The ruined sanctuary stands on the very crest and is about 
fourteen feet high and fifteen feet in diameter, octagonal in 
form, and constructed of large unburnt bricks. The roof was 
originally a dome, but most of this vaulted covering has fallen. 
As shown in the accompanying photograph, which I took from 
the western side, eight doorways look out toward the different 
points of the compass. 1 The brick and stucco columns which 
form the sides of the doorways and support the roof are so 
arranged that they give a pillared effect to the temple. There 
was no artificial foundation beneath the structure ; the build- 
ing was erected directly upon the natural rock, part of which 
thrusts itself into the middle of the floor. 

On entering the crumbling fane I noticed over each doorway 
a sunken niche, the lines of which curved symmetrically to a 
point in such a manner as to give an arched finish to each 
entrance on the inside. Traces of a brownish plaster or stucco 

1 The dimensions of the shrine are height of doorways, 7 ft. 3 in. (2.20 
nearly as follows: height, 14 ft. m.) ; width of doorways, 3 ft. 7 in. 
(4.00 m.) ; diameter, 15 ft. (4.50 m.) ; (1.10 m.). 



254 FROM RAMADAN TO ISFAHAN 

were preserved in these panels, and portions of the dome and 
walls which were above ordinary reach were still intact. 
Imagination alone could restore the original finish of the walls, 
whether a layer of tiles, a wainscoting of marble, or panels of 
stone around the columns. There was not a trace of an inscrip- 
tion, tablet, or sculpture anywhere to be seen, except modern 
Persian names written by those who had scrambled up to 
scrawl their signatures in its niches. The Oriental does not 
differ much from some of his cousins in the West in the repre- 
hensible fondness for inscribing his name in conspicuous places. 
The floor of the sanctuary was 13 feet 8 inches in diameter 
(4.16 m.); it was nearly circular in shape, and in the centre 
there was the curved outline of what was probably a mortar 
base on which the fire-altar rested. Beneath the debris I found 
ashes ; who knows whether some of them may not have kindled 
the fire of the Magi ? 

In addition to the shrine, the summit of the hill is capped by 
the remains of a series of buildings which are gathered about 
the fane itself, but stand a little below it, and occupy the south- 
eastern, eastern, and northeastern sides of the hill-crest. These 
form together a part of the general temple precinct, having 
probably served as an abode for the priests, a sanctum for the 
fire, and perhaps also as a temple treasury. The design and 
arrangement reminded me of the ruined sanctuary of fire which 
I noticed near Abarkuh on my journey to Yezd. 

To examine the crumbling chambers I had to descend a few 
paces from the sacred building which I had been inspecting. 
Some walls of a ruined edifice on the southeastern side of the 
summit first attracted my attention. They were the remnants 
of a succession of halls and rooms built of clay and brick ac- 
cording to a definite plan of construction, but they were all in 
a hopeless state of dilapidation. I turned from these to the 
northeastern side of the crest. Here I found a still more 
elaborate structure, but even in a worse state of ruin than the 
preceding. One room, about twenty feet square, was still in a 




RUINS OF THE SANCTUARY 255 

sufficient state of preservation to allow me to make out its gen- 
eral outlines, but it was filled halfway up the sides with dirt 
and rubbish. The walls were made of clay and bricks covered 
over with plaster, and two of the side walls, which had no 
doorway or window, were marked by blind arches indicated by 
columns of unburnt brick and capped by burnt 
bricks set in the manner here shown, to lend a 
decorative 'effect. 

There were also niches in the wall, resembling the familiar 
tdkchahs seen in modern Persian houses. The other two sides of 
the room, as shown in my photograph, had archways looking 
out over the plain, and the view from these, comprising the 
silver thread of the Zendah Rud River and a rich succession of 
gardens about Isfahan, was a contrast to the ruin and decay 
which reigned within the walls. Besides the chamber which I 
have been describing, there was still another ruined apartment, 
the stuccoed walls and outlines of which could be seen ; but it 
was smaller than the one I have just mentioned and had nothing 
particular about it to require special notice. 

After I had inspected the ruins I made some observations 
with regard to the general contour of the hill. The sides drop 
with a fairly steep declivity, except at one particular point, and 
there a defensive wall guards the approach to the shrine. 
No one knows how often it may have been of service in ages 
past. As I descended from the hill I noticed several arched 
recesses on the side of the crest. They looked like cells and 
were partly constructed of large sun-dried bricks, like those at 
Rei. Some of these recesses were small, others fairly large. 
One of them, which showed careful construction, was about 
twelve feet deep, three feet wide, and less than four feet high. 
I have no opinion to express regarding the original purpose of 
these cells. They were in too exposed a position to allow one 
to conjecture that they may have been used as places of deposit 
for the temple treasure, and it would be hazardous to surmise 
that they were intended to be hermit cells, or that they were 



256 FROM HAM AD AN TO ISFAHAN 

employed as shrines for consecrated lamps, or as repositories 
in which the wood for the sacred fire was prepared. 

This is but one of a number of interesting problems con- 
nected with the fire-temple and its history. For the very 
reason that the subject has been little dealt with, I have gone 
into more detail in describing the appearance of the ruined 
shrine, and I shall continue this by giving such an account of 
its historic past as I am able to present. I have not been able 
to find anything on the subject in five of the earliest European 
travellers : Josafa Barbaro, Pietro della Valle, Herbert, Olearius, 
and Mandelslo. So far as I can see, Tavernier, in the seven- 
teenth century, is the first Occidental to mention it. I shall 
give his statement and those of his successors, and after that 
I shall cite three or four Oriental authorities, which are cen- 
turies older than the early French traveller and will give con- 
siderable help toward identifying the temple historically. 

Tavernier, who made several journeys to Persia between the 
years 1638 and 1663, alludes incidentally to the ruins in ques- 
tion as a 'fortress,' but he does not associate them with the 
ancient religion of Persia. His statement is this : 

'Upon the south (southwest?), about two leagues from Isfahan, 
lies a very high mountain, on the top whereof toward the west are 
to be seen the remains of a very strong fortress, where Darius kept 
himself when Alexander gave battle to him in that plain. In the 
side of the rock is a grotto, either natural or artificial or both, out 
of which issues a natural spring of excellent water, where a Dervis 
usually inhabits.' l 

Daulier-Deslandes (1665) merely mentions the mountain 
in connection with Gabarabad, the Gabar suburb of Isfahan. 2 

Chardin (1666, 1677) confines his remarks to a few words 
about 'le Bourg des Guebres ' when he is describing the section 
known as the Gabar settlement near Isfahan, which was cleared 

1 Tavernier, Travels, p. 149, London, la Perse, p. 61 (cited from Hyde, 
1684. Historia Eeligionis Veterum Persa- 

8 Daulier-Deslandes, Les Beautez de rum, p. 359). 



EARLY DESCRIPTIONS OF THE TEMPLE 257 

by Shah Abbas in order to make room for a part of the royal 
residence in the environs. 1 

Bell (1715) repeats the story which speaks of Darius and 
Alexander. The Englishman's statement reads as follows : 
4 About three or four miles to the southward of the city are to 
be seen the ruins of a tower on the top of a mountain, where it 
is said Darius sat when Alexander the Great fought the second 
battle with the Persians.' 2 

Ker Porter (1821) devotes a paragraph to the description of 
the hill and concludes his observations with these words : 4 But 
as this Attush Kou is an artificial mount and stands close to the 
quarter of the city where the Guebres, and particularly those 
who followed the arms of Mahmoud, dwelt, no doubt they 
reseated themselves in a spot that had been inhabited by their 
ancestors from the first peopling of the banks of the Zein- 
derood ; and they found it thus marked by the High Place of 
their worship.' 3 Ker Porter's last statement is right in so far 
as he emphasizes the antiquity of the place ; but he is wrong 
in calling the mount ' artificial,' as an examination of the liv- 
ing rock which composes it would have shown him. 

Sir William Ouseley (1823) has a mere mention of the hill : 
4 In the view [from Julfa] is seen, above the bridge, a moun- 
tain distant five or six miles, on which are some remains of an 
edifice not very ancient but occupying, as tradition states, the 
site of a ruined Fire-Temple. Hence the mountain has been 
denominated Kuh dtesh Kadah or dteshgd h. ' 4 

Ussher (1865) gives a brief description of the mount and 
alludes to the large size of the bricks seen in the ruins; he 
adds that the fire-temple was 4 erected, it is said, by Ardeshir, 
or Artaxerxes.' 5 

Lord Curzon (1892) has but a couple of sentences on the 

1 See Chardin, Voyages, 2. 105, Am- Ker Porter, Travels, 1. 437. 
sterdam, 1735. Compare also Curzon, 4 Ouseley, Travels, 3. 49, and pis. 
Persia, 2. 47. Ivi, c, London, 1823. 

2 Bell, Travels in Asia, ed. Pinker- 6 Ussher, Journey from London to 
ton, 7. 308, London, 1811. Persepolis, p. 595, London, 1865. 



258 FROM RAMADAN TO ISFAHAN 

subject of this 4 isolated rocky hill, the summit of which is 
crowned by some ruins of mud-brick. This is called the Atash 
Gah, from a tradition that a fire-altar was here erected by 
Ardeshir (Artaxerxes) Longimanus. The tradition may be 
true, but the present ruins are not old.' l 

To these meagre statements I am now in a position to add 
more material which will throw fresh light on the subject and 
carry the history of the fire-temple more than a thousand years 
back of the present. This material is to be found in the Arab 
geographers of the ninth and tenth centuries, who call this 
pyrseum the fire-temple of Mdrabin, or Maras (according as we 
are to read the name and its variants), 2 the name of a village 
or district in the vicinity of Isfahan. 

The first of these Oriental witnesses is Ibn Khordadhbah 
(A.D. 816) whom I have already quoted on several other occa- 
sions. In describing the district of Isfahan he mentions 4 the 
village of Marabin (v.l. Maras) in which there is a citadel built 
by Tahmurath, and in it a temple of fire.' 3 This statement 
shows that even in Ibn Khordadhbah's time the temple was 
regarded as very ancient. 

The second author is the well-known writer Masudi, who 
died in the year 957. In his Meadows of Gold, written in 
943-944, Masudi refers to the fire-temple near Isfahan as the 
second of seven sanctuaries, which were originally idol-temples 
devoted to the worship of the sun, moon, and the five princi- 
pal stars. His statement is as follows : ' The second of these 

1 Curzon, Persia, 2. 58. historischen Topographic von Per- 

2 The reading of J^U in an un- sien ^ p> 17L 

pointed text will easily account for 8 Ibn Khordadhbah, ed. De Goeje, 

these differences in spelling, the forms Bibl. Geog. Arab. 6. 20, transl. p. 16. 

Maras, Mdras, being in fact the vari- In the same manner Hamadhani 

ant, and Mdrabin the more original. (ed. De Goeje, 5. 265) says, ' When 

It is true that Ptolemy, Geog. 6. 4. 4, Tahmurath reigned he built the vil- 

mentions a place, Mappd<rtov, near lages of Marabin (v. 1. Maras) and 

Aspadana (i.e. Isfahan), but we can- Ruvandast in the district of Isfahan' ; 

pot be sure that this was the and again (6. 263) he associates 4 Jei 

ancient Maras, cf. Tomaschek, Zur (i.e. Julfa) and Marabin.' 



IDENTIFICATION OF THE SITE 259 

temples is situated on the summit of a mountain called Maras 
(Mdrablri), near Isfahan. There were idols in this until they 
were removed by King Yustasf (i.e. Vishtaspa, the patron of 
Zoroaster) when he adopted the religion of the Magi and con- 
verted the shrine into a temple of fire. It is three farsakhs 
from Isfahan and is still held in great veneration by the 
Magi.' ! 

This important statement by Masudi proves conclusively that 
the temple was not in ruins in the early part of the tenth 
century of our era, and that tradition (like the previous al- 
lusion to Tahmurath) ascribed to it an antiquity even prior to 
Zoroaster's day, since it is said to have existed as a shrine of 
idols before Vishtaspa converted it into a fire-temple. I feel 
positive that the present ruins go back at least to Sasanian 
times, and there may be reasons for assigning them to a still 
earlier period. The site was certainly an ancient one, as is 
shown by the statements of both the Oriental authorities whom 
I have quoted, and there may be grounds for laying more stress 
on the tradition, cited above, which connects the temple with 
the name of Ardashir Dirazdast, known to us through the 
classics as the Achsemenian king Artaxerxes Longimanus. This 
monarch reigned B.C. 465-425, a century and a half after the 
date assigned by tradition to Vishtaspa, and it is said in the 
Bahman Yasht that he made the religion of Zoroaster current 
in the whole world. 2 

In corroboration of such a suggestion regarding Bahman's 
connection with this particular fire-temple I may cite the 
authority of a third Oriental writer, Ibn Rostah (about A.D. 
950) who says : ' Marabin borders upon the town of Jei. It 
was one of the pleasure-grounds of the early Khosrus. Kei 

1 For the original text of Masudi and repeats practically the same statement 

a French translation, see Les Prai- as that which I have quoted, see 

ries d'Or, ed. Barbier de Meynard, Vullers, Lexicon Persico-Latinum, 

4. 47, Paris, 1865. I may add that s. v. Maras. 

Muhammad Hasan Burhan (1651), in 2 BYt. 2. 17 ; see my Zoroaster, 

his Persian dictionary Burhan Kdta 1 -, p. 160. 



260 FROM HAMADAN TO ISFAHAN 

Kaus is said to have resided there and to have beautified the 
place. At his command a lofty and magnificent citadel was 
erected on the mountain peak there. It towered aloft so as to 
overlook the valley of the Zendah Rud ; l and from its summit 
there was a commanding view of the entire country. But King 
Bahman, son of Isfendiar (i.e. Vohuman Ardashir Dirazdast), 2 
took possession of it and burned it ; and he built below it a 
stronghold and established in it a shrine of fire which stands 
till this day, and even the fire remains in it.' 3 

A similar tradition connecting the temple with the name 
of Ardashir Bahman Dirazdast (Artaxerxes Longimanus) is 
preserved in the annals written by Hamzah of Isfahan (A.D. 
eleventh century), who, as a native, was well acquainted with 
the traditions of Isfahan. He writes as follows regarding 
Ardashir Bahman : 4 He founded in one day three fire-temples 
in the Province of Isfahan. The first was in the east, the 
second in the west, and the third in the middle. The first of 
these is situated near the citadel of Marin (i.e. Mar^al^m, or 
Mdras) and is the Fire of Shahr Ardashir, the word Shahr sig- 
nifying district, and Ardashir being, a name of Bahman ; the 
second is the Fire of Zervan Ardashir, located in the territory 
of Darak called Barkah ; and the third is the Fire of Mihr 
Ardashir, located in the territory of Ardistan of the same 



name. 



' 4 



One other Oriental passage is to be mentioned. It is found 
in a Persian history of Isfahan based on an Arabic original 
composed in the year A.D. 1030 by Mufaddal b. Seid b. Al- 
Husein Al-Mafarrukhi and entitled Risdlatu Mahdsini Isfahan. 
But it must be noticed that this source attributes the temple 

1 The text has Zarrinrudh. Barkah compare Yakut, pp. 99, 222. 

2 See my Zoroaster, pp. 157-163. Yakut (p. 509) has a brief mention 
8 Ibn Rostah, ed. De Goeje, Bibl. only of a place which he calls Mar- 

Geog. Arab. 7. 152-153. banan, situated one half farsakh ' 

4 See Hamzah of Isfahan, Annalium from Isfahan, but he makes no men- 

Libri X, ed. Gollwaldt, 2. p. 27, Leip- tion of the fire-temple, 
zig, 1848. For the names Darak and 



A FAMOUS FIRE-TEMPLE 261 

rather to Sasanian times, assigning it to the reign of Piruz, 
son of Yazdagard (A.D. 459-484), and ascribing its erection to 
Adhar-Shapur, the head man of the village of Muristan in the 
district of Marabin. 1 

Whether we ultimately decide to assign the origin of the 
temple to Achsemenian times or to assign the present ruins to 
Sasanian times, something has been added to our knowledge of 
the shrine and its site by carrying its history back for at least 
fifteen centuries. It is sufficient for the present to know that 
the sacred fire burned for ages upon its altar and we may leave 
it to the future to decide to which particular Zoroastrian divin- 
ity the temple may have been dedicated. 

1 For this reference, see the article tory of Isfahan, in JEAS. 1901, pp. 
by Browne, A Eare Manuscript His- 417-418. 



CHAPTER XVIII 
ISFAHAN, THE FORMER CAPITAL 

* Following on o r io r ney we came to a towne called SPAHAM, which hath been a 
notable town till of late. ' 

JOSAFA BARBARO (1474), Travels in Persia, p. 71. 

GARDENS, palaces and pavilions, mosques and madrasahs, 
bazaars, splendid bridges, and above all a magnificent Royal 
Square these are the impressions which the traveller carries 
away from Isfahan, and so vivid are they that time dulls them 
but slowly. The city has a special claim to attention, as it was 
ranked as the metropolis of Persia from the sixteenth to the 
eighteenth century, before Teheran usurped its place as capital 
of the Shah's dominion. Isfahan, nevertheless, retains its 
traditional title Nisf-i Jahdn, 4 Half the World,' even though 
the rest of the world to-day may know it only as a centre of 
trade in rugs, or possibly as the scene of the Persian novel, 
Hajji Baba of Isfahan. 

The city lies in a level tract of country which extends 
around it for miles. Entrance to the town is made through a 
maze of walled vineyards and orchards, whose variety of color 
resembles a Persian carpet. Mosques pierce the sky-line with 
their slender minarets, or rival the blue of heaven with their 
turquoise domes; poplars and plane trees lend grace and color 
to the scene; and distant hills form a serrated background for 
the picture. A peculiar feature of the landscape is the large 
number of pigeon-towers which line the sides of the road and 
dot the plain beyond. These turreted columbaries, which are 
built of clay and brick and look like windmills that have lost 
their arms and sails, provide shelter for myriads of pigeons and 

262 



OF THE 

UNIVERSITY 

OF 
















FLANDIN'S SKETCH OF THE HILL OF THE ISFAHAN FIRE-TEMPLE, WITH A 
PIGEON-TOWER IN THE FOREGROUND 









ISFAHAN AND THE BRIDGE OF ALI VERDI KHAN 



A PERSIAN DELHI 263 

form a lucrative source of revenue to those who sell the drop- 
pings of the birds to be used as a fertilizer on the neighboring 
fields. 

In some respects Isfahan offers to the traveller who is 
attracted by the modern phases of Persia more objects of inter- 
est than any city of Iran. It may be true that the city has lost 
much of the splendor that distinguished it three hundred years 
ago as the capital of Shah Abbas the Great, whose lavish hos- 
pitality to the foreigners that visited his court is described by 
the early European travellers, Tavernier, Chardin, Sanson, 
Fryer, and Kaempfer. It is equally true that the city never 
fully recovered from the blow that it suffered in the eighteenth 
century from, the Afghan invasion, which lost for it its prestige 
as capital and resulted in the transfer of the imperial seat to 
Teheran. Nevertheless, enough of the old lustre remains (even 
though the effect is sometimes produced by tinsel) to make 
Isfahan a Persian Delhi and a worthy rival to its modern suc- 
cessor on the Caspian littoral. I cannot quite agree, therefore, 
with some of the recent writers in their tendency to bemoan 
the decadence of Isfahan and lament over its vanished glory. 
I acknowledge that it has been dimmed, but I should be far 
from accepting that pessimistic view which would interpret 
the title Zil as-Sultdn, ' Shadow of the Sultan,' held by its gov- 
ernor, a brother of the Shah, as symbolizing the shade into 
which Isfahan has been thrown by Teheran. From what I 
could learn about the vigorous possessor of this umbrageous 
attribute, I should judge that he and the city resemble each 
other in being very substantial shadows and important factors 
in Persian affairs. The general condition of the people, the 
evidences of commercial activity, and the apparent signs of a 
growing trade, seemed to me to be indications of promise again 
for the future. 

I have called Isfahan a modern city; it is so in its present 
form, for it owes its surviving traces of beauty to Shah Abbas I, 
the contemporary of Queen Elizabeth, and to his successors 



264 ISFAHAN, THE FORMER CAPITAL 

in the seventeenth century. Historically, however, Ispahan, 
or Isfahan, as the natives more generally call it, can lay claim 
to great antiquity. The Greek geographer Ptolemy, in the 
second century after Christ, mentions it as Aspaddna, which 
would be the equivalent of an Old Persian word signifying 
'having horses as a gift.' 1 The Pahlavi texts designate the 
city as Spahdn, in one case mentioning it in a comment on an 
Avesta passage, 2 in another stating that the original governor 
was Sparnak, who was apparently a brother of the hero Rus- 
tam, 3 and in still another adding that Rustam and King Kaus 
defeated Afrasiab, the great enemy of Iran, in a battle near the 
borders of Isfahan. 4 

The most ancient section of the city (corresponding in 
part to the modern suburb Julfa) was called Jei by the Arabs, 
a name which answers to the Pahlavi Gral and the classical 
G-dbae, and is associated in some way with the name of Kavi 
the blacksmith, of legendary fame in Iran. 5 Tradition makes 
him a native of Isfahan and tells how he headed a rebellion 
against the tyrant Zohak, or Azhi Dahaka, of Babylon. Zo- 
hak is represented as a monster from whose shoulder grew 
two serpents that had to be fed each day with the brains of 
children. When the tyrant caused Kavi's two sons to be 
killed for this purpose, the blacksmith raised an insurrection, 
hoisted his leather apron upon a spear as a standard, marched 
with the hero Feridun to Babylon, and overthrew and slew the 

1 Ptolemy, Geographia, 6. 4. 4 (A<r- Gumanik Vijar, 2. 2 (West, SBE. 
vdSava). 24. 123, ed. Hoshangji and West, pp. 

2 Phi. Vd. 2. 23 (52) ttgun Spahdn. 11, 188). 

8 See Bd. 81. 10, tr. West. SBE. 5. 6 For the names, cf. Shatroiha-i 

140 ; and regarding the uncertain Airan, 53 (Gal or Gae), Strabo, Geog. 

reading of the governor's name as 728 (ed. Meineke, 1015. 2), and Ptol- 

' Sparnak,' see Justi, Iranisches Nam- emy, Geog. 6. 4. 7 (r<cu). See 

enbuch, p. 307 b. also Marquart, ErdnSahr, p. 29. The 

4 See Great Iranian Bundahishn, hero is called ' Gavah of Ispahan ' in 

41. 7, tr. Darmesteter, Le ZA. 2. 402 ; the Persian Sad Dar, 63. 5 (cf. West, 

cf. also West, Gryndr. iran. Philol. SBE. 24. 323). On Jei-Julfa, see 

2. 102. Another incidental allusion also Justi, Grundr. iran. Philol. 2. 

to the city is found in Shikand- 485, with references. 



HISTORY OF ISFAHAN 265 

monster. The leather apron mounted upon a spear became 
the national ensign of Iran, and the keeping of this treasured 
emblem was entrusted to Isfahan and remained its cherished 
privilege for ages. 1 

Another quarter of Isfahan in early times was called Yahu- 
diah, 4 Jewry,' having received its name, according to Persian 
tradition, from a colony of Jews who came to Isfahan as exiles 
from Jerusalem in the reign of Nebuchadnezzar. 2 The Sasa- 
nian king Yazdagard I had as his queen a Jewish princess of 
Gae, or Isfahan, descended from these exiles, 3 and there is still 
quite an extensive settlement of Persian Jews, numbering some 
five thousand, in Isfahan. 

The history of Isfahan from the time of the Mohammedan 
invasion differed little from that of other Persian cities which 
were conquered successively by Mahmud of Ghazni, Jenghiz 
Khan, and Tamerlane, down to the end of the sixteenth century, 
when Shah Abbas lifted the town to the rank of capital of the 
empire, a dignity which it had already enjoyed, though only 
over a limited province. Its central position is such as to 
qualify it well for a metropolis, even if Teheran has now an 
advantage in being more accessible to Europe, and it is certain 
that during the two centuries of its ascendancy Isfahan merited 
the renown for beauty to which the great Shah Abbas believed 
his capital was entitled. 4 

The topography of Isfahan is easy to grasp, if we bear in 
mind that the city lies on the north side of the Zendah Rud, 
and that Julfa, its Armenian suburb, lies on the south side of 

1 See Yakut, p. 43. Juives, 19. 41 ; idem, La Heine. 

2 Al-Hamadhani, ed. De Goeje, Shasydn Dokht, in Actes du Huitieme 
Bibl. Geog. Arab. 5. 261-262, and Congres International des Orienta- 
Yakut, p. 613. On Nebuchadnezzar listes, sec. 2. 193-198, Leiden, 1892. 
cf. also Gray, Kai Lohrasp and Nebu- 4 For additional material in a rare 
chadrezzar, in WZKM. 18. 291-298. Persian manuscript relating to Isfahan 

8 See Shatroiha-i Airan, 53 (ed. and its history, see Browne, JEAS. 

Modi, pp. 111-113), and compare 1901, pp. 411-446, 661-704. Cf. also 

Darmesteter, Textes Pehlms relativs Houtum-Schindler, Eastern Persian 

au Judaisme, 2. 41, in Eev. fitudes Irak, pp. 119-129, London, 1897. 



266 ISFAHAN, THE FORMER CAPITAL 

the river and is connected with the main part of the town by 
several fine bridges. 

The heart of the city, and central point of interest, is the 
magnificent Meiddn-i Shah, c Imperial Square,' which is men- 
tioned even in the Shah Namah 1 and is one of the most impos- 
ing plazas I have ever seen. Its length from north to south is 
more than a quarter of a mile, and its breadth from east to 
west is nearly an eighth of a mile. It is as level as a parade- 
ground, and as we canter over its smooth surface we are re- 
minded of the days, three hundred years ago, when the rulers 
of the capital used to have exhibitions of the traditional horse- 
manship of the Persians given here. A prize, sometimes a 
golden goblet, was set on the top of a pole in the midst of the 
vast arena and shot at as the marksmen galloped by ; or sides 
were taken by the princes and nobles in the ancient game of 
polo, ffui u chugdn, and a large marble goal-post is still 
standing at each end of the Meidan to mark the terminus 
toward which they drove the ball that, in the words of Omar 

Khayyam, 

' no question makes of Ayes and Noes, 
But Here and There as strikes the Player goes.' 

But polo is no longer played here ; only occasional parades and 
processions are held, and the caravans wend their slow way 
across it to unload their burdens in the bazaars. 

The four sides of the Meidan are bordered by low-galleried 
buildings, the uniform outline of whose roofs is broken at vari- 
ous points by stately edifices that have real architectural merit. 
These I shall describe only briefly, for it would be impossible 
for me to add anything to the many excellent and full descrip- 
tions which have been given from the days of Ta vernier and 
Chardin to Curzon and Browne. 2 

1 See Firdausi, Shah Namah, ed. reader to Lord Curzon's admirable 
Vullers-Landauer, 2. 746, and tr. account of Isfahan (Persia, 2. 18-69), 
Mohl, 2. 423. which summarizes all that is best in 

2 In general I should refer the the authorities that preceded him. 




MASJID-I SHAH, THE KING'S MOSQUE 




PALACE OF THE ZIL AS-SULTAN, PRINCE GOVERNOR OF ISFAHAN 



AN IMPOSING PLAZA 267 

On the north side, at a point leading into the bazaars, is the 
Nakdrah Khdnah, 'Music Hall,' 4 Band Tower,' where, as at 
Urumiah and other places, a noisily beaten signal and a blast of 
horns accompanies the rising and setting of the sun. 1 Conspicu- 
ous on the east side of the plaza is the blue enamelled dome of the 
mosque of Sheikh Lutf ullah, which dates back several centuries. 
On the southern side of the square, near the middle, is a grand 
arched portal leading to the handsome Masjid-i Shah, 4 Mosque 
of the King,' which stands somewhat back of it, to the south- 
west. This fine specimen of a Mohammedan sanctuary was 
founded in 1612 by Shah Abbas, but it is now unfortunately 
beginning to fall into decay. On the west of the great quad- 
rangle, but nearer the southern end, rises the Royal Palace 
with its grand entrance Ali Kapi, a sort of 'Sublime Porte.' 
The Ali Kapi, with its open porch above, and columns noticeable 
in the picture, served formerly as an audience-hall in which the 
Safavid kings received ambassadors. It was used also as an 
asylum of refuge for fugitive debtors and manslayers. But its 
importance has long since vanished. The Royal Palace itself, 
which stands back of this entrance, is now occupied by the Zil 
as-Sultan, the Prince Governor of Isfahan. This princely 
residence covers a considerable area with its gardens, courts, 
and pavilions, one of which, the Chahal Situn, ' Hall of Forty 
Pillars,' was famous as the veranda and throne-room of Shah 
Abbas. 

The bazaars in Isfahan lie behind the rows of buildings on 
the northern and eastern sides of the Meidan. It is possible to 
walk for two or three miles under their covered shade, or rather 
to push one's way through the crowded mass of camels, don- 
keys, packs, porters, buyers, sellers, and money-changers. The 
bazaars have all the characteristic features which I have already 
described as belonging to an Asiatic mart. Their trade has 
preserved for the city some at least of the prestige which once 
belonged to Isfahan as the emporium of Persia. Brocaded 

1 See p. 104, above. 



268 ISFAHAN, THE FORMER CAPITAL 

cloths, felt goods, saddles, native weapons and armor, lacquered 
ware, articles of silver filigree- work, and objects made of metal 
are among the wares exposed for sale. I still can hear the 
deafening rattle and beat of the coppersmith's mallet and the 
brass- worker's hammer, busily engaged in the manufacture of 
useful culinary utensils and of vessels that were often artistic 
specimens of Persian metal-work. 

To the west of the Meidan, and beyond the precincts of the 
Royal Palace, is the parklike section of the city with its grand 
avenue that leads to the river. In the extreme western portion 
of this quarter, and beginning at the grand promenade, is the 
section of gardens known as the Hasht Bahisht, ' Eight Para- 
dises.' In the midst of this 'paradise' region there stands a 
pavilion built by Shah Suleiman, about 1670, and once a mas- 
terpiece of the creative art of the Safavid dynasty, but, like 
its surroundings, it has been allowed to run down, so that its 
beauty and sumptuousness are a thing of the past. 1 

The grand avenue itself is called the Ohahdr Heigh, ' Four 
Gardens,' from the vineyards which Shah Abbas devoted to 
this purpose when he embellished his capital, and it is some- 
times referred to as the 'Champs Elysees of Isfahan.' It is a 
long boulevard, nearly three quarters of a mile from end to 
end, and two hundred feet broad, laid out with watercourses 
and fountains through its centre and with promenades shaded 
by alleys of poplar and sycamore trees on each side. Little 
is left to tell the story of its former beauty ; neglect and 
decay are all too evident. 

On the eastern side of this shaded avenue, between it and 
the Meidan as we ride toward the river, an impressive build- 
ing attracts the attention. It is the Madrasah-i Shah Husein, 
an educational institution, built about the beginning of the 
eighteenth century by Shah Husein, whose name it bears, and 
designed as a college for the training of mullahs and dervishes. 

1 For a description of this Garden 2. 36-38 ; Brugsch, Im Lande der 
of Nightingales,' see Curzon, Persia, Sonne, pp. 317-319 (after Chardin). 




MADRASAH OF SHAH HUSEIN, FROM THE CHAHAR BAGH 




WITHIN THE COURT OF THE MADRASAH 



THE CHAHAR BAGH AND THE GREAT BRIDGE 269 

The handsome portal, with its doors encrusted with brass and 
chased with silver, calls forth admiration, and its turquoise 
dome, girdled with arabesques in rich yellow, is as beautiful as 
it is graceful. But much of the exquisite tiling on the dome 
has dropped off, and some of the marble panels on the outer 
wall of the building itself have disappeared, while parts of the 
framework and lattice of the windows are broken. But within 
the walls of the Madrasah, in its arched cells, Mohammedan 
students continue to learn the Koran, or find leisure, in their 
recreation hours, to smoke their kalians around the shady tank 
in its courtyard. 

At the end of the avenue of Chahar Bagh is a great bridge, 
which is one of the five that lead over the Zendah Rud to Julfa. 
This bridge is called Pul-i Ali Verdi Khan, after the name of 
the distinguished general of Shah Abbas, and it is one of the 
finest bridges of its kind in the world. It is about twelve yards 
broad and three hundred and eighty-eight yards long and it 
spans the river with a succession of thirty-four arches solidly 
built of brick and stone masonry. The design and construc- 
tion of the bridge is such that it offers at least three viaducts 
for traffic at the same time. The main causeway is for 
mounted passengers and for the occasional vehicles that trav- 
erse it. The galleried arcades on each side of this are for the 
use of those on foot. A vaulted passage, in addition to these, 
pierces the stone arches on which the bridge is built and may 
be employed as an extra means of transit, if needed. All day 
long, till nightfall is signalled by the curfew horn of the Na- 
karah Khanah, this great bridge is thronged by an unceasing 
crowd, and the other bridges are almost equally frequented in 
the daytime. 

It was toward evening when I first crossed the Zendah 
Rud and had a view of the picturesque surroundings. The sun 
sent long slanting rays across the river, whose water was then 
high from the melting snow and whose shimmering surface was 
stirred by the breeze into a thousand sparkling ripples. The 



270 ISFAHAN, THE FORMER CAPITAL 

shores of the river were decked in the rich verdure of spring, 
and the banks were spread with bright cloths which the dyers 
had laid out in the air to dry after their day's work, and these 
heightened the color-effect of the picture. In less than half 
an hour I reached Julfa, the Armenian suburb of Isfahan and 
residence of most of the Europeans, but once a sort of Persian 
Versailles, the royal pleasure-grounds of the Safavid kings. 

Although Julfa occupies an older site, which possibly even 
corresponds in part to the ancient Gabse and Jei, it is neverthe- 
less, in its present form, hardly older than three centuries, and 
owes its name and Armenian population alike to the fortunes 
of war and the wisdom of Shah Abbas the Great. About the 
year 1603 that memorable monarch gained some successes 
in battle over the Turks on the northwestern frontier of his 
realm, whereupon, for political and other reasons, he trans- 
planted several thousand families of Christian Armenians 
from Julfa on the Araxes, which I mentioned in the third 
chapter, 1 and settled them in a new Julfa on the outskirts of 
his capital. Here they nourished under his liberal treatment, 
but not so under his successors. The latter were less generous 
in their policy toward these colonists, whose number conse- 
quently decreased considerably and has continued to fall off 
down to the present time, when the settlement is estimated 
at not more than three thousand souls. They are still Chris- 
tians, have a cathedral (built under the auspices of Shah 
Abbas), and one or two places of worship besides, and among 
them the Mission of the Church of England is doing earnest 
work. 

I had with me letters of introduction to the Mission House, 
and I chanced to overtake my prospective host and hostess on 
the way to their home. They had returned only the day before 
from a fortnight's journey to Shiraz and were so sunburnt that 
I almost mistook them for Eurasians an experience which I 
myself met with later in the journey, when my own face had 
1 Of. pp. 22-23, above. 




THE MEIDAN OF ISFAHAN 
The Royal Mosque and the Porch of the Palace 




PAVILION OF MIRRORS 



JULFA, THE ARMENIAN SUBURB 271 

become almost as dark as a Persian's from weeks of exposure 
under the tropical sun of Southern Iran. These English cousins 
received me hospitably at the Mission, and we soon felt like old 
friends. I had an opportunity to learn something about their 
evangelical labors and also about the philanthropic work in 
medical and surgical lines. The sick and suffering come some- 
times from a distance of several hundred miles for treatment 
and relief; some, alas, come suffering from the effectg of bar- 
barous mutilations, inflicted as punishments, an instance of 
which I saw during my short stay in Isfahan. 

I had almost forgotten the rumor I had heard on the road 
some days before, to the effect that the mail had been robbed. 
On reaching Isfahan, I was told that the presumable culprits 
had been captured, but I thought nothing more about the mat- 
ter. Next morning, as my host and I were riding into town, 
we met a small company of excited people following a man who 
was seated upon a donkey and looked deathly pale. He held up 
his right arm, which was covered with a cloth, and from beneath 
this a purple stream of blood was flowing. We took little heed 
of the fact until, a short distance beyond, we met another, whose 
ashen pallor contrasted with the ruddy drops that trickled into 
the dust. He was on foot and alone. A hundred yards behind 
him followed a third man seated upon a small white donkey, 
which was stained with a similar crimson stream that poured 
from a rag-swathed arm. Some women followed behind him 
and were beating their breasts and tearing their cheeks with 
their nails. The men that marched behind the sufferer ges- 
ticulated and gave vent to imprecations. Seeing that we were 
foreigners, the bleeding man cried out in Persian, k Masters, let 
the sword of your Government be sharpened to avenge the out- 
rage of this unjust punishment ! ' In an instant it dawned upon 
us that these were three of the men who had been convicted of 
the robbery. Their right hands had been cut off. 

The manner in which the punishment is inflicted is as brutal 
as it is summary. The public swordsman, emboldened to his 



272 ISFAHAN, THE FORMER CAPITAL 

task by a dose of hashish, seizes the prisoner by the arm and 
with his scimitar slashes the hand from the wrist. No care is 
taken, as in a surgical operation, to draw back the skin before 
amputating the member, and no attention is paid to the offender 
after the hand is severed, except to thrust the bleeding stump 
of the arm into melted butter and let the victim go. The three 
men whom we had just passed were on their way to the surgeon 
of the Christian Mission for treatment. In each case he had 
first to saw off enough of the bone to allow the skin to cover 
the lopped-off member, before he could begin treatment for the 
injury. 

This was not all of the incident. When we reached the 
Meidan, there was a crowd gathered. A fourth prisoner had 
been punished, but his sentence was death. In such cases the 
execution is swift, but inhuman. Steel hooks are thrust into 
the doomed man's nostrils and his head is quickly drawn back 
by these; a hasty gash then cuts the throat, and the bleeding 
victim is tossed upon the ground to go through the death-agony 
before the eyes of the gaping crowd. I was told that the 
sisters of the dead man were weeping over his lifeless body, but 
I turned away to be spared the sight. It is true that the 
majesty of the law had been upheld and justice administered by 
the sword, and that, too, in the public square near the doors of 
the very bank whose post these highwaymen had robbed. The 
road was now safe rah saldmat bud but what a way of 
making it safe ! 

Capital punishment of this kind may be a necessary measure 
in Persia, but its barbarousness is none the less revolting. 
Nor does it effectually eradicate crime, for it was not long 
after this incident that eight men were punished in a similar 
manner in the same public square. Four of these had their 
throats cut, two had their hands cut off, and the remaining two 
were hamstrung. One of these latter had previously suffered 
mutilation, some eighteen years before, by losing a hand. I did 
not hear whether eventually he died from being hamstrung, 



BARBAROUS PUNISHMENTS AND EXECUTIONS 273 

but this cruel punishment frequently results fatally. In much 
the same fashion, as I learned later, three bakers at Shiraz 
had their tongues cut out for selling their bread at too high a 
price. But, as my correspondent concluded, Iran hamin ast 
4 Persia is always the same ! ' 

While on the subject of Persian punishments, I may men- 
tion one of the death penalties, inflicted by torture, which is 
still in use. - This is the method called ' gatching ' (from gach, 
'gypsum, mortar '). The malefactor is plastered up in gypsum, 
so as to form a pillar by the roadside, but with his face exposed 
to the public gaze, and there left to die. At the time when 
the late Shah was assassinated, five persons who had been 
accused of treacherous designs were thus put to death on the 
main road near Shiraz. At Taft near Yezd, moreover, I passed 
a place where a convict had thus been dealt with a month 
before ; but, more mercifully for the victim, he was immured 
in the mortar with his head downward, so that his sufferings 
were less prolonged. 1 Instances of other barbarous methods 
of execution might easily be multiplied. 

During my short stay in Isfahan I found time to inquire 
whether there were any Zoroastrians engaged in business there, 
as I thought this would be probable in so large a place. It 
seemed the more likely also because there once was a suburb of 
Isfahan called G-abarabdd, 4 Settlement of the Gabars,' to which 
the German traveller Olearius alluded, three centuries ago, 
giving a picture likewise of the Tower of Silence (dakhmah) in 
the vicinity. 2 Thus far on my journey through Persia I had 
not met with a single Zoroastrian, but had only heard of the 
one at Miandoab, who called himself a 4 Babi,' 3 and had learned 
that there were two or three Gabars at Sultanabad, a town 
largely devoted to the weaving of Persian rugs, but I had not 

1 It is not improbable that some of 2 See Olearius, Persianische Reise- 

the pyramids of human skulls left as beschreibung, Hamburg, 1696, p. 293 

monuments by Jenghiz Khan and (* Kebrabath'), p. 296 ('Dakhmah'). 
Timur Lang may owe their origin to 3 See p. 119, above, 

this ghastly practice. 



274 ISFAHAN, THE FORMER CAPITAL 

been able to visit them on my way to Isfahan. 1 This, therefore, 
was my first opportunity to see some of the Persian followers 
of the Prophet of Ancient Iran. 

I found that, although there were some six of them doing 
business in the bazaar, only three resided regularly in Isfahan ; 
the rest were Gabars from Yezd. I have designated them as 
G-abars, after the native fashion, but this term is derogatory, 
being equivalent to 'unbelievers,' and is never employed by the 
Zoroastrians themselves. They designate themselves as Zar- 
dushtidn, 4 Zoroastrians,' sometimes as Bah-lHndn, 4 those of the 
Good Religion,' 2 or again Fdrsls, i.e. 'Parsis,' from Fars, or 
Pars, the old province of Persia Proper. As for the name ' Fire- 
Worshipper ' (Atash-Parast), the Zoroastrians in Persia as well 
as in India object to that title. They claim that they regard 
fire as a symbol or manifestation of Ormazd, as an emblem of 
purity and power, and not as a divinity. It would be equally 
logical, they urge (and I was prepared to hear this argument), 
for Christians to be called 'Cross- Worshippers' after the symbol 
of their faith. 

Through the ever ready kindness of the English Director 
of the Imperial Bank at Isfahan, I obtained the address of the 
principal Gabar merchant, who bore the old Zoroastrian name 
Bahman Jamset, that is, Vohuman Jamshid, and I proceeded 
to call upon him at his shop, as he happened to be in the city at 
the time. He was a man over six feet in height, and large in 
proportion, and he was dressed in a snuff-colored garment 
peculiar to the Gabars. His face, which was smooth-shaven, 
except for the black mustache, was round and full, and his 
features showed a marked contrast to the Mohammedan physi- 
ognomy, in which an admixture of foreign blood is often 
noticeable. His appearance reminded me somewhat of the 
types in the Old Persian and Sasanian sculptures at Behistan 

1 See also my article, The Modern 2 The name Bah-Dlnan is generally 

Zoroastrians of Persia, in Homiletic used by the Parsis in India to denote 
Review, 48. 14-19, New York, 1904. the laity in the Zoroastrian community. 



A ZOROASTRIAN MERCHANT OF ISFAHAN 275 

and Tak-i Bostan, especially the rough bas-relief figures carved 
on the boulder near the famous inscription of Darius, as I have 
already described. l His manners were polite and dignified, 
but I did not understand at the moment a certain reserve in his 
demeanor, nor did I appreciate his almost concerned look when 
I began to question him regarding the subject of religion. I 
afterwards discovered the cause : there were a number of 
Mohammedans present, and he hesitated to speak freely in 
their presence about his faith. A meeting, however, was planned 
for the next day, but I was then unfortunately prevented from 
keeping the appointment, so he sent word forward to his 
brother, Rustam Shah Jahan, at Shiraz, asking him to extend 
hospitality to me in case I should call, and telling him about my 
interest in Zoroastrianism and in its present followers. 

At Isfahan there were a number of other matters which I 
should have liked to investigate, for I felt interested in the 
city in spite of the signs of decay and notwithstanding the fact 
that the Isfahanis have always borne a reputation for untrust- 
worthiness and superficiality, like the tinsel and veneer on some 
of their buildings. I should likewise have been glad to have 
an opportunity of learning more about the condition of the 
people themselves, as contrasted with the past, but that may 
well be left to others, better qualified than myself, to treat. 
I neglected, moreover, to perform what would be regarded as 
a ' Baedeker duty ' in European travel : this was to see the 
Shaking Minarets of Abdallah's Shrine. These slender towers, 
which rise from the roof of the tomb, oscillate back and forth, 
describing an angle of several degrees, when simply pushed by 
the hand. I felt privileged to forego this piece of sight-seeing 
in order to hasten my departure again southward to visit the 
historic sites of Pasargadee and Persepolis. 

I found that for this journey, and as far as Shiraz, I could 
obtain the regular post-horses, as I was on the main route over 
which the Persian mail is carried. I decided, therefore, to give 
1 See pp. 210-212, above. 



276 ISFAHAN, THE FORMER CAPITAL 

up my caravan and allow Shahbas, who had been my chdrvdddr 
and guide for a month, to return to Urumiah with his horses. 
I was sorry to say good-by to Rakhsh, whom I had ridden so 
long ; Safar parted also somewhat reluctantly with his mount, 
which was known as the ' brown horse ' ; a farewell was taken 
of the gray pack-horse who had carried his heavy load so well ; 
and Shahbas made ready to ride his own stumbling jade, 
accompanied by the others. I then wrote a letter of recom- 
mendation for Shahbas, stating how he had served me on the 
journey. Owing to the broad smile which lighted up his 
round face, as it had so often during the four weeks, I prob- 
ably made the report somewhat more favorable than I might 
otherwise have done. I paid him in full for his month's work, 
gave him a gratuity of from two to four krans (twenty to forty 
cents) for each day that he had served me particularly well, 
and added an extra toman (dollar) for every day that he 
gained for me by reaching the several destinations earlier than 
the time scheduled. This concluded our regular business 
transactions ; but I wished to have two of the pack-ropes 
from the baggage because they had been used in drawing me 
up the Behistan Rock. Shahbas still had an eye to the main 
chance, and he made me pay a stiff price for the ropes, but I 
am glad now that I have them as a memento of the climb and 
of the journey. 

Having dismissed Shahbas, I completed my arrangements 
with the head of the chdpdr-khdnah for my first relay of post- 
horses ; but, as it was already late in the afternoon when we 
started, we made only ten or twelve miles, covering the dis- 
tance at a swift gallop, and spent the night at the hamlet of 
Marg. The second day was a record-breaker; we rode sev- 
enty-seven miles in the hours between 5.10 A.M. and 10.45 P.M., 
when we reached Yezdikhast, one of the most curiously situ- 
ated places that can be imagined. It is perched on top of a 
rocky height that looks, as one approaches it, like some gigan- 
tic ship that has been turned to stone in the midst of a river- 



FROM YEZDIKHAST TO MESHAD-I MURGHAB 277 

bed that has been dry for ages. Our third night was passed at 
the walled village of Abadah, eleven farsakhs, or forty-two 
miles, farther on ; and the fourth at the small settlement of 
Deh-Bid, with poor quarters, but the convenience of a telegraph 
office. Finally, at noon on the fifth day, we reached Meshad-i 
Murghab, the nearest halting-place to the Tomb of Cyrus and 
the scenes of the past glory of the Achsemenians. 



CHAPTER XIX 

ANCIENT PASARGAD^E AND THE TOMB OF CYRUS THE 

GREAT 

* The paths of glory lead but to the grave ' 

GRAY, Elegy, 36. 

PASARGAD^E, the subject of this chapter, and Persepolis, the 
topic of the next, are sad themes, in a measure, for both are 
silent cities of a dead past, although each was in turn the capi- 
tal of ancient Persia Pasargadee, the royal seat of Cyrus and 
Cambyses; Persepolis, that of Darius and his successors on the 
Achsemenian throne. Cyrus and Darius still remain in effigy 
of stone, and the vestiges of royal halls, untenanted for more 
than two thousand years, bear witness to the departed splendor 
of a period of grandeur ; but ruin reigns supreme, and even the 
tombs that housed the bodies of the dead kings have been 
crumbling for ages. Time's relentless touch has worn away 
the clear-cut features of these monuments and destroyed the 
beauty of their lineaments, yet they still endure to mark daily 
by their shadows the advance of centuries across the dial of 
eternity and to give tangible evidence of the ancient magnifi- 
cence of the Persians, whose law, like that of the Medes, knew 
no change and whose sceptre once swayed the fortunes of the 
Eastern World. 

V 

The brilliant career of Cyrus, the founder of the Achaeme- 
nians, took Asia captive by its splendid triumphs, and his suc- 
cesses have thrown such a halo about his memory that it is 
sometimes difficult to view the events of his reign in their true 
light and separate facts from the legends gathered about his 
name. This is due in large measure to an early, and for the 

278 



OF THE" 

UNIVERSITY 

OF 





THE TOMB OF CYRUS THE GREAT 




THE PLATFORM OVERLOOKING PASARGAD^J 



THE ROYAL SEAT OF CYRUS 279 

most part erroneous, identification of Cyrus with the shadowy 
figure of Kei Khosru an identification which lingers still in 
the hearts of the Persian people. Whatever value we are to 
attach to the picturesque accounts of the youth of Cyrus as 
told by Herodotus, Xenophon, and other classical writers, his 
real elevation to power began with his defeating Astyages and 
then overthrowing the Median Kingdom, before the year B.C. 
550. l This vanquisher of Media next conquered Croesus and 
the realm of Lydia, subdued Ionia, Lycia, and Caria on the 
west, reduced a part of the Hindu Kush region in the east, and 
at last humbled the proud city of Babylon, thus rendering firm 
the foundations of his vast kingdom before death stayed his 
hand. He chose as the seat for his capital the spot where 
he had first gained his victory over the Medes. Here he built 
Pasargadse, the royal city whose ruins still cover several miles, 2 
and here he erected his tomb. 

The location of the classic Pasargadse is now acknowledged 
to have been on the Plain of Murghab, about six miles from 
a present small settlement called Meshad-i Murghab, which 
I reached on my fifth day from Isfahan and where I stayed 
over night on three different occasions. Pasargadse is not 
visible from Meshad-i Murghab, as it lies beyond a low range 
of hills that encircle the plain around it. An hour's easy ride, 
crossing once or twice the intersecting channels of the Polvar, 
or ancient Medus, brought me to the foot of the ridge. The 
ascent was rough, but not at all difficult, and as the horses sur- 
mounted the rocky crest a sharp bend in the old caravan-road 
threw open to the view the historic Plain of Murghab in all 
its rich fertility, spreading its green expanse fully nine miles 
in one direction and fifteen in another, and surrounded on 
all sides by hilly barriers. 3 As a battlefield for the hosts of 

1 For details see my article Cyrus 3 For a topographical map of the 
the Great, in New Internal. Encyclop. vicinity, see Perrot and Chipiez, His- 
5. 582-583. toire de V Art, 5. 444. 

2 Strabo, Qeographia, 15. 3. 8. 



280 PASARGAD& AND THE TOMB OF CYRUS 

Astyages and Cyrus it must have been superb, and I could imag- 
ine the women of Persia gathered on the hilltops to the south 
to watch in breathless anxiety the issue of the eventful fray. 1 
And here in the plain I could see memorials of the victory still 
surviving in the granite remains of the capital which Cyrus 
had founded. 

As the rider begins to descend from the northern ridge, 
the first object that catches the eye is a massive platform of 
stone, built on a terrace to the left of the way and overlooking 
the plain below. It is outlined in immense blocks of masonry 
and is believed to have been intended for the foundations of 
an audience-hall of Cyrus. If so, Mohammedan tradition has 
obliterated the historic truth by dubbing the structure Takht-i 
Suleiman, 4 Solomon's Throne,' after the usual manner in which 
it has renamed most objects and places of Ancient Persian or 
Zoroastrian date. This solid piece of masonry is over two 
hundred feet long and fifty feet broad, and in many places the 
beautifully matched stone blocks of the facing are in perfect 
order and still show the mason's building-marks upon them, 
while in others they have been torn away in great rows, to 
furnish material for buildings in after ages. In every instance 
the great cramp-irons that secured the blocks have been gouged 
out, leaving holes that afford nesting-places for hundreds of 
pigeons and other birds. 

Descending from the ridge, we come, in a few minutes' ride 
southward, to the first group of the ruins that are scattered over 
the plain. 2 They form the remains of a solid square building 
which must have been more than forty feet high, but only one 
of its shattered walls is standing. The natives conveniently 
call it ' Solomon's Prison ' (Zinddn-i Suleiman) ; Dieulafoy 



1 Compare the accounts of the battle position of the ruins, see Flandin and 
given by Nicolaus Damascenus and Coste, Voyage en Perse, Ancienne, 
Polyaenus, cited by Gilmore, Frag- 4. pi. 194 (reproduced in Perrot and 
ments of Ktesias, pp. 115-128. Chipiez, Histoire de VArt, 5. 596). 

2 For an outline map showing the 



THE APPROACH TO THE RUINS 281 

believes that it was the Tomb of Cambyses, the father of 
Cyrus ; Curzon agrees that it was a sepulchre, even if he does 
not go so far as to assign it definitely to the father of Cyrus. 
All scholars unite on one point, in comparing it with a similar 
edifice near the tombs of the kings at Naksh-i Rustam, and I 
believe that most of them are correct in supporting the view 
that the edifice was an Achaemenian shrine of fire, as I shall 
maintain in the next chapter. 1 But scarcely a stone of the 
only wall that survives is in its exact position to tell the story 
of the past. The present dilapidation of the building, the hard, 
cold whiteness of the stone, and the contrast which it showed 
to the soft green of April that freshly decked the plain, as it 
does ever anew, made a vivid impression upon me. 

Several hundred yards farther southward is a solitary shaft, 
nearly twenty feet high and broken at the top. It is com- 
posed of three blocks, as shown by my photograph, and looks 
as if it might have formed part of a doorway. Near the sum- 
mit of this column are carved in cuneiform script, in three 
languages, the simple but dignified words, 'I am Cyrus, the 
King, the Achsemenian ADAM KURUSH KHSHAYATHIYA 
HAKHAMANISHIYA.' The same device is repeated on the 
angle-piers of a ruined edifice or court several hundred yards 
farther to the southeast, and it once decorated the top of a high 
round column within this enclosure, but in the latter case the 
inscription has disappeared, and the whole structure is in utter 
ruin. The like is true of a fourth collection of ruins still far- 
ther to the east- southeast on a raised flooring of white stone 
sustaining the pedestals of former columns and bases of ruined 
doorways. 2 But to these I paid less attention, because my 

1 See p. 302, below, and compare 2 All the ruins have been so well 

Justi, Grundr. iran. Philol. 2. 422. described by Curzon, Persia, 2. 71-75, 

For the view that the edifice was a with measurements and observations 

tomb, see Dieulafoy, L'Art Antique regarding their position and state of 

de la Perse, 1. 14-21, cf. also pi. 5; preservation, that I have contented 

and for a summary of the opposing myself with a brief outline without 

theories consult Curzon, Persia, 2. 73. elaborating the notes I made on the 



282 PASARGAD^E AND THE TOMB OF CYRUS 

thoughts were riveted upon a monolith standing alone in the 
plain some distance to the east of the circular column and 
paved court that I have mentioned. I had long known about 
it and had always looked forward to the time when I should 
see it. 

This impressive monument is a huge slab, over twelve feet 
high, five feet broad, and about two feet in thickness. Upon 
its weathered front is carved in low relief the figure of a king. 
On his head there rests a curious crown which shows traces of 
the influence of Egyptian art; but the most striking feature 
of the image is a double set of immense vanlike wings that rise 
from the shoulders and droop almost to the feet. 1 The sculp- 
tured form is the very idealization of sovereignty. The top 
of the monolith, which once was inscribed, is broken off, but 
the missing device, as we know from the drawings of the earlier 
travellers, consisted of the simple words, ' I am Cyrus, the King, 
the Achsemenian ' in keeping with the dignity of the sur- 
roundings. 2 

spot, except where I could add some- the fact that they prove that the face 

thing new. of the king is slightly turned toward 

1 For a good description of the bas- the spectator, as both eyes apparently 
relief as it appeared almost a hundred are shown. The head is not there- 
years ago, see Ker Porter, Travels, 1. fore in profile as all the drawings 
492-496. (including those from Ker Porter, 

2 The inscription is clearly drawn in op. cit. pi. 13, to Dieulafoy, IS Art 
the sketch made in the last century by Antique, 1. pi. 17) represent it. In 
Ker Porter, Travels, 1. 492, pi. 13, addition to this I do not believe that 
and is seen in other early pictures. Dieulafoy is justified in representing 
( For references compare Justi, Grundr. the figure as holding some object 
iran. Philol. 2. 422.) I looked in vain in the hand ; a careful study of en- 
for some trace of the broken piece that largements of my best photograph 
held it. There was a stone lying some convince me that Ker Porter was 
fifty feet to the southeast, but, although right in regarding the hand simply 
it showed some chisel marks, it did not as raised (like the hand of Darius at 
appear to match the capstone of the Behistan and Naksh-i Rustam) and 
monument. The two photographs that Dieulafoy has mistaken the 
which I took (one of which is here feather-tips of the smaller wing for 
reproduced) are interesting not alone an instrument grasped in the hand of 
because they show the present condi- the king. 

tion of the bas-relief, but also from 




I, CYRUS, THE KING, THE ACH^MENIAN 



THE SCULPTURED MONOLITH OF CYRUS 283 

I remounted my horse and, followed by Safar and the guides, 
turned his bridle in the direction of the lonely mausoleum that 
forms the principal object in the plain. To reach the road 
that leads to it, I had to ride nearly a mile west and southwest 
over fields freshly turned up by plows of the primitive Jainshid 
type. 1 In Persia one has little hesitation about riding over 
newly sown ground, for a drop of water from the irrigation 
trenches quickly restores each trampled blade. My thoughts 
were centred only upon the massive stone structure in the 
distance, which looked towerlike enough to merit the name 
ptirgos applied to it by Strabo. 2 

I had long been interested in the accounts which the Greek 
and Latin authors have given of the death of Cyrus, and in 
their descriptions of his tomb. I had devoted considerable 
time some years before to investigating the mooted question 
whether this 4 Grave of Solomon's Mother/ or ' Mosque of the 
Mother of Solomon,' as the natives call it, was actually the 
vault of the great king. 3 After a careful and unbiased study 
of the classical testimony on the subject and a thorough exam- 
ination, on three different occasions, both of the site and the 
building itself, I became convinced that no doubt should be 
entertained on the subject, and that we should accept the gen- 
erally current view that it is the authentic tomb of the founder 
of the Achsemenian dynasty. I shall briefly present the main 
facts that lead to this conclusion, and then describe the sepulchre 
itself. 4 

1 Cf . pp. 85-86, 246, above. pendently of Curzon, Persia, 2. 75-90. 

2 Strabo, Geographia, 15. 3-7: irtpyov On completing my investigations I 
ov /j^yav. found that he had covered the field 

3 The Persian designations are so thoroughly and come to the same 
Kabr-i Mddar-i Suleiman and Masjid-i results, that I could confine myself 
Mddar-i Suleiman. For traditions largely to the main points, adding 
referring to the tomb as the burial- comments wherever it seemed neces- 
place of a woman, see Curzon, Persia, sary. I may remark, for example, 
2. 78, 84. that Curzon, op. cit. p. 78, correctly 

4 I made my studies of the classical assigns the first real identification of 
writers on the subject wholly inde- the tomb to Ker Porter (1818), Travels, 



284 PASARGAD^E AND THE TOMB OF CYRUS 

In the first place I may speak of the death of Cyrus. 
According to Herodotus, who wrote less than a century after 
the event, Cyrus met with defeat and death at the hands of the 
Scythian hordes led by Tomyris, queen of the MassagetsB, 
about the year B.C. 530, and the victress thrust his severed 
head into a wine-skin filled with human blood and bade him 
glut to the full his thirst for gore. 1 The Father of History 
adds that this is only one of several accounts of Cyrus's death, 
but the one which seemed to him nearest the truth. The same 
story in an abridged form, but drawn evidently from Herodotus, 
is repeated in the first century B.C. by Diodorus Siculus, who 
states, however, that the Amazon caused the vanquished king 
to be impaled. 2 The narrative, with the details as in Herod- 
otus, is repeated by Justinus (c. A.D. 150) in his epitome of 
the history of Pompeius Trogus, and briefly sketched by Polyse- 
nus (c. A.D. 163). 3> The early historian Berosus (c. B.C. 280) 
says that Cyrus perished 'in the plain of the Dahse.' 4 The 
still earlier authority Ctesias (c. B.C. 400), who knew Persian 
traditions well, states that Cyrus was mortally wounded in 
battle against the foreign hosts of the Derbicse (apparently 
somewhere in the east of Iran), and died three days afterward, 
and his body was conveyed to Persia by his son Cambyses. 5 
Less confidence may be placed in Xenophon's historical novel, 
the Cyropaedia, which depicts Cyrus as passing away at a 
good old age among sorrowing friends, to whom he imparts 

1. 502-508, but Pietro della Valle torica, 2. 44 (Tauchnitz edition). 

( Viaggi, 2. 276 ; Travels, ed. Pinker- 8 See Justinus, Historiae Philip- 

ton, 9. 112) was on the right track picae, 1. 8. 11-13; Polyaenus, Strata- 

when he wrote, under the date July gemata, 8. 28. 

22, 1622, that he arrived at two o'clock 4 Berosus, quoted by Eusebius, ed. 

in the morning at the site of the an- Schone, 1. 30, cf. Miiller, Fragmenta 

cient 'Passargada, where, according Historicorum Graecorum, 2. 605 a, 

to Pliny and Quintus Curtius, was the Paris, 1848. See also Justi, Grundr. 

place of sepulture of Cyrus the Great.' iran. Philol. 2. 421, and Gilmore, 

1 Herodotus (B.C. 484-408), His- Fragments of Ktesias, p. 136, n. 
tory, 1. 201-214. & Ctesias, Fragments, 29 (38-40), 

2 Diodorus Siculus, Bibliotheca His- ed. Gilmore, pp. 136-137. 



THE DEATH OF CYRUS 285 

the sagest counsels and whom he urges to commit his body to 
the ground in the simplest manner possible. 1 Cyrus was about 
seventy-one years old, and the year of his death is placed 
at B.C. 580. On the whole Ctesias's account of the event 
may be regarded as the most trustworthy, if compared also 
with the statements of Arrain, Strabo, Pliny, Quintus Curtius, 
and Plutarch regarding the tomb. 2 From the statements of 
these writers, which I shall next give, we may accept it as a 
fact that the body of Cyrus was laid to rest here at Pasargadse, 
and it is not impossible that it was first coated with wax in the 
custom of the Persians, or perhaps embalmed after the manner 
of the Egyptians. 3 

Arriaii, the Greek historian who lived early in the second 
century of the Christian era and drew material from the writ- 
ings of Aristobulus, who accompanied Alexander the Great 
on his Eastern Campaign, gives an excellent description of the 
tomb of Cyrus and of the visit which Alexander paid to it. 4 He 
describes the mausoleum as standing in the midst of a park 
surrounded by a grove and rich meadows of grass. The tomb 
itself stood on a rectangular base of stone and the sepulchre 
is described as ' a stone house, 5 roofed over, and having a door 
so small as to be difficult to enter even for a man of no large 
stature.' In this 4 house' the body of Cyrus was laid in a 

1 Xenophon, Cyropaedia, 8. 7. 1-28. in the next chapter, may be consulted. 
There is a suggestion of a dying- For the suggestion regarding embalm- 
speech also in Ctesias, Fragments, ing, see Curzon, Persia, 2. 80, n. 1. 
29 (39). * Arrian, Anabasis, 6. 29. 4-11. A 

2 See also Katz, Cyrus des Perser- version of the passage may be found 
konigs Abstammung, Kriege und in Curzon (Persia, 2. 79-80), and for 
Tod, pp. 36-42, Klagenfurt, 1895 ; that reason I merely paraphrase the 
and compare Justi, Grundr. iran. contents, adding the original Greek 
PhiloL 2. 421, n. 3. wherever it seems necessary. For 

8 I have referred elsewhere to the another rendering of the original, see 

statement of Herodotus (1. 140) that Anabasis of Alexander and Indica, 

the Persians coated the bodies of their translated by E. J. Chinnock, pp. 340- 

dead with wax before interring them. 341, London, 1893. 
The remarks upon the tombs of Darius 6 Gk. ofroj^a \idtvov. 

and those of the later Achsemenians, 



286 PASARGADsE AND THE TOMB OF CYRUS 

' golden coffin ' l that rested ' upon (or beside) a couch, the 
feet of which were of hammered gold.' 2 Under this catafalque 
carpets of royal purple were spread, and over it was laid a 
covering of Babylonian tapestry, while around it were lying 
rich vestments of purple and other colors, costly jewels and 
precious stones, placed doubtless on the ' table,' which is also 
mentioned. 3 When Alexander visited the tomb, he found that 
it had been rifled of its treasures, the body had been thrown 
out of the coffin, which was broken and battered, for the 
plunderers found it too heavy to drag with them, and only 
the casket and catafalque remained. 4 On discovering the out- 
rage Alexander gave orders that the body should be replaced 
in the coffin and that everything should be restored to its 
former condition. He obliterated the doorway, closed up the 
entrance with a stone, cemented it with mortar, and sealed it 
with his own signet. 5 

The account of Strabo (c. B.C. 30) is substantially the 
same, bat somewhat less detailed ; he calls the building ' a 
tower of no great size,' and adds, on the authority of Onesi- 
critus, who was with Alexander, that this tower had 'ten 
tiers, or stages,' and the body of Cyrus lay in the ' upper- 
most stage,' an allusion evidently to the terraced courses and 
plinth on which the mausoleum stands. 6 Plutarch (about A.D. 
50), like his two predecessors, speaks of the inscription which 

1 Gk. irve\ov xpwijv. the COUCh ' (TT\T)V TTJS irvt\ov Kal TTJS 

2 The Greek words literally mean jcXfpip) . 

that the couch stood ' beside the coffin ' 5 Gk. rb vrweiov rb f3a(ri\iK6i>. 

(K\IVTJV Trapi ry True'Xy), whereas a 6 Strabo, Geographia, 15. 3. 7, 8, 

statement that follows seems to imply -n-vpyos and rbv ntv irvpyov 8ei<d<rTeyoj> 

that the allusion is to a catafalque . . . r^ avwrdTw <rrtyri. Compare also 

' upon which (lit. ' in the midst of the Falconer's translation of Strabo, 3. 

couch') rested the coffin that contained 133-134, Bonn's Classical Library, 

the body of Cyrus' (iv /dry 8t rfc London, 1857. The idea of the 

K\tm)$ i) TTI>\OS fKeiro ?/ rt> ffd^a. Tov ' tower ' appears again in the Latin ver- 

Kvpov %x v * a }- sion of Callisthenes made in the third 

8 Gk. Kal rpdirefa l/ceiro. century A.D. by Julius Valerius, who 

4 Lit. ' he found that everything had uses the term turris in his Alexandra 

been carried off except the coffin and Polemi, 2. 29. 18 (Teubner edition). 



CLASSIC DESCRIPTIONS OF THE TOMB 287 

Cyrus had bidden to be placed on the tomb, and states that 
the name of the ghoul who had desecrated the vault was 
Polymachus. 1 The later and less trustworthy Quintus Curtius, 
on the other hand, says that Alexander was disappointed at 
finding Cyrus simply interred with his shield, two Scythian 
bows, and a sword, and not with silver and gold, as was re- 
ported; whereupon the Macedonian placed a golden crown 
upon the coffin and covered the sarcophagus with his own 
rich cloak. 2 The incidental statement of Pliny the Elder 
(d. A.D. 79) adds information that is important both for the 
identification of the tomb and its site, when he says : ' The 
Magi hold the fortress of Pasargadse, in which is the tomb of 
Cyrus.' 3 The allusion is in harmony, moreover, with Arriaii 
and Strabo, who say that the Magians were the hereditary 
guardians of the tomb, dwelling near it and offering a sheep 
every day and a horse each month as a sacrifice. 4 

As we ride nearer to the sepulchre the details of the classic 
descriptions become even more clear. The structure, which 
in the distance might be spoken of as a purgos, now looks 
like 4 a house of stone, roofed over,' as Arrian says, and the 
terraced steps (though seven, not ten, in number) answer to 
the statement of Onesicritus in Strabo's account. 5 The door, 
which faces west, or rather northwest, is strikingly small, and 
around the tornb there are the remains of what was once a 
colonnade that formed a rectangular enclosure around the 
tomb. With the exception of a few pillars which, though 
broken, show where the entrances must have been, most of 
the columns have disappeared or lie tumbled about in confusion 
upon the ground. 6 A score of Mohammedan graves have 
been crowded within the area next to the tomb so as to be as 

1 Plutarch, Alexander, 69. 1-2. Geographia, 15. 3. 7 (the latter says 

2 Quintus Curtius, Hist. Alex. 10. 1. they 'received ' these animals). 
30-32. 6 On this subject, see also Curzon, 

3 Pliny, Historia Naturalis, 6. 26. Persia, 2. 82. 

29, 116. Ker Porter, Travels, 1. 499, re- 

4 Arrian, Anabasis, 6. 29. 7 ; Strabo, ported seventeen columns as ' still erect, 



288 PASARGAD^E AND THE TOMB OF CYRUS 

close as possible to this shrine of 'Solomon's Mother.' In 
further confirmation of Arrian's statement concerning 4 a small 
house for the Magi who guarded the tomb of Cyrus,' l we find, 
about a hundred yards north of the mausoleum, the foundation 
of a building which may have been at the same time a sanctu- 
ary and a residence for the priests; but practically nothing 
remains of the structure except the base upon which it stood, 
and this is now partly buried by a mass of wretched hovels. 2 

From a distance the tomb of C}^rus looks dwarfed by the 
vastness of the surrounding plain, but when viewed near-by, its 
true size becomes apparent, and the nobility of its lines, the 
symmetry of its proportions, and the striking whiteness of the 
marblelike stone of which it is constructed, come out in full 
effect. It stands high upon a terraced base, seven steps of 
which are now visible, and the stones which compose both the 
substructure and the tomb are very massive. The lowest 
stage of the seven terraced steps is a plinth over two feet high, 
nearly fifty feet long, and more than forty feet broad. 3 Each 
of the other stages above this flooring is proportionately smaller 
in area, but not in height, and the combined elevation of the 
pedestal thus formed is more than sixteen feet. The mauso- 
leum itself is about eighteen feet high, the point of its roof 
being nearly thirty-five feet from the ground ; the length of the 
building is about twenty feet and its width seventeen feet. 
I give the more precise measurements in metres below, as I 
took them with considerable care. 4 

but heaped round with rubbish and 8 In Ker Porter's time this was 

barbarously connected with a wall of almost covered, so that he calls it a 

mud.' Each generation will find less 'sort of skirting-stone' and counts 

to record. For a theoretical restora- only six steps (Travels, 1. 499; and 

tion of the colonnade, see Dieulafoy, cf. Curzon, Persia, 2. 77, n. 1). Exca- 

V 'Art Antique, 1. pi. 18, and compare vation might perhaps reveal the ' ten 

Perrot and Chipiez, Histoire de VArt, steps' of Onesicritus. 

6. 698. * The plinth, as nearly as I could 

1 Arrian, Anabasis, 6. 29. 7. measure it, is 13.50 m. long, 12.20 m. 

2 For a picture, see Stolze and broad, and .70 m. high. The next 
Andreas, Persepolis, 2. pi. 130. stage is also .70 m. high ; the third, 



THE QUESTION OF AN INSCRIPTION 289 

The mammoth blocks which make up the tomb and base alike 
are set together with the utmost precision. There is no evi- 
dence of the use of mortar, but iron clamps were employed in 
uniting the masonry, as I afterward learned from Mr. J. R. 
Preece, British Consul at Isfahan, who informed me that some 
years ago a friend of his actually found in situ, on the east side 
of the tomb, one of these great clasps binding two blocks to- 
gether. Nature has added to her destructive forces an extra 
one : several bushes of the evergreen type have taken root in 
the interstices worn between the great stones of the terraced 
steps, and a small tree has sprouted out from the roof ; both of 
these agents, as time goes on, will take part in bringing about 
the general ruin of the monument. 

It is natural to suppose that an inscription of some sort 
adorned this resting-place of the mighty dead, and we have 
the authority of the Greek writers for asserting that there 
was such an epitaph. Both Arrian and Strabo say that Aris- 
tobulus, who was appointed by Alexander to restore the tomb 
after it had been desecrated, mentions a Persian inscription, to 
this effect : l 

O MAN, I AM CYRUS, THE SON OF CAMBYSES, WHO 
FOUNDED THE PERSIAN EMPIRE AND WAS KING OF 
ASIA. GRUDGE ME NOT THEREFORE THIS MONU- 
MENT. 

Strabo, after repeating this epitaph, which he says Aristobu- 
lus quoted from memory, adds that Onesicritus says 'the in- 
scription was in Greek, engraved in Persian characters . . . 
and there was another in Persian of the same import ' : 2 

I, CYRUS, KING OF KINGS, LIE HERE. 

1.02 m.; the fourth, also 1.02 m. ; the total height of about 11 m. above the 

fifth, .53 m. ; the sixth, .54 m. ; the level of the ground. 

seventh, .53 m. The sepulchre is l Arrian, Anabasis, 6. 29. 8. 

6.24 m. long on the outside; 5.26 m. 2 The Greek words form a hexameter 

broad ; and about 6 m. high to the line : 'Ev0d8' ^yd> KeT/j.at \ Kupoj pa<ri\efo 

top of the pointed roof, which has a /SatrtX^wv Strabo, Geog. 15. 3. 7. 



290 PASARGAD^E AND THE TOMB OF CYRUS 

Plutarch has a statement to the same effect, and relates that 
when Alexander found the tomb violated by Polymachus and 
read the epitaph, he ordered the inscription to be engraved in 
Greek letters underneath, and it read as follows : 1 

O MAN, WHOSOEVER THOU ART, AND WHENCE- 
SOEVER THOU COMEST (FOR I KNOW THOU WILT 
COME), I AM CYRUS, WHO FOUNDED THE EMPIRE 
OF THE PERSIANS. GRUDGE ME NOT THEREFORE 
THIS LITTLE EARTH THAT COVERS MY BODY. 

The tone of the inscriptions sounds genuine, especially in the 
combination of proud glory and deep humility, and the brief 
line given by Onesicritus i, CYRUS, KING OF KINGS, LIE 
HERE would answer admirably to a presumable Persian 
inscription ADAM KURUSH KHSHAYATHIYA KHSHAYATHIY- 
ANAM, 4 1, Cyrus, King of Kings, (lie here),' with the customary 
omission of the verb, which Onesicritus would naturally supply, 
even if the authenticity of the version be questioned because the 
Greek words form a hexameter. A single glance at the fagade 
of the tomb shows a large stone over the doorway, which looks 
as if designed for holding an epitaph, but I could not find a trace 
of a letter upon it, though I examined it with care, nor do any 
characters appear in the photographs which I took. But the 
original letters may have been destroyed, or possibly they 
were carved on tablets attached to the wall, as seems likely 
if we notice the holes on each side of this slab over the 
entrance. In fact I see no reason to doubt the existence of 
an inscription when Aristobulus and Onesicritus visited the 
tomb with Alexander, any more than we might now argue that 
the bas-relief of Cyrus never had an inscription, because one 
no longer exists, although there was one less than a hundred 
years ago. 

The entrance to the tomb is low and narrow, as the Greek 
authorities state. The height of the doorway is only 4 ft. 
1 Plutarch, Alexander, 69. 2. 



THE INTERIOR OF THE SEPULCHRE 291 

2 in. and its width 2 ft. 7 in., 1 and it is necessary to 
crouch in order to pass through, as Arrian affirmed. The 
original door to the vault was probably a heavy stone swing- 
ing on pivots, such as may be seen at the supposed tomb of 
Esther and Mordecai at Hamadan and also in the square monu- 
ment at Naksh-i Rustam, 2 but I do not recall seeing the socket- 
holes, as in those cases. Dieulafoy's ingenious suggestion of 
two doors, an outer and an inner door, so arranged as to open 
upon each other, but not both at the same time, seems plaus- 
ible 3 ; and even if the original means of closure have long 
since disappeared, there are actually to be seen two rickety 
wooden doors which guard the ingress, as the thickness of the 
wall allows space for two. But the second of these wretched 
board structures was off its hinges, when I visited the tomb, 
and was lying in one corner of the vault. 

Crawling through the low entrance, without having the slight- 
est objection raised by the bystanders, who I thought might 
prevent me, I found myself within the chamber where death 
had held his court. The room is 10 ft. 5| in. long, and 7 ft. 
7 in. wide, and 7 ft. 11J in. high. 4 Two enormous blocks that 
look like marble form the floor, and the side walls and flat ceil- 
ing are composed of stones equally large. In one part of the 
eastern wall, facing the doorway, there is a great gaping chink, 
and the block of the floor at the same end shows a rough sort 
of depression which suggests the idea that something had been 
scooped out or forcibly wrenched away, perhaps by robber 
hands. It may have been that the sarcophagus stood here. 5 I 
could picture its position and I lay down to measure my length 

1 The dimensions in centimeters as 4 The measurements as I took them 

1 took them are: height, 126 cm.; in meters are : length, 3.18 m.; width, 
width, 80 cm. Curzon gives 4 ft. 3 in. 2.30 m. ; height, 2.12 m. They are 
by 2 ft. 3 in. ; Ker Porter, 4 ft. by practically the same as Curzon's, who 

2 ft. 10 in. gives: length, 10 ft. 5 in.; width, 7 ft. 

2 See p. 168, and note 2, above. 6 in. ; height, 6 ft. 10 in. 

3 See Dieulafoy, V Art Antique, 1. 6 See Arrian's description, p. 286, 
48, fig. 54. above. 



292 PASARGAD^E AND THE TOMB OF CYRUS 

near the depression in order to determine whether the space 
were in proportion to a human form. I found there would 
indeed have been room enough, and space besides for the table 
or couch holding the purple vestments, the sword, and the other 
regal insignia which Arrian and Strabo enumerate. 

The condition of the royal chamber is not what it once was. 
The wall to the right, or on the south side, has been desecrated 
by a modern Persian inscription and verses from the Koran. 
They are handsomely carved, it is true, and are surrounded by 
an ornamental border shaped like a prayer-niche, but they are 
out of place in the tomb of Cyrus. Innumerable graffiti, 
scratched by Orientals with the scribbling craze, cover the 
walls. One of the names was in Pahlavi characters and inter- 
ested me, as it was that of a Zoroastrian priest, Mobed Ormaz- 
dyar Bahram. As a lineal descendant of the race of the Magi 
he had more right perhaps than the others to carve his name in 
the tomb, for the Magi were the traditional guardians of the 
tomb of Cyrus. A worn copy of the Koran lay upon the floor, 
and the wind sweeping through the vault turned its pages. 
A manuscript of the Avesta would have been more appropriate 
in this chamber. But most inharmonious of all, for it hung 
over the place where the body must have lain, was a cord with 
an incongruous collection of worthless trash in the way of 
votive offerings. A piece of rag, a bit of brass, a fragment of 
a lamp, a bell, a copper ring, and what not, made up the motley 
string. They were the customary donations of pilgrims as 
mementos of the journey or as talismans for a safe return. 

Happily the note so out of tune with the historic surround- 
ings lasted only for a moment. An instant later the setting 
sun streamed through the doorway and flooded one corner of 
the dingy vault with a gorgeous splendor. The image of the 
kdvaya hvarenah, the 4 Kingly Glory,' or symbol of sovereignty 
in the Avesta, flashed across my mind. In ancient times a 
reflection of this same divine light was believed to shed a halo 
about the person of the King of Kings. Its sacred majesty 



THE OBSEQUIES OF THE KING 293 

was shining this instant from heaven around the tomb of 
Cyrus and made it seem too hallowed to remain. As I slowly 
descended the deep steps and mounted my horse, the sun sank 
low behind the western hills. I turned for a last look at the 
historic shrine. A vision seemed to rise before my view, and 
I saw in fancy the scene of the last obsequies of the Great 
King. From the classical writers we know with what pomp 
and magnificence the processions of the Persian kings were 
conducted during their lifetime; the richly caparisoned 
horses, chariots of state, purple robes, heavy accoutrements, 
and barbaric splendor were always present in gorgeous pro- 
fusion. With no less pomp, but with greater solemnity, must 
the sorrowing nation have borne their dead hero to the tomb. 
His body, not left in Magian fashion to be torn by dogs 
and birds, but covered with wax or perhaps embalmed, was 
laid in a gold-adorned casket and carried to the sepulchre, 
attended by the flower of the Persian army. I still could hear 
the tramplings of the horses that led the funeral train ; the 
measured tread of the soldiers in clanking armor rang dully on 
my ear; the smoke of imagined incense rose heavenward to 
Auramazda from the huge urn holding the sacred fire ; and 
the chanting voice of the Magian priest intoning perchance 
the Zoroastrian psalm Kam nemoi zftm ' to what land am I 
going ' beat rhythmically through my brain. The Great 
King Cyrus was no more ! The shroud of darkness fell like a 
pall upon the plain and the moon rose slowly over the distant 
hills. 



CHAPTER XX 
PERSEPOLIS AND ITS MONUMENTS 

4 Among the ruined temples there, 
Stupendous columns, and wild images 
Of more than man, where marble demons watch 
The Zodiac's brazen mystery, and dead men 
Hang their mute thoughts on the mute walls around.' 

SHELLEY, Alastor, 116-120. 

THE scene now shifts from Pasargadse to Persepolis, the 
royal seat of Darius, Xerxes, Artaxerxes, and their successors, 
who bore in turn the title 4 King of Kings,' having inherited 
the throne of Cyrus through a side line, as his son Cambyses 
had died childless. 1 These monarchs located their capital 
some forty miles south of Cyrus's city, at the site now 
marked by the great platform of Takht-i Jamshid and the 
ruins of the city of Stakhra, as well as perpetuated by their 
tombs at Naksh-i Rustam. The new capital may have been 
called Pdrsa-karta, ' City of the Persians,' like the older Pasar- 
gadee, for the Greeks appear simply to be paraphrasing the name 
when they refer to the city as Perse-polish It is not improb- 
able, moreover, that the name Stakhra, 4 Strong,' still preserved 
among the natives as Stakhr or Istakhr, and which we can 



1 See pp. 26, 180, above. The date 2 The earliest occurrence of 

of Cambyses's death was B.C. 522. 71-0X1$ in Greek appears to be in the 

Herodotus (History, 3. 61-66) states fifth century B.C., after the Persian 

that it occurred at the Syrian Ecba- War, as we then find the word used 

tana ; Ctesias (Fragments, 43-44, ed. by ^Eschylus, Persians, 65, apparently 

Gilmore, pp. 144-145) says that the with a punning allusion to 'destroy- 

event occurred at Babylon, and adds ing (TT^O-IS) cities (TnSXts).' This I 

that the body was brought back to Per- believe to be the best interpretation of 

sia. Yet to this day no man knows A the passage in question. 
the place where Cambyses is buried. () 

294 



I 
^ 9 

H ^ 



w 

II 

I 1 

H 





FROM PASARGAD^E TO NAKSH-I RUSTAM 295 

trace back for centuries, if not to Achaemenian times, may 
have designated the city in the plain north and west of the 
platform, that is, the abode of the people in distinction from 
the residence of the kings on the grand terrace. 1 Be that as 
it may, the monuments in this vicinity are the most interesting 
and historic in all Persia ; Susa alone can make any claim to 
comparison with them. 

To reach Persepolis, we strike southward toward the Plain 
of Mervdasht. The road runs at first through a mountain 
gorge, picturesque in wild scenery, but dangerous at night 
because of its rugged track and robbers. The river Polvar, 
the classic Medus, pushes its way with turbulent stream 
through the craggy defile. A part of the road above its rocky 
bed exhibits one of the most remarkable pieces of ancient 
engineering in the Orient. For a considerable distance, 
through the solid limestone rock, a narrow causeway was 
hewn ages ago to afford, as it still does, a passage for cara- 
vans on their route from the south to the north of Iran, and 
an ingress more than once for great armed forces. It is known 
as the Sang -Bur, 4 Rock- Cutting,' or the Tang-i Buldghi, 
'Water-stream Pass,' and is thought by some to be identical 
with the mountain gorges of Vash-Shikuft, mentioned in the 
Bundahishn. 2 

1 On the problem of the names 2 Bd. 12. 2, 21 ; so Justi, in Indo- 

Persepolis and Istakhr, compare Cur- germanische Forschungen, Anzeiger, 

zon, Persia, 2. 132, n. 2, 133,148, 187; 17. 106. For a picture of the pass, 

also Noldeke's article on Persepolis see Stolze and Andreas, Persepolis, 2. 

in Encyclopaedia Britannica, 9th ed., pi. 127 ; and for descriptions, cf. 

18. 557-560. The Greeks and Latins Curzon, Persia, 2. 90 ; Browne, A 

naturally did not observe this distinc- Year Amongst the Persians, p. 243. 

tion, and we can understand how the In the opinion of Justi (IF. Anzeiger, 

Tabula Peutingeriana should in late 17. 106, cf. Grundr. iran. Philol. 2. 

Parthian times speak of ' Persepolis, 425) it was on a mountain in this 

the emporium of Persia,' although vicinity that the Median pretender 

this could only refer to the city itself, Gaumata, the False Smerdis, first 

because the place had lost its prestige asserted his claim to the sovereignty 

as a capital under the Arsacids. See of Persia, only to be overthrown by 

Tomaschek, Zur historischen Topogra- Darius. 
phie von Persien, pp. 166-175. 



296 PERSEPOLIS AND ITS MONUMENTS 

From the ravine we emerge into a succession of valleys 
between hills and cliffs. At this point my cavalcade overtook 
a band of Cossacks in the employ of the Shah. These finely 
mounted horsemen had been sent down to clear up the road 
which was infested with highwaymen (rdh-zari), and they 
were not long in finding an opportunity to exercise their 
functions. A shepherd came past with tears in his eyes, com- 
plaining that he had been robbed of a sheep by a peasant who 
was acting as 'a guard of the road.' The Cossacks pursued 
the offender to the hills, quickly caught him, pinioned his arms 
behind him, and marched him at the rifle's muzzle to the 
nearest village, where punishment, I presume of the cruellest 
kind, was inflicted upon the culprit. 

At Sivand, the station beyond, I found no occasion to wait 
for a longer time than to change horses, and then resumed the 
trail near the Polvar, though I regret that I missed seeing the 
famous Pahlavi inscription in the hills near the village of 
Hajiabad, some miles below Sivand. I have in my possession, 
however, a picture of the tablet taken by Mr. A. O. Wood, 
of the bank at Isfahan, and there is a large photograph of the 
inscription in Stolze's work on Persepolis, besides copies that 
have been made by others. 1 The writing on the stone is in 
Chaldseo-Pahlavi and Sasanian Pahlavi, and it appears to record 
a remarkable shot with an arrow by King Shahpur, or Sapor I, 
of the House of Sasan. 2 

The afternoon was considerably advanced when I reached 
Naksh-i Rustam and the tombs of the AchaBinenian kings. 
Here in the face of a long high bluff are hewn four sepulchres 
belonging to the elder kings of the second line, Darius, Xerxes, 
Artaxerxes I, and Darius II. These rock-cut vaults lie five or 
six miles to the north of the great platform where once stood 

1 See Stolze and Andreas, Per- hesh, pp. 83-84, Copenhagen, 1851 ; 

sepolis, 2. pi. 126; Ker Porter, Travels, cf. also Curzon, Persia, 2. 116. 
1. 613 ; Flandin and Coste, Voyage en 2 See West, Grundr. Iran. Philol. 

Perse, Ancienne, 2. pi. 164 ; Wester- 2. 77. (Hajiabad is not to be confused 

gaard, in the appendix to his Bunde- with the place mentioned on p. 252.) 



THE ROYAL TOMBS AT NAKSH-I RUST AM 297 

the palaces of the kings and behind which are three other 
Achsemenian tombs of a somewhat later date. 

The rocky cliff in which the sepulchre of Darius and those 
of his son, grandson, and great-grandson are carved, resembles 
a jagged wall, over five hundred feet long and between one 
and two hundred feet high. It extends in a generally easterly 
and westerly direction, but makes a rather sharp turn at the 
eastern end, so that we can understand how Ctesias came to 
speak of it as a 'double mountain' Sicro-bv opo?. 1 The 
natives call it Husein Kuh, ' Hill of Husein,' or more often 
Naksh-i Rustam, 4 Rustam's Picture,' from a mistaken idea that 
the equestrian statues of the Sasanian kings at its base repre- 
sent Rustam and his famous charger. 

The four tombs which are hewn in the bosom of the rock 
are practically of the same size and dimensions and absolutely 
uniform in their exterior design. The shape of each fagade is 
roughly that of a Greek cross, some seventy feet high and 
sixty feet wide, 2 and the arms are hewn deep into the stone. 
In the middle of each facade a door with decorative lintel is 
cut, but only the lower half is pierced so as to furnish a small 
aperture, the upper part being left solid as a screen. Two 
columns, cut in high relief, stand on each side of the doorway. 
They are capped with the heads of bulls after the characteristic 
manner of the Persepolitan architecture and they support an 
entablature, with ornamental architrave, frieze, and cornice, 
forming a base for the elaborately sculptured panel that fills 
the upper limb of the cross. Here, carved in two rows, one 
above the other, are bas-reliefs representing the vassal nations 
as supporting the staging upon which stands the king, who 
thus makes his enemies his footstool. The monarch is por- 
trayed in the same manner as he is seen on the Behistan sculp- 
tures, bow in hand, 3 but his attitude is now that of worship 

1 Ctesias, Fragments^ 46 (15), ed. height and. the breadth was doubtless 
Gilmore, p. 150. designed for the sake of effect. 

2 The disproportion between the 3 See p. 180, above. 



298 PERSEPOLIS AND ITS MONUMENTS 

before the sacred fire, over which floats the familiar winged 
effigy of Auramazda with the emblem of the sun shining in 
the background. 

The identity of only one of the four tombs is positively 
known ; this is the tomb of Darius. It is the third sepulchre 
to the right, near the point where the cliff makes the sharp 
angle. The identification is made by means of two trilingual 
inscriptions, of like contents, carved near the figure of the 
king and around the doorway. 1 In some sixty lines the king 
glorifies Auramazda, enumerates the nations that acknowledge 
his sway, and exhorts the people not to depart from 'the 
Way which is Right.' All the bas-reliefs, including the sculp- 
ture of the king, the two figures behind him, which are known 
from inscriptions to be Gobryas and Aspathines, 2 as well as 
the effigies of the vassal nations, have suffered much from the 
elements and from lapse of ages, but the corresponding carvings 
on the other tombs give considerable aid in restoring them. 3 
By comparing the national garb, the characteristic features, 
and the position of the figures with the names enumerated in 
the adjoining inscription, we may identify to-day almost every 
one of the nations represented on the bas-relief. 4 The entrance 
to the sepulchre is so high from the ground that it is impossible 
to reach it except by the aid of ropes or ladders. The interior 
of the vault has been several times examined and described by 
travellers. 6 It consists of a passagelike chamber into which 

1 These are the well-known inscrip- long for the publication of Dr. F. 
tions Naksh-i Rustam a and b (Weiss- Sarre's admirable pictures, referred 
bach, Die altpers. Keilinschr. pp. to by Andreas, in Verhandl. des 13. 
34-36 ; Spiegel, Die altpers. Keilin- Internat. Orientalisten-Kongresses, p. 
schr. pp. 52-57). The lower one (b) 96, Leiden, 1904. 

of the two inscriptions is now almost 4 See Andreas, op. cit. pp. 90-07, 

illegible. and cf. Justi, (,'rnndr. Iran. Philol. 2. 

2 See Weissbach, op. cit. p. 36 ; 454-455. MM. Babin and Houssay, 
Spiegel, op. cit. p. 58, and compare collaborators of M. Dieulafoy, in 1885, 
p. 181, above. actually discovered names carved un- 

8 Photographs may be seen in the der seven of the figures ; cf. Perrot 
well-known works of Stolze and of and Chipiez, Histoire de VArt, 5. 622. 
Dieulafoy, and we may look before 6 For a plan of the interior of the 






THE SEPULCHRE OF DARIUS 299 

the low door opens, and opposite the entrance is a recess whose 
floor is higher than the level of the passageway. Into the 
stone flooring of this are hewn three deep troughs to serve as 
sarcophagi, probably for the king and whichever two persons 
he regarded as nearest to him, while in the extension of the 
passage to the left six more such loculi are chiselled. All the 
receptacles are entirely empty, and the sole tenants of this 
lofty catacomb are bats and birds. 

Two interesting stories in connection with this sepulchre are 
told by Ctesias, who must have had them at first hand during 
his residence in Persia as Greek physician to Artaxerxes. In 
his brief notice regarding the tomb of Darius he says (and I 
translate fairly literally) : ' Darius ordered a tomb to be made 
for himself in the Double Mountain, and the work was brought 
to completion ; but when he wished to inspect it, he was dis- 
suaded from so doing by the Chaldseans (the Magian sooth- 
sayers) and by his parents. His parents, however, were 
anxious to go up to see it. As they were being drawn up, the 
priests who had hold of the ropes saw some serpents and became 
so frightened that they let go the ropes, and the parents of the 
king fell and were killed. The grief of Darius was so great 
that he caused to be beheaded forty of those who had pulled on 
the ropes.' 1 The other story that is told by Ctesias is to the 
effect that Bagapates, the favorite eunuch of Darius, lived by 
his dead master's tomb for seven years until death released him 
from his devoted charge. 2 

The other tombs apparently belonged to Xerxes, Arta- 
xerxes I, and Darius II ; but in the absence of inscriptions we 
can only surmise how they were occupied respectively. 3 It is 
natural to suppose that the tomb of Xerxes was next to that 

tomb, see Flandin and Coste, Voyage 2 Ibid. 59 (19), ed. Gilmore, p. 152. 

en Perse, Ancienne, 4. pi. 170 ; cf . 3 See, for example, Nb'ldeke's article 

also Perrot and Chipiez, op. cit. on Persepolis, Encyclop. Brit. 9th ed. , 

5. 626. 18. 558. A less likely assignment may 

1 Ctesias, Fragments, 46 (15), ed. be found in Dieulafoy, L* An Antique, 

Gilmore, p. 150. 3. 2, n. 1. 



300 PERSEPOLIS AND ITS MONUMENTS 

of his father, but its position, whether to the right or the left, 
would affect the assignment of the other two. If we assume, as 
I think we are entitled to assume, that the three sepulchres 
which are in the main face of the rock were cut one after the 
other in regular order, 1 and that the one in the bend to 
the extreme right (the so-called 'first tomb') was cut last, 
and partly for that reason is better preserved, then the vault 
of Xerxes would be on the left of that of Darius, as we face 
the hill, the vault of Artaxerxes Longimanus at the end, 
and the vault of Darius II, latest of them all, in the angle at 
the extreme right. 2 On the other hand, if the so-called ' first 
tomb ' at the extreme right be assigned to Xerxes, then Arta- 
xerxes would have occupied the ' third tomb ' and Darius II 
the last. 8 

Owing to the peculiar conformation of the cliff that has been 
already referred to, the 4 first tomb ' faces almost to the west, 
whereas the other three face nearly to the south. The more 
protected position of this tomb, and its greater inaccessibility, 
for it is the most difficult of the four to reach, may account for 
its being better preserved, even if one is not prepared to grant 
that it is more recent than the others. Passing by the tomb of 
Darius, already described, I may remark that the facade of the 
' third tomb,' the one on its left, is comparatively well preserved, 
much better in fact than that of Darius, and we must regret 
that we cannot be sure whether we are looking at the sepulchre 
of Xerxes or not. 4 The fourth and last of the group is nearer 
the ground than the others and is the most damaged of them all. 
Like the others, moreover, it is empty. 

Along the base of the rock below the tombs there is carved a 
series of seven panels which date from a later dynasty, as they 

1 It is worth noting that the Sasa- 8 Such seems to be the view of An- 
nian bas-reliefs are sculptured only in dreas, op. cit. p. 96. 

the base of this main wall, and not 4 For a good photograph of this and 

below the tomb in the cliff at the bend. the other facades, see Dieulafoy, VArt, 

2 So also Justi, Empire of the Per- Antique, 3. pis. 1-3. A picture of the 
sians, p. 203. first tomb is also given below. 




FIRST SASANIAN SCULPTURE AT NAKSH-I RUSTAM 




FOURTH SASANIAN SCULPTURE AT NAKSH-I RUSTAM 



SEVEN S AS AN I AN SCULPTURES 301 

bear Sasanian bas-reliefs of the third and fourth centuries of the 
Christian era. If we adopt the same order of enumeration from 
east to west as for the tombs, we may describe the first bas- 
relief as located between the first and second sepulchre and 
adjoining a large incised space that is vacant, except for three 
rectangular holes and an unimportant modern Persian inscrip- 
tion dated early in the eighteenth century. 1 The bas-relief itself 
represents a Sasanian royal group, one figure in which is a 
woman. The scene portrayed has been variously interpreted 
as representing Shahpur I (A.D. 241-272) and his queen, or as 
Varahran II (A.D. 275-293) and his chief royal consort, or else 
as figuring the marriage of Varahran V, Bah ram Gor (A.D. 420- 
438), with an Indian princess. 2 The second and third bas-reliefs 
(the latter nearly buried by the sand blown against it) 3 are eques- 
trian sculptures, carved one above the other in the space below 
the tomb of Darius. They represent two stages of a combat on 
horseback, in which the king on the left triumphs over his 
enemy on the right, whose spear is broken and whose horse is 
worsted in the fray. These panels may, perhaps, commemorate 
victorious engagements of Varahran IV of Kermanshah (A.D. 
388-399). 4 The fourth panel, or that in the lower space be- 
tween the tomb of Darius and its neighbor on the left, has a 
peculiar interest, as it portrays the surrender of the Roman 
emperor Valerian (A.D. 260) to the Sasanian monarch Sapor, 
Shahpur I. The victor, mounted upon a war-horse, triumphantly 
receives the submission of the suppliant who kneels before him. 
The fifth sculpture, like the second and third, represents in 

1 For a note on this modern Per- Persia, 2. 118-119 ; Modi, JEAS. 
sian inscription, see Browne, A Year Bombay Branch, 19. 58-75, Bombay, 
Amongst the Persians, p. 248. The 1895 = Asiatic Papers, pp. 67-84, 
blank space with the three holes, seen Bombay, 1905. 

in the photograph which I reproduce, 3 This partially buried panel is 

appears in the seventeenth-century shown in a photograph by Stolze 

drawings of Chardin, Voyages, 2. pi. and Andreas, Persepolis, 2. pi. 121, 

74 and p. 175. and in the drawings of Flandin and 

2 For the various views, see Justi, Coste, 4. pis. 174, 184. 
Grundr. Iran. Philol. 2. 519 ; Curzon, * Compare also p. 221, above. 




302 PERSEPOLIS AND ITS MONUMENTS 

a spirited manner an engagement on horseback, but the per- 
sonages in the combat have not yet been identified. The sixth, 
near the lower end of the crag, portrays Varahran II and his 
courtiers, and the seventh, or last of the group, represents in 
effigy King Ardashir (A.D. 226-241), the founder of the Sasa- 
nian line, mounted on horseback and receiving from the god 
Ormazd, who is similarly mounted, a ring that symbolizes the 
gift of sovereignty. Trampled beneath the feet of the horses 
lie the prostrate forms of Volagases and Artabanus, the last of 
the Parthian dynasty. 1 

Opposite the fourth tomb, and about twenty yards away, 
is a square building that evidently dates back to Achaemenian 
times and recalls the ruined structure at Pasargadse to which 
I drew attention in the preceding chapter. 2 The natives call 
it Ka'-bah-i Zardusht, ' Shrine of Zoroaster.' Lord Curzon, 
however, like Dieulafoy, has stoutly maintained that the struc- 
ture served as a royal tomb and was, perhaps, the mausoleum 
of Hystaspes, the father of Darius. 3 The majority of Iranian 
scholars, including so distinguished a specialist as Justi, are 
agreed with Ker Porter in regarding the building as a fire- 
shrine, like the modern Parsi sagri that commonly adjoins 
a Tower of Silence. Any one who has visited Malabar Hill 
at Bombay, or crossed the harbor to inspect the dakhmah at 
Ooran, after studying the history of Persia's sacred fire in 
connection with the representations on the coins, will be in- 
clined to agree with this Zoroastrian explanation. Without 
again mentioning the Pasargadse building, it is sufficient to 
compare the kindred structures near Naubandajan 4 and at 
Firuzabad, and the representation of fire-shrines on coins of 
the Parthian dynasty, to become convinced of the sacred 

1 For a fuller description of all 8 Curzon, Persia, 2. 144-147. 
these sculptures, together with abun- * Near Fasa below Fahliyan in 
dant material in the way of bibliog- western Farsistan. For a photo- 
raphy, consult Curzon, Persia, 2. graph of the fire-temple at Nau- 
117-126. baudajan, see Stolze and Andreas, 

2 See p. 280, above. Persepolis, 2. 147. 




SIXTH AND SEVENTH SASANIAN SCULPTURES AT NAKSH-I RUSTAM 




STONE EDIFICE NEAR THE TOMBS AT NAKSH-I RUSTAM 



THE SQUARE BUILDING AND THE FIRE-ALTARS 303 

character of the building. 1 Although not strictly a temple, 
since the Persians had no true temples like the Greeks, it is 
precisely the sort of building that would have been adapted 
to the purpose of preserving the sacred fire which was kept 
burning in some hallowed urn. The absence of windows (for 
the window-spaces are blank) and of a smoke-vent is no 
convincing argument against this view, because smoke was 
regarded as a creation of the evil spirit and every effort was 
doubtless made to provide against its formation. 2 

Leaving this square building and riding around the lower 
end of the bluff we come to two Atash-G-dhs, 4 Fire-Altars,' 
carved out of the living stone and dating back to Achsemenian 
times, according to the generally accepted view, from which 
there is no occasion to dissent. They recalled to me the dditya 
gdtu, or fire-altar, of Avestan days, and I could fancy the 
Magian priest heaping high the incense and sandalwood upon 
the sacred flame on some solemn occasion when the Great 
King, mindful of death in the midst of all his earthly pomp, 
came to offer sacrifice at the royal tombs. 

Ascending the hill so as to overlook the tombs, we see 
near the lower edge of the bluff a low pillar, about five feet 
high, hewn out of the solid rock. 3 This shaft may have com- 
memorated some event no longer discoverable, as it is not 
recorded by an inscription, but I may refer to the fact that 
Darius set up inscribed pillars at Tell al-Maskhutah in Egypt 
and at Chaluf, and also, according to Herodotus, in Thrace. 4 

1 See especially the points made by the surface or as cavities for sheets 
Justi, Grundr. Iran. Philol. 2. 456, and of metal plates or for tiles. 

Empire of the Persians, pp. 203-206. 3 For a somewhat imperfect sketch 

2 Cf. also Justi, Empire of the of the cliff and the pillar, see Flandin 
Persians, p. 205, and Grundr. iran. and Coste, Voyage en Perse, Andenne^ 
Philol. 2. 456. The peculiar pitted 5. pi. 89. 

appearance of the exterior of the 4 For the 0-7-^X77 in Thrace, see 

building is a puzzle to arcbseol- Herodotus, History, 4. 91 ; and for 

ogists. Justi ( 1. c. ) has suggested the others compare Dr. L. H. Gray's 

that it may be caused by spaces appendix to my article in JA08. 21. 

left for securing a stucco spread over 183-184. 



304 PERSEPOLIS AND ITS MONUMENTS 

Still farther up on the bluff there are some level spaces, 
hewn out of the rock and peculiar in their character. They 
are presumed to be the tables on which the dead were exposed 
to be devoured by dogs and birds in accordance with the ordi- 
nances of the Avesta. Because of their possible connection 
with the question of the religion of the Achsemenians, I paid 
more attention to them than I might have done otherwise, and 
I visited the spot twice. I counted three of these repositories 
on my first visit and noticed two more on my second, and pos- 
sibly there may be others that I overlooked. The one to which 
I paid most attention was finished with more care than the rest, 
and the rock had been cut in such a manner as almost to form 
a divan, with a level flooring in front of it, measuring perhaps 
eight feet by ten, but hewn in the roughest style and shape. 
By the side of this quasi-couch of stone and at the head as well 
as near the feet were holes, which suggested to me the idea that 
they might possibly have been intended for use in securing the 
body, as enjoined by the Vendidad, 4 so that the corpse-eating 
dogs and birds might not carry away any of the bones of the 
dead and thus defile the water and the trees.' 1 The difficulty, 
however, is to determine precisely by whom and at what par- 
ticular period this mortuary platform was used, if we regard it 
so and similarly construe the other spaces which are not so 
distinctly marked. The bodies of the kings themselves can 
hardly have been exposed here before being laid in the 
tomb, as those catacombs were evidently designed to hold 
large coffins, if we may judge from the size of the loculi and 
from the description of the tombs back of the platform as given 
by Diodorus, who says that the body was raised by 4 machines ' 
expressly devised for the purpose. 2 Nor does it seem likely 
that the bones were first denuded of the flesh any more than in 
the case of Cyrus, whose 4 body ' (o-w^a) is spoken of by Arrian 
and Strabo when describing the mausoleum at Pasargadse. 3 

1 Vd. 6. 46. See pp. 285-286, above. 

2 Diodorus Siculus, History, 17. 71. 



OF THE 

UNIVERSITY 

OF 




PRESUMABLE ANCIENT DAKHMAHS ON THE CLIFF 305 

The Persians in Achgemenian times, moreover, had not gener- 
ally adopted the Magian fashion of exposing the dead, but 
rather buried the body, merely coating it with wax, as a con- 
cession to the Magi, who followed the strict Zoroastrian law 
for exposing corpses ; 1 and it was not until the time of the 
Sasanians that the custom of exposure became universal among 
the orthodox throughout Iran. For that reason, if we are to 
assign these presumable dakhmahs to Achsemenian times, we are 
entitled, perhaps, to suggest the possibility that they may have 
been first employed by the Magian priests themselves per- 
haps to give currency to this tenet of the religion and then 
may have come into more general use later, especially when the 
city of ^Stakhr was under Sasanian rule. But this is only a 
conjecture, and the whole subject requires further consideration. 

On the highest point of the bluff, and just above the sepul- 
chres, is one other noticeable object. It is a sort of parapet 
cut near the rocky ledge and ascended by five roughly hewn 
steps. It appears to date from the same period as the other 
cuttings and looks as if it might have been for use on some 
occasion when sacrifice was offered to ' Auramazda and the 
other divinities,' 2 or when the king was laid in the tomb, but 
nothing is actually known of its real character. 

After finishing my examination of the necropolis cliff, its 
altars, and its presumably early dakhmahs, I proceeded down- 
ward and remounted my horse to ride toward the river Polvar, 
which winds its way through the plain between Naksh-i Rus- 
tam and the great platform of Persepolis. My ultimate goal 
was the small station of Puzah, on the other side of the Polvar, 
but to reach it I had first to cross some deep irrigation-canals 
and then ford the stream. To make the progress of our little 
caravan slower, the horses persisted in stopping to nibble the 
tempting tops of the barley which had grown high from the 

1 We may deduce this from the ac- 2 Bh. 4. 61, 63, etc. Cf. also Ker 

count given by Herodotus, History, Porter, Travels, 1. 570. 
1. 140. 

x 



306 PERSEPOLIS AND ITS MONUMENTS 

effect of the spring rains, but in this way I had a better oppor- 
tunity to survey the plain. 

The north side of the Plain of Mervdasht is dotted here and 
there with ruins of the ancient city of Stakhra, the capital as op- 
posed to the platform, or capitol. 1 Although the name Stakhra 
has not been traced back beyond Sasanian times, it must be an 
old Iranian word meaning 4 strong,' as applicable to the strategic 
character of the place in ancient times, and we have traditional 
authority for even assigning the founding of the city to the 
early dawn of the legendary Pishdadians, since Yakut says, 4 Its 
commencement is attributed to Istakhr, son of Tahumars 
(Tahmuraf).' 2 Mustaufi (A.D. 1340) reports a twofold legend, 
stating that, 4 according to some, Istakhr was built by Keiomars ; 
but according to others, it was founded by his son Istakhr, 
enlarged by Hoshang, and completed by Jamshid.' 3 The still 
earlier writer Ibn Haukal, in the tenth century, acknowledges 
that 4 Istakhr is a city neither small nor great, but it is more 
ancient than any city whatsoever in Farsistan. ' 4 Firdausi, in 
his epic, presupposes the existence of the city in the age of the 
legendary Kei Kaus, for, according to the poet, it had ' a palace 
that was the glory of the royal family.' 5 Tabari (d. A.D. 923), 
writing in the same century as Firdausi, but earlier, even claims 
that it was to a place in Istakhr, called Dizh-i Niplsht, c Strong- 
hold of Records,' that Zoroaster's patron, Vishtasp, sent the 
original copy of the Avesta, which was engrossed in letters of 
gold. This tradition, also found elsewhere, seems to agree with 
the Pahlavi account of the archetype copy of the scriptures 
which was deposited in the 4 treasury of Shaplgdn ' 6 and burned 

1 See p. 294, above. Kulub, cited by Barbier de Meynard, 

2 Yakut, p. 49. Tahumars, or Tab- Diet, g'eog. p. 48, n. 1 ; and compare 
muraf, is the same as the Avestan king Le Strange, Persia under the Mongols, 
Takhma Urupi, the predecessor of in JRAS. 1902, p. 519. 

Yima Khshaeta (Jamshid), and, ac- 4 Ibn Haukal, tr. Ouseley, p. 100. 

cording to legend, his brother. See 5 Firdausi, Shah Ndmah, tr. Mohl, 

on this point Darmesteter, Le ZA. 2. 2. 428. 
683, n. 18. The spelling and reading of this 

3 Hamdallah Mustaufi, Nuzhat al~ name is various. 




RUINS OF STAKHRA 




TAKHT-I TAUS, OR RUSTAM'S THRONE 



RUINS OF ANCIENT STAKHRA 307 

by Alexander the Great ; but we are not sure whether this 
4 Stronghold of Records ' was in the city on the plain, or located 
on the platform itself, as is more probable, having been trans- 
ferred thither under the later Achsemenians. 1 In Sasanian 
times the city seems to have been well known as Stakhr ; 2 but 
it appears to have lost its prestige with the lapse of time, and 
the place was in ruins when Pietro della Valle visited it in 
1621. 3 Broken columns, bases of pillars, and the remains of 
an ancient gateway alone now mark its site. To one who is 
acquainted with the clay-built dwellings alike of the rich and 
poor in Persia to-day, it is easy to understand how such a city 
could crumble into dust, with the exception of the few stone 
columns that mark its site, particularly as Yakut (A.D. 1220) 
expressly says, 4 the houses of Istakhr are built of clay or of 
stone covered over with plaster.' 4 

The southernmost point of this wide-extended but ill-defined 
settlement of the past appears still to be marked by a well- 
preserved small granite staging, which the peasants call Takht-i 
Td'iis, i Peacock Throne,' or Takht-i Rustam, 4 Rustam's Throne,' 
near the little post-house at Puzah. This raised flooring of 
stone is between seven and eight feet high and about forty 

1 For the statement of Tabari (Lei- talische Studien, pp. 1031-1033, Strass- 

den edition, p. 675) see the translation burg, 1906. 

by Gottheil in my Zoroaster, pp. 97, 2 See the Pahlavi works Shatroiha-i 
224, n. 2. The statement is repeated Airan, 41, tr. Modi, p. 97 (and notes) ; 
by Bundari and by Thaalibi, tr. Zoten- Karnamak, 4. 11, ed. Darab D. P. San- 
berg, Histoire des Hois des Perses, jana, p. 21. 

p. 257. Tabari also says there was ' a 3 See Pietro della Valle, Viaggi, 2. 

fire-temple of Anahedh (Anahita) at 248; Travels, ed. Pinkerton, 9. 101, 

Istakhr' (tr. Noldeke, Geschichte der and cf. Curzon, Persia, 2. 134, 136. 
Perser und Araber, p. 4). For the * Yakut, p. 49. On my journey 

tradition about the archetype copy of southward to Persepolis, for example, 

the Avesta, see Denkart 3. 3 ; 7.7.3 n.; I noticed an instance where an entire 

5. 3. 4 ; and compare West, SBE. 37. village had been abandoned in com- 

p. xxxi ; SBE. 47. 82, 127, and my paratively recent times and fallen into 

Zoroaster, p. 224, n. 1, consulting like- a mass of dust and rubbish, but was 

wise my article, Some Additional Data supplanted by another settlement on a 

on Zoroaster, in the volume in honor new site half a mile beyond, 
of Professor Theodor Noldeke, Orien- 



308 PERSEPOLIS AND ITS MONUMENTS 

feet square, and is composed of two layers of white blocks, 
some of them ten feet long, laid in terrace fashion. Not a 
trace of a column or base of a pillar is seen on the floor of 
the structure, as would have been the case if it were the sty- 
lobate of a small audience-hall (for we should hardly expect 
to look for a true temple among Achsemenian remains) ; we 
may presume, therefore, that it was really a throne-platform, 
as tradition says, like the one mentioned above in the chapter 
on Tak-i Bostan. 1 As such it was probably not intended for 
levees like those on the Grand Platform, but may have been 
designed as a reviewing-stand for great military gatherings, 
such as the assembly on the muster-field of Castolou Pedion, 
under the younger Cyrus, alluded to by Xenophon. 2 

The level surface of this granite staging, or platform, 
near Puzah is broken at the northwest corner by a large 
block, which might possibly have been a rostrum. About 
two hundred yards distant from this there rises from the ground 
a solitary block which looks as if it might have been used as an 
altar or a pulpit, for in front of it there is a stepping-stone, 
standing on which I could rest my note-book conveniently upon 
the top of the block a statement which will convey some idea 
of the height. In the light of the sinking sun the adjacent 
white terrace looked like the purest marble and was outlined 
in sharp detail against the black tents of the nomad Iliats who 
were encamped near it. 

On the way to the post-house I had time to make a brief 
inspection of the Sasanian sculptures close by the rocks of 
Naksh-i Rajab, as a preliminary to a re-examination of them 
next day. These bas-reliefs, three in number, are carved in a 
recess in the rocks, and their position is so little obtrusive that 
they would easily escape notice unless one were looking 
for them. They belong to the earlier period of the Sasanian 
dynasty, and two of them represent its founder, Ardashir 
Papakan (A.D. 226-241), in the act of receiving the crown at 

1 See p. 228, above. 2 Xenophon, Anabasis, 1. 1. 2 ; 1. 9. 7. 




SASANIAN SCULPTURE IN THE REAR OF THE RECESS AT NAKSH-I RAJAB 




SASANIAN SCULPTURE AT THE LEFT OF THE RECESS AT NAKSH-I RAJAB 



THE S AS AN I AN SCULPTURES AT NAKSII-I RAJAB 309 

the hands of the god Ormazd. In the first of these, the one 
to the right, or on the western side, Ardashir and the god are 
mounted on horseback, in a manner similar to that in the seventh 
bas-relief at Naksh-i Rustam, as shown in the photograph pre- 
viously given. 1 In the second, that on the rear wall of the niche, 
both personages are on foot, and there are several other figures 
in this group. Between the king and the deity stand two small 
boys, whose statues were originally less sharply carved and are 
now nearly obliterated. They are supposed to represent sons 
of the king. Behind the monarch there stands a serious-faced 
unbearded personage, who points to an inscription back of 
the king. He is intended possibly to represent a eunuch. 
Behind him again stands a heavily bearded person who is either 
a bodyguard or a vizir. On the other hand, to the right 
and back of the god, there are two smooth-faced figures that 
look like a queen and her maid retiring from the scene. 2 We 
do not know precisely the details of the royal incident here 
depicted, but the names Shahpur and Varahran, given in a 
Pahlavi inscription to which the unbearded personage points 
his finger, assigns it to Sapor I (A.D. 241-272) or his son 
Bahram II (A.D. 2T6-293). 3 The third panel, the bas-relief 
on the left, or north, side of the recess, represents Sapor I on 
horseback and attended by a bodyguard on foot. An inscrip- 
tion in Pahlavi and Greek serves to identify the king. 4 

The night was passed in the tiny chdpdr-khdnah of Puzah, 
the smallest in all Persia, I believe, and by daylight next morn- 
ing I was ready to mount for the short ride to the great 

1 See p. 302, above, and the pic- 4 1 have reproduced from my col- 
ture of the seventh bas-relief. For a lection photographs of the second and 
photograph of the Naksh-i Rajab third of these bas-reliefs ; for other 
equestrian sculpture, see Stolze and photographs of these, including the 
Andreas, Persepolis, 2. pi. 100. first, see Stolze and Andreas, Persepo- 

2 They did not look to me like Us, 2. pis. 100-104 ; Dieulafoy, U Art 
eunuchs, and my guides spoke of Antique, 5. pi. 17. The light in the 
them as sculptures of women. recess is unfortunately not favorable 

3 For the inscription and references, for making successful photographs of 
see West, Grundr. iran. Philol. 2. 77-78. the sculptures. 



310 PEESEPOL1S AND ITS MONUMENTS 

platform of Persepolis, Takht-i Jamshid, ' Jamshid's Throne,' as 
the Persians call it, or Chahal Mindr, ' Forty Columns,' the 
name by which it was more generally known in books of 
travel three or four centuries ago. This magnificent terrace 
is the foundation upon which stood the palaces of Darius, 
Xerxes, Artaxerxes, and their successors ; here Alexander 
held revel in the deserted halls of his adversary Darius 
Codomannus, the last of the Achsemenians, and in his drunken 
orgy, as is believed, burned the lordly edifices and the royal 
library that housed the scriptures and ancient records of Iran. 1 
To-day majestic ruins alone crown the height, and the natives 
know nothing about the historic associations connected with 
the platform and think of it only as one of the scenes of Jam- 
shid's departed glory. 

The platform itself lies at the base of a rocky row of hills 
called Kiih-i Rahmat, 4 Mountain of Mercy,' and in former times 
apparently also Shdh-Kuh, ' Royal Mountain,' the /3acn\iicbv 0/005 
of Diodorus, 2 which rises to the east and whose spurs have 
been partly cut away in the building of the terraced elevation. 
A noble wall, varying from twenty to fifty feet in height and 
constructed of stone quarried from the hillside, encloses the 
area on the three exposed sides, for no barrier was needed at 
the rear because of the natural fortification of the hills, as will 
be clear to any one who has visited the scene or consulted the 
panoramic photograph by Dieulafoy and the drawings of Flan- 
din. 3 The configuration of the terrace-height is such that three 
distinct levels are clearly noticeable, the highest being in the 
middle, which is made still more elevated by a mound in its 
midst. Over the surface of the platform are spread the 
remains of the architectural glories of the Achsemenians. 

We have a general description of the platform, written in 

1 See p. 307, n. 1, above. Diodorus 2 See Diodorus Siculus, History, 

Siculus (History, 17. 72) draws a vivid 17. 71. 

picture of the orgy and of the burning 3 See Dieulafoy, U "Art Antique de 

of Persepolis, a scene familiar through la Perse, 2. pis. 4-7 ; Flandin and 

Dryden's 'Alexander's Feast.' Coste, Voyage en Perse, 2. pi. 07. 



THE PLATFORM OF PERSEPOLIS 311 

Greek nearly two thousand years ago, by Diodorus Siculus 
(c. B.C. 50). Since the passage is important as a means of 
identification of the site, I translate the paragraph which relates 
to the construction of the terrace and the tombs of the kings, 
preserving in my rendering the interchange of tenses, present 
and past, that is found in the original Greek. 



' The citadel (aKpa) is worthy of mention. It had a threefold 
wall surrounding it, the first (section) of which was constructed 
with stately bastions (dvaX^/x/xan TroAvSaTravw) and adorned with 
battlements (e7roAe<n) and it had a height of sixteen cubits. 1 The 
second has a similar arrangement to that of the preceding, but double 
its height. The third enclosure is rectangular in shape, and its wall 
is sixty cubits high and constructed of solid stone so perfectly set 
as to last forever. On each side it has brazen gates and, beside them, 
brazen bulls, twenty cubits high, 2 the latter being intended to 
inspire awe in the beholder, and the former designed for security. 
On the side of the citadel toward the east, and four hundred feet 
distant, is the so-called Royal Mountain, in which were the tombs 
of the kings. The rock was hewn out and had in its bosom several 
sepulchres in which were the vaults of the dead. 3 There were no 
specially prepared means of access, but the corpses were hoisted up 
by machines (opyavwv) devised for the purpose, and thus received 
burial. 4 In the citadel itself there were many sumptuously equipped 
residences for the king and his officers, and likewise treasuries 
well adapted for the safe-guarding of wealth. 75 

The original plan and the main construction of this noble 
platform (which more than a thousand years ago was com- 

1 1 believe that the ' threefold wall ' 2 Reading ravpote for <rTavpot>s, ac- 

refers to the three distinct levels shown cording to Mr. Cecil Smith's excellent 

in the terrace rampart, although Blun- emendation, cf. Curzon, Persia, 2. 

dell (Persepolis, in Transactions of the 187, n. 1, and Blundell, op. cit. p. 553. 

Ninth International Congress of Ori- It is possible that the now ruined 

entalists, 2. 553) interprets this as re- bull-flanked portals may have been 

ferring to actual walls of circumvalla- actually gilded in ancient times. 
tion. Instead of 'bastions,' which I 3 Literally 'has in its midst several 

understand as alluding to the various houses (ofoous) . ' 

bays in the rampart, we may perhaps 4 Literally ' the vaults receive burial 

refer the words a vaX^nan iroXvSairdvy of the corpses that have been hauled up.' 
to the sumptuous edifices on the terrace. 6 Diodorus Siculus, History, 17. 71. 



312 PERSEPOLIS AND ITS MONUMENTS 

pared with Baalbec and the architectural remains of Palmyra 
and Egypt, 1 and fabled to be the work of Solomon's genii) 
was due to Darius. In one of his inscriptions he definitely 
states that he ' built this fortress on a place where no fortress 
had been built before,' and that he did so by the grace of 
4 Auramazda and the other gods.' 2 Darius erected at least two 
of the noblest buildings, but the elaboration of the design was 
due to Xerxes and its completion to his successors. Though 
far grander in its magnificence than any ordinary fortress, it 
must have been easily guarded by armed patrols on the walls 
and by platoons of soldiers stationed at all points of access, and 
reasons have been advanced for believing that its strength was 
re-enforced by walled fortifications or turrets in front of it on 
the plain. 3 The southerly position of the Palace of Darius and 
the fact that it faces southward has led, not unreasonably per- 
haps, to the assumption that there was originally an approach 
from the south or southeast, whereas the regular means of 
access, which must have been unchanged since the time of 
Xerxes, is by a great double staircase constructed in the wall 
near the northwest angle of the platform. 

This Grand Staircase (A) consists of a double ramping flight, 
each series numbering more than a hundred steps, with an angle 
of ascent so gentle and a width so broad, that a troop of horse- 

1 See the Mohammedan writer Ista- ress ' (Elam. halvarras, the same word 

khri (c. A.D. 950), ed. De Goeje, Bibl. that is employed to render OP. dula in 

Geog. Arab. 1. 150 and 1. 123, cf. Bh. 2. 39) is found only in the Elamitic 

Schwarz, Iran im Mittelalter nach version of the inscription on the side of 

den Arabischen Geoyraphen, 1. 13- the rampart, mentioned below, p. 318. 

14, Leipzig, 1896 ; and Mokadassi, or 8 See reference already given to 

Makdasi (A.D. 984), ed. De Goeje, Blundell, Persepolis, in Ninth Inter- 

Bibl. Geog. Arab. 3. 420, 435, 446, cf. nat. Congress of Orientalists, 2. 547- 

Noldeke, Encyclopaedia Britannica, 656. As explained above, I am inclined 

9th ed., 18. 558, notes 1 and 10. A to explain the threefold wall of Diodo- 

still earlier description of Istakhr is rus (17. 71) as referring rather to the 

given by Masudi (A.D. 944), Les Prai- three main elevations, and to under- 

ries cf Or, ed. Barbier de Meynard, 4. stand that the bull-flanked portals may 

76 seq. actually have been gilded as implied in 

3 This statement regarding the ' fort- the ' brazen gates ' and ' brazen bulls. ' 



THE GRAND STAIRCASE AND THE PORTAL OF XERXES 313 

men, ten abreast, could ride up it. 1 As we surmount the top- 
most step and cast the eye over the surface of the platform, we 
are struck by a succession of stately portals, broken columns, 
capitals, pedestals, stone steps, sculptured friezes, and doorways, 
spread about in confusion or gathered into disordered groups. 
So often have these ruins been described, and so fully have 
they been illustrated that I can do little here except point out 
the salient features and possibly add a suggestion or two 
regarding the historic significance of these relics of the past. 2 

Directly opposite the Grand Staircase is the Porch of 
Xerxes (-#) This imposing propylseum is guarded at each en- 
trance, back and front, by colossal winged bulls of stone, after 
the Assyrian manner. Two of these colossi face westward out 
over the plain ; the other two (a photograph of which I have 
given in the fourth chapter) look eastward toward the hills 
behind the platform. Near the top of each of the massive 
pylons of this portico there is a trilingual inscription in cunei- 
form characters, stating that the portal is the work of Xerxes 
and ascribing praise to Auramazda for all the blessings of his 
divine favor. 3 Two of the original four fluted columns are 
still standing between the stately piers of this triumphal arch, 
the ' Portal of All Nations,' as Xerxes himself called it, 4 through 
which at No-Ruz the envoys from tributary lands marched in 
solemn procession to bring gifts to the Great King, as por- 
trayed on the sculptured stylobate some fifty yards to the south. 



1 Attention was called to this fact, and Chipiez, and Dieulafoy, all of 
centuries ago, in the Zinat al-Majlis, which have been referred to many 
pt. 9, cited by Barbier de Meynard, times before. 

Diet. geog. p. 48, n. 2 ; also by Justi, 8 See Weissbach and Bang, Die 

Empire of the Persians, p. 189, and altpers. Keilinschr. p. 40, and Spiegel, 

by others. Die altpers. Keilinschr. p. 58. 

2 For the best presentation of all 4 In the old Persian language, 
that has been written on this subject, duvarthi visa-dahyu, see Xerx. Pers. 
see Curzon, Persia, 2. 148-196 ; for a. 11. For a conjectural restoration 
illustrative material, consult the stand- of the Portal, see Perrot and Chipiez, 
ard works of Texier, Flandin and Histoire de VArt, 5. pi. 3, p. 404. 
Coste, Stolze and Andreas, Perrot 



314 PERSEPOLIS AND ITS MONUMENTS 

This latter terrace, with its elaborately carved frieze and 
cuneiform inscriptions on the walls of the four staircases that 
approach it, served as a stylobate for Xerxes' lofty Audience- 
Hall ((7), the ruined columns of which gave rise to the native 
designation Chahal Mindr, 'Forty Pillars.' But the original 
number of columns was seventy-two, and of these only thir- 
teen are standing, to mark with their tall fluted shafts the 
aisles that led to the spot where Xerxes held levees within 
its once tapestry-hung walls. 1 The ruin and desolation form a 
pathetic contrast to the proud vaunt of the king in the cunei- 
form tablet carved on the stairway of approach, ' I am Xerxes, 
the Great King, the King of Kings, King of the Nations with 
their many peoples, King of this Great Earth even to afar,' and 
a sadder comment on the pious fervor of the words that follow, 
* Thus saith Xerxes, the Great King : Everything that has 
been made by me here and all that has been made for me else- 
where, I have made by the grace of Auramazda ; may Aura- 
mazda with the other divinities protect both my kingdom and 
all that I have made.' 2 

Walking about fifty yards to the south we come to the ruins 
of the Palace of Darius (Z>) situated on the highest part of the 
platform, and directly before a mound (Z). Although smaller 
and less imposing than either of the main edifices raised by his 
son Xerxes, the Palace of Darius is better preserved than the 
others. Here, several times repeated, are inscriptions record- 
ing the fact that the building was the ' palace \tachara), ' house ' 
(vitTi), or the 4 abode ' (hadisK) 3 of King Darius, whose figure 

1 For the problem of the walls ac- bach and Bang, Die altpers. Krilin- 
cording to Fergusson's architectural schr. p. 40 ; Spiegel, Die altpers. 
ideas, see Blundell, Persepolis, in Keilinschr. p. 62). 

Ninth International Congress of Ori- 8 The word hadis is added by 

entalistSj 2. 642-547, as opposed to Xerxes in two inscriptions on his 

Curzon, Persia, 2. 164-165, and the father's palace, once on a shaft in the 

reconstructions by Dieulafoy, IS Art southwest corner of the building, and 

Antique, 3. pi. 0, and Perrot and Chi- once on the southern wall of the stylo- 

piez, Histoire de VArt, 5. pis. 4, 5, 6. bate on which the palace stands (see 

2 See Xerx. Pers. b. 12-30 (Weiss- Dar. Pers. a, c; Xerx. Pers. ca[cb]; 



PERSEPOLIS 




SUBJECT NATIONS BRINGING TRIBUTE TO XERXES 



THE PALACES OF DARIUS AND AETAXERXES 315 

is sculptured in bas-relief, as fighting with some monster, whom 
he slays, thus triumphing over the power of evil, or as attended 
by servants who bear the royal umbrella, the fly-flap, and other 
insignia of sovereignty. To me the most interesting of all the 
inscriptions was a short device carved around the stone lintels 
of the windows through which the king looked out upon his 
people and over the fine panorama that stretched before his 
view. The cuneiform letters are deeply chiselled and they 
form a narrow band of text, originally repeated eighteen times 
because of the number of windows, of which only thirteen 
now remain. The brief sentence reads, ardastdna dthangaina 
Ddrayavaush khshdyatJiiyahyd vithiyd karta, 4 a structure of 
stone built in the house of King Darius.' 1 

Proceeding again southward, across a space that was once 
an open court below the palace, we enter the ruins of the 
Palace of Artaxerxes III, Ochus, (U) which faced directly 
north toward the latter. 2 An inscription, which is thrice re- 
peated on the double stairway that forms the approach on the 
north, and is reproduced again on the west, bears the name of 
Artaxerxes III, or Ochus. In this the king gives his genealogy 
and, after declaring that he has built this stone structure, 
closes with the words, 'may Auramazda and the god Mithra 
protect me and my country and all that has been made by me.' 3 
These inscribed tablets are separated by handsome panels that 
are decorated with bas-reliefs of the royal guards ; but excepting 
these stairway-friezes and the bases of a group of columns, there 
is little to show that the ruins among which we are standing 

Weissbach, pp. 6, 8, 32, 43 ; Spiegel, compare Perrot and Chipiez, Histoire 

pp. 50, 62 ; and compare Justi, Qrundr. de VArt, 5. pi. 9 (p. 644). 

iran. Philol. 2. 451-452). a Artax. Pers. a[b] (Weissbach, pp. 

1 Dar. Pers. c (Weissbach, pp. 5, 9,46; Spiegel, pp. 68, 69). The desig- 
34 ; Spiegel, p. 50 ; cf . also Justi, nation for the stairway, or possibly the 
Grundr. iran. Philol. 2. 451, and stylobate, is ustasana aQa n -ga[i]na, 
Bartholomae, Air. Wb. p. 193). lit. 'up-building of stone,' cf. also 

2 For the court see Blundell, Per- Justi, Grundr. iran. Philol. 2. 452, 
sepolis, in Ninth International Con- and Bartholomae, Air. Wb. pp. 64, 
gress of Orientalists, 2. 641-542, and 407. 



316 PERSEPOLIS AND ITS MONUMENTS 

are the remains of a palace. The small size of the ground-plan 
and the unfinished appearance of the surroundings have led 
some scholars to question whether the building was actually 
intended as a royal residence at all and whether it was ever 
completed. 1 

Directly to the east, on a large rectangular stylobate formed 
partly of the natural rock,' 2 stand the ruins of the sumptuous 
Palace of Xerxes (^), the grandest of the buildings in design 
except his own Audience-Hall and the Hall of a Hundred Col- 
umns belonging to his father. The stairways that lead up to it 
are richly decorated with sculptured panels, slabs, ornamented 
friezes, inscribed tablets, and pillars. 3 Fragments of columns, 
doorways, and windows remain to mark the courts of the king, 
who still walks in effigy of stone; 4 but like the grandiloquent 
titles which he hung upon the now crumbled walls they are 
merely mute witnesses of a dead past. 

Crossing some fifty yards eastward over uneven ground 
behind the palace, we find the remains of a smaller building, 
the so-called Southeast Edifice (6r), the identity of which is 
not positively known. Apparently it was a royal abode of 
some sort if we may judge from the images of the king carved 
on the doorways and representing him in combat with conven- 
tionalized monsters or as attended by slaves who carry the royal 
umbrella and fly-flap. 6 We may even go further and presume 
that it was the abode of Xerxes as crown-prince, if we may 
judge from the physiognomy of the king as portrayed here and 
on the walls of his palace. 6 

About forty yards north-northwest from this point and 

1 On this point compare the re- 4 For illustrations of Xerxes at- 
marks of Curzon, Persia, 2. 172-173 ; tended by his servants or in combat 
but see Justi, Grundr. iran. Philol. with mythical animals, see Stolze and 
2. 452, and Empire of the Persians, Andreas, 1. pi. 13 seq. 

p. 197. 6 For photographic illustrations, see 

2 Blundell, op. cit. p. 539. Stolze and Andreas, 1. pis. 1-4. 

8 See for example the photographs 6 Justi (Empire of the Persians, p. 

in Stolze and Andreas, Persepolis, 1. 198) says 4 the portrait of Xerxes is 
pis. 24-26. fairly recognizable j it shows a long, 



THE HALL OF A HUNDRED COLUMNS 317 

directly behind the mound in the rear of the Palace of Darius, 
we see a ruined entrance-hall, decorated with the conventional 
bas-reliefs of the king seated upon his throne. This small 
structure is known as the Portico of Darius (#"), though some- 
times called the Central Edifice. 

Adjoining it on the east there stands the last and the 
largest of all the palatial buildings, the Hall of a Hundred 
Columns (J), erected by Darius for holding ceremonial func- 
tions. The main entrance was on the north side through a 
vestibule whose roof was supported by sixteen columns that 
led the way into the throne-hall itself. This superb edifice 
covered an area two hundred and twenty-five feet square, and 
formed a magnificent structure raised on a hundred columns, 
ten rows each way, but of these not a single one is now stand- 
ing. The doorways east and west still represent Darius in the 
act of slaying animals of monstrous shape, and the entrances 
north and south depict him crowned with a tiara and mounted 
upon his throne supported by three or even five tiers of subject 
nations who carry arms in defence of their ruler, over whom 
hover the wings of his god. 1 The chamber walls of the royal 
hall were probably of sun-dried brick plastered over with a 
glaze or coated with enamelled tiles ; but they crumbled into 
dust ages ago, and only fragments of columns, stone door- 
jambs, and window-sills remain, together with bits of carbonized 
cedar buried beneath a mass of debris and ashes, to tell that 
the pillars once supported a roof with a thousand beams. 2 In 
this case, as in the case of the Audience-Hall of Xerxes, we 
are led to wonder whether it was the hand of the drunken 

bearded face with a prominent hooked Stolze and Andreas, 1. pi. 51 ; cf. also 

nose.' See also Justi's plan, op. cit. Curzon, Persia, 2. pp. 176, 178. 
p. 187, and compare his remarks in 2 The Avesta, by its architectural al- 

Orundr. iran. Philol. 2. 452. lusions ( Vd.18. 28 ; Yt. 5. 101 ; Ys. 57. 

1 1 reproduce a photograph of the 21) seems to refer to magnificent 

North Doorway of the Hall of the structures such as this. With regard 

Hundred Columns ; for a picture of to the charcoal and decomposed gach, 

the less imposing South Doorway, see or plaster, see Blundell, op. cit. p. 540. 



318 PERSEPOLIS AND ITS MONUMENTS 

Alexander and the torch of his revelling soldiers that brought 
about the desolation which reigns supreme. 

Some sixty or seventy yards north of this famous hall are 
seen a few blocks and mutilated columns of what was once a 
bull-flanked propylyeum, or Porch, (7) that led to the audience- 
hall itself; but all the rest of its pristine glory is lost for- 
ever. In addition to this portal there is near the Porch of Xerxes 
a rock-hewn Cistern (K) which must have fed a fountain 
whose jets sprang from the midst of a tank like the hoz in a 
modern Persian courtyard. Besides the tumulus, or Mound 
(), previously referred to, there are beneath the surface of the 
platform also several underground passages, water-channels, and 
drains, which have not yet been fully excavated, together with 
some minor evidences of unfinished work in the past, but they 
still await the spade of the archseologist. 1 

Before closing the chapter I must add two more paragraphs 
on matters of historic interest. The first relates to two impor- 
tant inscriptions carved by Darius on mammoth blocks set in 
the southern retaining wall of the platform and known to 
scholars as Dar. Pers. d and e. Each tablet contains twenty- 
four lines. In the former inscription the king glorifies Aura- 
mazda, gives thanks to him for his blessings, and prays that he 
and the other divinities may ever protect the land. The same 
idea is repeated in the Babylonian version, which is largely a 
paraphrase, and partly also in the Elamitic section, which adds, 
however, some interesting information, not found elsewhere, to 
the effect that Darius was the first to fortify the place, which 
was not previously a stronghold. 2 The adjoining inscription 
on a block to the right is written only in old Persian and 

1 For some results from compara- see Weissbach and Bang, pp. 5, 34, 
lively recent diggings in the mound and Spiegel, pp. 46-50; and compare 
and some excavations among the ruins, Weissbach, Die Achdmenideninschrif- 
see Blundell, op. cit. pp. 537-559. ten Zweiter Art, p. 76; Bezold, Die 

2 For a photograph of this inscrip- Achaemen. Inschr. p. 39. Consult like- 
tion, see Stolze and Andreas, Per- wise, Justi, Chrundr. iran. Philol. 2. 
sepolis, 2. pi. 95 ; and for translations, 448. 



TOMBS OF THE LATER ACHsEMENIANS 319 

enumerates the conquests of Darius, concluding with the prayer 
that ' Peace may come from Aura.' 1 As I stood at the foot of 
the rampart to collate the tablets high above and make notes 
which I hope later to publish, the sun's rays were so scorching 
and the flies so pestiferous that I could understand the king's 
need for a chowri and an umbrella in ancient days ! 

The remaining point of historic interest connected with 
Persepolis is the series of three tombs hewn in the rocky hill 
of Kuh-i Rahmat behind the platform. They are the sepulchres 
of three later kings of the Achsemenian line arid resemble the 
four elder tombs at Naksh-i Rustam, which have been already 
described; but owing to the nature of the hillside where they 
are cut they differ from the latter in minor details, especially 
in being less high from the ground and therefore easy of access. 
The first of the three is hewn in the face of the rock almost 
directly back of the Hall of a Hundred Columns and is com- 
monly known as the North Tomb and presumed to be the 
mausoleum of Artaxerxes II, Mnemon (B.C. 404-358). The 
second lies in a recess in the mountain-side somewhat south- 
east of the lower end of the platform and is designated as the 
Middle Tomb, 2 and is believed to be the vault of Artaxerxes III, 
Ochus (B.C. 358-337). The third is cut in a rock more than 
half a mile farther to the south and is easily accessible from the 
ground, but was never finished. It may have been commenced 
by Darius III, Codomannus (B.C. 335-330), the last of the Achse- 
menian kings. If that be true, it is reasonable to suppose that 
his overthrow by Alexander and his subsequent tragic death 
were the cause of its never having been completed. 3 

1 Dar. Pers. e. 1-24. See also pre- as I have given it, is the generally ac- 
ceding references. cepted one. See for example Curzon, 

2 The position of the tombs may Persia, 2. 183 ; Justi, Grundr. iran. 
best be gathered from Flandin and Philol. 2. 455. For architectural 
Coste, Voyage en Perse, Ancienne, 2, details consult Perrot and Chipiez, 
pi. 65 (reproduced in Perrot and Histoire de VArt, 5. 617-638 ; and for 
Chipiez, Histoire de VArt, 5. 454). photographs refer to Stolze and 

8 The assignment of the three tombs, Andreas, Persepolis, 1. pis. 70-73. 



320 PERSEPOLIS AND ITS MONUMENTS 

As we gaze upon this tomb and the others, and then cast our 
eyes toward the ruins of Persepolis, we can but think with a 
heart-pang of Omar Khayyam's lines : 

' They say the Lion and the Lizard keep 
The Courts where Jamshyd gloried and drank deep.' 

Here stood the palace of Darius, there the throne-room of 
Artaxerxes, yonder the pillared halls of Xerxes, and not far 
distant the tombs of the kings. But all are in ruins ; all are 
relics of glory past. Yet who knows ? Out of the shadow of 
by-gone days, out of the dust of departed ages, out of the ashes 
of the Simurgh's fire, out of the fragments of shattered Iran, 
there may arise one whose master hand will restore the glory 
of the ancient Persian kingdom, illumine again the pages of 
Persia's chronicles, recall what was noblest in the Parthian rule 
and Sasanian empire, and make splendid once more the land 
and people of the Lion and the Sun. 




THE PALACE OF DARIUS 




A PORTAL AND THE HALL OF A HUNDRED COLUMNS 
(The North Tomb in the Background) 






HA 

OF THF 

UNIVERSITY 






CHAPTER XXI 
SHIRAZ, THE HOME OF THE PERSIAN POETS 

* There's a bower of roses by Bendemeer's stream 

And the nightingale sings round it all the day long ; 
In the time of my childhood 'twas like a sweet dream, 

To sit in the roses and hear the birds' song. 
The bower and its music I never forget, 

But oft when alone in the bloom of the year 
I think is the nightingale singing there yet ? 

Are the roses still bright by the calm Bendemeer ? ' 

MOORE, Lalla Eookh. 

SHIRAZ lies about forty miles south of Persepolis, but the 
two stages of the journey are not easy, so I arranged to make 
my start from the desolate and ruined halls of the Achas- 
menians in time to reach the native city of Hafiz and Saadi 
before nightfall. The first relay of horses for the journey 
I found good, which proved an omen for the second ; and a 
series of long and hard gallops, with only occasional halts to 
adjust the load on the pack-horse, brought my little cavalcade 
in two hours to the end of the hill-girt marshy plain of Merv- 
dasht. At no great distance from this point the road crosses 
a bridge over 'Bendemeer's stream.' This watercourse owes 
its name Band-i Amir, 4 Dam of the Amir,' to an arched cause- 
way constructed by Azad ad-Daulah, who governed Fars in 
the tenth century, and who also adorned the banks of the river 
at various places with parks and palaces. 1 The latter, unfortu- 
nately, have vanished generations ago. 

A few miles beyond the Bendemeer bridge the village of 
Zargan, or Zergun, lies nestled at the foot of a mountain, where 

1 See Yakut, pp. 313, 480. 
Y 321 



322 SHIRAZ, THE HOME OF THE PERSIAN POETS 

the chdpdr khdnah is located, and here a change of horses may 
be obtained. As the weather was warm the post-quarters at 
the time had been moved out on the main trail in the plain 
and were lodged in a couple of small tents, so I was saved the 
extra ride to the Jialting-place. An unexpected delay, however, 
occurred. At the moment when we were about to change 
mounts the horses stampeded and scampered away a mile or 
more before the muleteer could recapture them. Meanwhile 
we had time to rest and console ourselves for the loss of 
time with a good glass of tea. The half hour passed quickly 
in taking notes of the surroundings and in observing the differ- 
ent types among the natives, for the inhabitants of Farsistan 
impressed me as being the handsomest Persians I had seen. I 
was interested also in the primitive tankards of goatskin in which 
the water for the tea was brought. These rude vessels were 
made from the undressed hide of a goat, with the animal's hair 
left on the outside and the skin drawn tightly around a wooden 
rim and a circular board bottom so as to form a bucket, while 
three sticks were used as fastenings to give firmness to the 
whole and as props for the uncouth vessel to stand upon. I pre- 
sume it was from tankards such as these that the hardy soldiers 
of Cyrus used to drink, before luxury taught them the use of 
silver beakers and the accompanying vices which sapped away 
the vigor that had conquered kingdoms. 

Zergun remains clear in my memory because of an accident 
to the postilion, who was seriously kicked, on the return 
journey, by my pack-horse, a vicious stallion, as most of the 
Persian horses are. At first I thought that the man's leg was 
broken, but on examining the wound I found that the kneecap 
was not shattered, though I fear that the injury to the bone 
may have proved in some way a permanent one. 

After leaving Zergun the hard stage of the road began. 
Nature has thrown up a barrier on the north to protect the 
approach to her chosen city of Shiraz, or perhaps to set bounds 
to the too enthusiastic admiration that might be bestowed upon 




KING DARIUS ON HIS THRONE 




BRIDGE LEADING INTO SHIRAZ 



FROM Z ERG UN TO THE GATE OF SHIR A Z 323 

its beauty. This obstacle is in the form of a steep and inde- 
scribably stony mountain road winding up and down and 
hither and thither, past ruined forts and dilapidated habita- 
tions, before reaching the stream of Roknabad. This small river 
was so named after the Buyid ruler, Rokn ad-Daulah Hasan, in 
the tenth century, who conducted its water by special courses 
to Shiraz and the beautiful suburb Musalla. It owes its fame, 
however, to Hafiz, who sang its praises and compared it with 
the rivers of Paradise. But the Roknabad has now shrunk to 
so small a measure that Hafiz seems guilty of a strange hyper- 
bole in the verse, 

1 In Paradise thou wilt not find 

The beauteous banks of Roknabad 

And the rose-bowers of Musalla.' 

Nevertheless the landscape about the stream is Arcadian, and 
the scenery calls forth admiration from the rider as he winds 
his way downward through the mountain glade that lies beyond 
the stony section of the road. 

Suddenly through a great notch in the mountains, Shiraz in 
all its beauty bursts upon the view. It seemed, as I saw it, to 
rise like an island in a sea of emerald bordered in the distance 
by purple hills. The coloring was as rich as it was harmoni- 
ous, and it seemed no longer a wonder that the proud Shirazis 
have given to the city gateway built here at Nature's grander 
portal the name Tang-i Alldhu Akbar, because the beholder, on 
viewing such ,a scene, instinctively cries out ' God is Most 
Great ! ' l The panorama was a magnificent one ; plain and 
hills, gardens and cypress-groves, towers and walls, domes and 
spires, were bathed in a mellow light. The vision grew in 
beauty as I rode forward. The wayside was lined with myriad 
poppies; the gardens were abloom with the jasmine and the 
rose, for it was the beginning of May, and the rose had begun 
to blush a few days before in obedience to the nightingale's 

1 See also the descriptions of 359, and Browne, A Year Amongst 
Brugsch, Im Lande der Sonne, p. the Persians, p. 260. 



324 SHIRAZ, THE HOME OF THE PERSIAN POETS 

behest 4 her sallow cheek to incarnadine ' ; the trees were vocal 
with the note of the bulbul : the setting sun threw long shadows 
from the tall minarets and slender sarv trees ; and Shiraz for 
the instant was the realization of that Persian elysium of 
which the poets sang, and it awaited only the flood of moon- 
light which swept in after sunset to complete the enchantment. 

I confess that this is written in a Persian mood. In calmer 
moments I can pause to consider that the city is not a Paradise 
after all. Ahriman has marred its perfection by his blight, as 
he did in Airan Vej of old. The climate at times becomes 
extremely hot and exhausting, as the zone is tropical, and 
fevers are frequent and deadly, since hygienic laws are fatally 
neglected in the city. The very architecture of the buildings 
also leaves much to be desired ; and the Shirazis, though pleas- 
ure-loving and clever, have a traditional reputation for bigotry 
and conceit not out of keeping, perhaps, with the official title 
of the town, which is 'Abode of Knowledge' (Ddr al-'-Ilm). 1 
But these are the judgments of a colder moment and foreign 
to the proper frame of mind for a visit to Shiraz. 

Although the city is the capital of the historic province of 
Fars and by right of inheritance the successor to the glory of 
Persepolis, the claim which Shiraz can make to eminence by 
reason of antiquity is not comparable with that of either 
Hamadan or Rei in Media of old. The general location of the 
city, it is true, is probably an ancient one, as shown by the 
vestiges of Achsemenian and Sasanian ruins in the vicinity, and 
Iranian legend and Mohammedan fable are even ready to 
ascribe the founding of the city to a son of Tahumars or to 
a great-grandson of Noah, but the more sober Moslem authors 
say that Shiraz was 'founded or rebuilt by Mohammed ibn 
Yusuf Takali after the Rise of Islam ' in the seventh century 
of our era. 2 The Arab traveller Ibn Haukal mentions its 

1 On this title see Browne, Episode 2 For the statement regarding Tahu- 

of the Bab, 2. 294, n. 1, 364, n. 2, Cam- mars, see Yakut, p. 362, and for the 
bridge, 1891. more conservative view, see the re- 




THE NEW MOSQUE (MASJID-I No) AT SHIRAZ 




OVERLOOKING SHIRAZ 



THE HISTORY OF SHIRAZ 325 

citadel (Kohandiz) in the tenth century, and Yakut (c. A.D. 
1220) states that the Buyid ruler Abu Kalanjar Sultan ad- 
Daulah fortified Shiraz with strong walls in the eleventh cen- 
tury (A.H. 440). 1 These fortifications, however, were of no 
avail against the Mongol conqueror Tamerlane, when he 
sacked the town two hundred and fifty years later. Successive 
rulers restored and embellished the city, but their work was 
usually destroyed later by the forces of nature or through the 
capture of the town by enemies. Shiraz owes most of its 
architectural beauty to-day to Karim Khan (1751-1779), who 
governed it as regent under the Safavid dynasty in the latter 
half of the eighteenth century. Many of the effects of his 
refining influence were nullified by the eunuch ruler, Agha 
Mohammed Khan, who razed its stone ramparts to the ground, 
replaced them by mud walls, and reduced the city to a rank 
unworthy of its traditional prestige. 

Among the architectural monuments of Shiraz, the oldest is 
a mosque which dates from the latter part of the ninth century 
and was built by the Safarid dynast Amr ibn Leith. Belonging 
to a period two centuries later is the New Mosque (Masjid-i No). 
Se'id ibn Zangi (1195-1226) reconstructed this out of his own 
palace, which he is said to have converted to the service of 
God as the result of a pious vow made in behalf of the life 
of his son. 2 The flat-roofed cloister around its court is seen 
in the photograph which I reproduce, while conspicuous in the 
background is the faience-traced dome of Shah Chiragh, the 
beauty of which is unfortunately impaired by a popular com- 
parison of its swelling cupola with the head of some gigantic 
asparagus. Beneath its vaulted roof lie the remains of one 

marks of Mustaufi, cited by Barbier de Meynard, Diet. geog. p. 362, n., 

de Meynard, Diet. geog. de la Perse, and Browne, Literary History of 

p. 362 n. Persia, p. 352. Elbe" and Horn, 

1 Ibn Haukal, tr. Ouseley, p. 93, Grundr. iran. Philol. 2. 218, 560, 
and Yakut, p. 365. 661, give the dates of Amr ibn Leith 

2 See Curzon, Persia, 2. 102, and as A.D. 878-900. 
compare Mustaufi, cited by Barbier 



326 SHIRAZ, THE HOME OF THE PERSIAN POETS 

of the sons of Imam Musa, a champion of Islam. 1 Yet in 
architectural merit neither this nor any of the other religious 
edifices, madrasahs, mausoleums, or baths can rival those of 
several other cities in Persia. The grand bazaar, Bdzdr-i Vakil, 
4 Regent's Bazaar,' is a fine structure, due again to the munifi- 
cence of Karim Khan, and it carries on a fairly flourishing trade; 
but the caravansarais- are not particularly spacious, nor are the 
streets of the city beautiful ; the Ark, or Citadel, on the other 
hand, is rather imposing. The best-constructed of the modern 
buildings in the town is that occupied by the offices of the 
Indo-European Telegraph Company. It was formerly a palace 
and has a fine courtyard of stone, while its hallways and roomy 
chambers seemed to me Western in their style of architecture 
rather than Eastern. 2 This touch of the West brought me in 
another way also nearer home, for I found an opportunity to 
send a cablegram to America a welcome experience after 
having been cut off from direct communication with home 
since I left Urumiah. 

Buildings of brick, mortar, and stone are not the glory of 
Shiraz ; it owes its renown rather to the causes which I shall now 
enumerate. In the first place the natural beauty of its environs 
is greatly enhanced by cultivation and by art. The entire 
plain surrounding the city is well cultivated, and owing to its 
tropical situation (for Shiraz is nearer to the equator than is 
the northern part of India) it yields abundantly to tillage and 
irrigation. The vineyards around the city produce the best 
wine in Persia, a product for which Shiraz has ever been 
famous. There are two varieties of this wine, a red and a 
white ; the taste of the white wine reminded me somewhat of a 
Marsala. 

The gardens and rose-bowers of Shiraz are still more famous. 
Within the city and on its outskirts there are dozens of these 

1 On the latter point compare also this building, see Weeks, From 
Curzon, Persia, 2. 102. Black Sea, p. 116. 

2 For some of the artistic points of 



GARDENS AND ROSE-BOWERS 327 

pleasure-grounds, some of which still retain their beauty 
despite the neglect into which they have fallen. The Persian 
garden in general is somewhat different from its counterpart in 
other lands and is more like an orchard, a horticultural 
enclosure, than a garden in the narrower landscape sense ; in 
fact the ordinary Persian word for ' garden,' Idgh, may some- 
times best be rendered by our word 4 orchard,' with little of the 
connotation of 4 flower-garden.' Instead of being winding paths, 
the walks are usually laid out in straight lines, with brick and 
tile borders, while terraces also are constructed whenever possi- 
ble, as in our own gardens, and finished with stonework and 
masonry. A reservoir of water, even if its basin be only a 
small tank, necessarily graces the area, and luxury may add 
a fountain and cascades falling over stone slabs, but water is a 
precious article, and lavishness in this regard is equivalent to 
extravagance, even if nature responds liberally to the smallest 
drop of the precious liquid. Shade trees like the poplar 
(kalam), willow (bid), cypress (sarv), plane tree or sycamore 
(chinar), line the walks or mark off the grass-plots, while the 
shrubbery varies considerably according to the latitude. 1 

The main road into Shiraz from the Allahu Akbar Gate, is 
lined on either side with gardens, two of which on the east, 
the Ohahal Tan, 'Forty Bodies,' and the Haft Tan, 'Seven 
Bodies,' are rather large pleasure-groves and a resort for 
dervishes, and may be seen in the photographic reproduc- 
tion, with the Tomb of Hafiz in the background. 2 On the 
western side there are corresponding enclosures, and one of 
the most characteristic of these is the BdgJi-i Takht, 4 Garden 
of the Throne,' to which I paid a special visit. It stands on 
rising ground overlooking the city from the northwest, and 
was laid out by the victorious Kajar ruler Agha Mohammed 

1 For illustrations of Persian gar- 2 See also the comments of E. G. 

dens, see Mumf ord, Glimpses of Mod- Browne, A Year Amongst the Persians, 

ern Persia, in House, and Garden, 2. p. 278. 
175-191, 360-373, Philadelphia, 1902. 



328 SHIRAZ, THE HOME OF THE PERSIAN POETS 

Khan, who constructed it on the site of an older garden 
admirably adapted for the purpose by its location. Terrace 
rises above terrace, and fountain, channel, and stream pour 
their waters in cascades over slabs of marble into reservoirs 
faced with stone. The watercourses are edged with masonry, 
and the walks bordered with cypress and orange trees. At the 
time of my visit the large reservoir in the centre was full, 
though the cascades no longer flowed, and I understand that in 
the drought of summer everything dries up and dust prevails 
everywhere. The walls around the enclosure and leading up 
the terraces were not kept up and were consequently beginning 
to crumble, while the pavilion which once graced the upper 
terrace was deserted and in ruins. Yet there remained enough 
of by-gone luxury to tell how beautiful this little Luxembourg 
must have been formerly, and it still offers to the Shirazi an 
attractive place to visit in the cool of the evening. 1 

The true renown of Shiraz, as I have implied, rests not upon 
the beauties of nature, which I have been describing, but upon 
the fame of her poets and the distinguished men she has given 
to Iran. Not the least known among the latter class is one of 
recent memory, the Bab, whose religious reform in the past 
century I have mentioned in an earlier chapter. 2 The list of 
notable Shirazis was already a long one when Yakut wrote, 
seven hundred years ago (c. A.D. 1220), yet even he had not 
lived to be aware of the future greatness of his younger con- 
temporary, Saadi, or to know that one of the world's greatest 
lyrists, Hafiz, would be born in Shiraz. 

Hafiz, whose birth occurred some time in the first half of the 
fourteenth century, is known almost as well, by name at least, 
in the West as he is in the East, where every Persian is familiar 
with his odes, which have made Shiraz a synonym for poetic 
inspiration. The beauty of his language, the charm of his 

1 For descriptions of the gardens, the Persians, p. 279 ; Curzon, Persia, 
see also Weeks, From the Black Sea, 2. 104. 
p. 116 ; Browne, A Year Amongst 2 See pp. 48-50, above. 



THE PERSIAN POET HAFIZ 329 

style, the sweet flow of his verse, and the passionate expres- 
sion of his feeling, whether it be in the lyrical outpouring of his 
own love, or in the mystic ecstasy of a spiritual devotion veiled 
under the guise of material images, entitle Hafiz to rank even 
in the Occident as a poet's poet and to hold a prominent place 
in the best literature of the world. 1 His youth may have been 
Anacreontic, but he must have been a faithful student, as he 
won by his memory and learning the title Hafiz, ' mindful,' a 
distinction bestowed only upon those who knew the entire Koran 
and its interpretation by heart, and he received also an appoint- 
ment as instructor to the family of the ruling House of Muzaf- 
far, as well as a position in the royal madrasah, which was 
founded expressly for him. Even a prince of India, Mahmud 
Shah Bahmani of the Deccan, invited him to his court as a 
permanent guest. Hafiz accepted the invitation and started 
on the journey, but proved unequal to facing the perils of a 
journey by sea, and abandoned his plan, excusing himself 
by writing a handsome panegyric of his would-be patron, and 
delicately urging his preference for a life amid the enchant- 
ments of Shiraz. According to an interesting tradition, further- 
more, even the stern conqueror Tamerlane came under the charm 
of Hafiz's verse, but scholars generally discountenance the 
legend as they in the same manner reject the fable that he 
received poetic inspiration from a cup of nectar placed to his 
lips by an aged man as a reward for his devotion to the love 
of a beauteous maiden. 2 

Hafiz was prolific as a writer and the manuscripts and 
printed editions of his works contain more than five hundred 
ghazals, or odes, which have his name deftly woven into the 
last stanza of each. Regarding the poetic merit of his verses 
there is no question, but there exists, as there existed even in 
his lifetime, a diversity of opinion regarding the interpretation 

1 See my article Hafiz, 1 in War- 2 See Sir Gore Ouseley, Notices 

ner's Library of the World's Best Lit- of the Persian Poets, pp. 35-37, Lon- 
erature, 12. 6793-6806, New York, 1897. don, 1846. 



330 SHIRAZ, THE HOME OF THE PERSIAN POETS 

of his poetry whether it is to be taken in a literal or in 
a spiritual sense. Some readers see in his praises of love and 
wine, of musky tresses and cypress forms, the passion of an 
Ovid or a Tibullus ; but others, especially some of his Oriental 
admirers, read beneath those physical images the spiritual 
thoughts of Divine Love and the Soul. Wine is the spirit, 
not the juice of the grape, and the cup drained in the tavern 
is but a draught of that self-oblivion which brings one into 
complete union with the supreme essence. There is undeni- 
able truth in the possibility of so interpreting the verses 
according to the doctrines of Sufiism, just as there is in the 
mystic interpretation of the Song of Solomon, and parallels 
might even be cited from the poems of Donne, Vaughan, and 
Crashaw, in the English literature of the seventeenth century ; 
but it is equally true that some of the verses of Hafiz, per- 
haps those of his youth, can hardly allow of anything but a 
material and passionate interpretation. As an illustration of 
his lyrical style I choose an ode translated by the late Professor 
Cowell, who taught 4 Omar' FitzGerald his Persian. 

' The fairest of roses no longer is fair, 
If she who possesses my heart is not there; 
If wine, the bright ruby, be ever forgot, 
The spring hath no charms and delighteth us not ; 
The walks of the garden are lonely and drear, 
If the song of the nightingale strikes not my ear. 
The cypress may wave, and the roses may bloom, 
But in vain if the queen of my heart does not come ; 
The wine and the roses are charming, I own, 
But if she is absent, the charms are all gone. 
The most lovely designs which art can devise, 
Without my fair mistress delight not my eyes. 
O Hafiz, thy life is but useless at best, 
Scarce worth a nivar to be thrown to a guest.' l 

The peculiar structure of Hafiz's verse and his repetition of 
rhymes has been well imitated in some English renderings by 
Mr. Walter Leaf. I reproduce the latter's translation of one 
1 See Professor Cowell's Life and Letters, p. 24, London, 1904. 




CULTIVATED FIELDS AROUND SHIRAZ 




GARDENS AROUND SHIRAZ 



THE POETRY OF HAFIZ 331 

of the poet's best-known odes, a favorite especially because of 
its refrain tazah bah tazah, no bah no. 

'Minstrel, awake the sound of glee, joyous and eager, fresh and free ; 
Fill me a bumper bounteously, joyous and eager, fresh and free. 

O for a bower and one beside, delicate, dainty, there to hide ; 
Kisses at will to seize and be joyous and eager, fresh and free. 

Sweet is my dear, a thief of hearts ; bravery, beauty, saucy arts, 
Odours and unguents, all for me, joyous and eager, fresh and free. 

How shall the fruit of life be thine, if thou refuse the fruitful vine? 
Drink of the wine and pledge with me, joyous and eager, fresh and free. 

Call me my Saki silver-limbed, bring me my goblet silver-rimmed; 
Fain would I fill and drink to thee, joyous and eager, fresh and free. 

Wind of the West, if e'er thou roam, pass on the way my fairy's home, 
Whisper of Hafiz am'rously, joyous and eager, fresh and free.' 1 

The opponents of Hafiz maintained that his philosophy of 
life was too much akin to free-thinking and his scorn of the out- 
ward semblances of piety too undisguised, to say nothing of lax 
verses on wine-bibbing and odes perfumed with the tresses of 
his loves, so that when he died, in 1389, this feeling found open 
expression among the Mullahs, who refused to accord to his 
remains the last rites due to a true Mohammedan. A contro- 
versy arose, and to settle the question it was agreed to leave the 
solution to the poet's own writings. A number of his verses 
were taken at random and a child was selected to draw one of 
these out of an urn. Happily the stanza read : 

* Forbear thou not to shed a tear 
Compassionate on Hafiz' bier, 
For know that though now deeply 'mersed in sin 
To Paradise he yet shall enter in.' 2 

The omen was favorable ; Hafiz was granted a Mohammedan 
burial, and his tomb, the Hafiziah, has since become a shrine 
for pilgrims from far and near. 

1 Leaf, Versions from Hafiz, an Es- 1898; cf. also Payne, Hafiz, 1. 45. 
say in Persian Metre, p. 23, London, 2 Ode 60. 7 ; cf. Payne, Hafiz, 1. 76. 



332 SHIRAZ, THE HOME OF THE PERSIAN POETS 

The sepulchre lies about two miles northeast of the city, and 
a short gallop, after we leave the outskirts of the town, brings 
us to the walled enclosure that surrounds it. Passing through 
the gateway we find ourselves within a spacious square bordered 
on three sides by low buildings, which afford shelter for priests, 
dervishes, and pilgrims ; the area is shaded by poplars, cypresses, 
and maples, and beneath their shadows a small reservoir is 
seen. The tomb of Hafiz stands in the middle of the garden 
and is surrounded by a number of graves, since burial near the 
poet's dust is now a special privilege. The place is well kept 
up, being no longer neglected as when Ker Porter described it 
in the last century, 1 and the increasing number of pilgrims that 
yearly visit the shrine speaks for the growing fame of the poet, if 
not of orthodox Islam. A handsome oblong block of marble 
covers the grave and takes the place of the original slab, which 
Karim Khan is said to have placed in the Jahan Namah Garden 
when he replaced the stone by the present sarcophagus. 2 The 
block is beautifully carved with verses from the poet's writings 
and at the top is an Arabic inscription, the tenor of which is 
the transitory character of human things and the eternal nature 
of God ; while at the bottom is added the date of the poet's 
death, which is given as the year 1389 (A.H. 791). 3 The pres- 
ent governor of Shiraz has taken pains to have the sepulchre 
protected by a large iron grating which is more imposing than 
the old metal cage that formerly enclosed it, and the scroll- 
work and design show some artistic taste. The stanchions and 
corner-posts, however, are iron telegraph poles, received from 
the Indo-European Telegraph Company, and the Shirazis 
seemed to be almost as proud of these and of the little metal 
flags that decorate the top, as of the inscribed slab over the 
poet's dust. 

1 See Ker Porter, Travels, 1. 694- 2 See Curzon, Persia, 2. 109. 

695. Sir Gore Ouseley (Notices of 8 For the inscriptions see Browne, 

the Persian Poets, p. 40, London, 1846) A Year Amongst the Persians, pp. 280- 

spoke of the grave as being ' in excel- 281. 
lent order' when he saw it in 1811. 



THE TOMBS OF HAFIZ AND SAADI 333 

The tomb of Saadi lies about a mile farther northward in a 
slight hollow of the plain and is called the Saadiah. Like the 
tomb of Hafiz, it is in an enclosed garden, and a grove of pop- 
lars, cypresses, fragrant shrubs, and rose bushes surrounds the 
building which contains the remains of Persia's great moralist 
and poet. It is a fitting resting-place for one who gave the 
titles of ' Rose Garden ' ( G-ulistdn) and 4 Garden of Perfume ' 
(Bostdri) to his two chief works. Within this precinct Saadi 
alone is buried at least I saw no other graves and the 
sepulchre itself is now enclosed within a building. The cham- 
ber in which the sarcophagus stands is entered through a stout 
door, and the poet's remains lie in a heavy stone case surrounded 
by a metal network. The room itself is without decoration, 
but is richly carpeted with a Persian rug, on which the foot 
falls noiselessly as one moves about the sarcophagus to do hom- 
age to the memory of the dead. The same Arabic inscription 
about the immutability of God, as on the grave of Hafiz, is 
chiselled on the stone, and verses are added from Saadi's own 
poems, a handsome manuscript of which is preserved in the 
building. 

Saadi's life was a long and eventful one. Born in 1181 or 
1184, or nearly a century and a half before Hafiz, he is said to 
have rounded out a full fivescore and more of years, as his 
death is recorded as having taken place in A.D. 1291. Although 
we may be uncertain with regard to dates, we know that his 
life was one of many experiences, and that he was a man widely 
travelled. He had journeyed east, west, north, and south 
throughout his own country, and had made more than a dozen 
pilgrimages to the shrine at Mecca, besides travelling in India, 
Asia Minor, and Africa. On one occasion he was taken 
prisoner by the Crusaders in Tripolis, enslaved, and set to 
digging in the trenches. The story goes that a wealthy mer- 
chant of Aleppo took compassion on his wretched plight, ran- 
somed him for ten dinars, and later gave him his daughter in 
marriage with a dowry of a hundred dinars. The marriage 



334 SHIRAZ, THE HOME OF THE PERSIAN POETS 

did not prove a happy one, owing to the wife's bad temper, 
and once she reviled him with the reproach : 4 Art thou not 
the slave whom my father bought for ten dinars ? ' 4 Yes,' 
replied Saadi, 4 he ransomed me for ten dinars and sold me to 
you for a hundred.' And he added in verse : 

' I've heard that a man of high degree 
From a wolf's teeth and claws a lamb set free. 
That night its throat he severed with a knife, 
When thus complained the lamb's departing life : 
" Thou from the wolf didst save me then ; but now 
Too plainly I perceive the wolf art thou." ' * 

As Saadi in the Gulistan tells the story about himself, it can 
hardly be thought gossipy to repeat it. 

As a literary work Saadi's Gulistan is a storehouse of anec- 
dotes as well as wise maxims, good counsel, and poetic thought. 
In this didactic work of mingled prose and verse the author 
himself says that 4 the discourse is combined with pleasantry 
and cheerful wit, the pearls of grave counsel are strung on the 
thread of diction, and the bitter medicine of advice is blended 
with the honey of mirthful humor.' Some of the stories are truly 
amusing, and the substance of one or two is worth giving as an 
example of Oriental humor, for the Persians possess a sense of 
humor, with which they are not always credited. Almost 
modern in its point is the anecdote of the man whose disagree- 
able voice in reciting his prayers in the mosque was annoying 
to everybody. One day some one asked him how much he 
was paid for reciting. 4 Paid ! ' he responded, ' I am not paid ! 
I recite for the sake of Allah ! ' 4 Then,' replied the other, 
4 for Allah's sake don't ! ' 

A sequel in the same vein is told by Saadi to prove the 
occasional value of a disagreeable voice. A certain muezzin 
in the mosque had so harsh a voice that his call to prayer only 
kept the worshippers away from service. The prince who was 
the patron of the mosque, being tender-hearted and not wish- 
1 Saadi, Gulistan, tr. Eastwick, pp. 101-102, 2d ed., London, 1880. 



SOME ANECDOTES FROM SAADI 335 

ing to offend the man, gave him ten dinars to go somewhere 
else, and the gift was gladly accepted. Some time afterward 
the fellow returned to the prince and complained that an 
injustice had been done him by the smallness of the donation: 
4 for,' said he, 4 at the place where I now am, they offered me 
twenty dinars to go somewhere else and I'll not accept it.' 
4 Oh,' laughed the prince, 4 don't accept it, for if you stay longer 
they will be glad to offer you fifty.' 

A single other illustration of Persian humor from Saadi may 
be added. The point of the story is this : A man who was 
suffering from inflamed eyes went to a horse-doctor for treat- 
ment. The veterinary gave him some of the salve that he used 
on animals and the man lost his eyesight. He then brought 
a suit in court to recover damages. The judge, after weighing 
the evidence in the case, handed down his decision as follows : 
4 There are no damages to be recovered ; the man would never 
have gone to a veterinary if he had not been an ass ! ' 

This and a score of other instances might be cited to show 
the light touch of Saadi's wit beside his acknowledged poetic 
talents. His rare gift as a poet is seen particularly in the 
Bostan and in the Divan, a collection of his short poems in 
the lyrical strain, which justify his title of the 4 Nightingale of 
Persia.' Some of the metrical stanzas in the Gulistan are 
gems of poetic thought, and as an example of Saadi's fancy I 
quote from it the following lines, which show his fineness of 
imagination and delicacy of touch. 

* I saw some handfuls of the rose in bloom, 
With bands of grass suspended from a dome. 
I said, " What means this worthless grass, that it 
Should in the roses' fairy circle sit?" 
Then wept the grass, and said, " Be still ! and know, 
The kind their old associates ne'er forego. 
Mine is no beauty, here, or fragrance true; 
But in the garden of the Lord I grew." ' x 

The hills beyond the grave of Saadi and to the east of the 
1 Saadi, Gulistan, tr. East wick, p. 115. 



336 SHIRAZ, THE HOME OF THE PERSIAN POETS 

Allahu Akbar Gate possess one or two points of interest that 
may be mentioned. One of these is a large hollow in the rock, 
partly natural and partly artificial, called from its shape the 
4 Cradle of the Demon ' (Kahvdrdh-i Dlv), although its precise 
origin is not known; the other is a ruined structure situated 
somewhat east of it and known as ' Bandar's Fortress ' (Kal'ah-i 
Bandar), which is supposed to be the remains of a Sasanian 
castle. Near this are two very deep wells, one of which is 
known as 4 Ali's Well ' (Chdh-i Murtazah All) and described as 
a pool at the bottom of a series of steps surmounted by a build- 
ing which gives the place the character of a shrine. It is said 
to occupy the place of an old fire-temple, and the story goes 
that the well sprang up as a miracle to quench the flame of the 
old Zoroastrian faith when the true religion of Mohammed 
came into Persia. 1 There are also some Achsemenian remains 
about four miles southeast of Shiraz and still farther beyond 
there are some sculptures of Sasanian kings who were Zoroas- 
trians, but I did not inspect them. 2 

My allusion to Zoroastrianism leads me to speak of the so- 
called Gabars, or fire-worshippers, of Shiraz, as the city knew 
only their religion in Sasanian times, whereas scarcely fifty of 
their faith now live there. 3 I took the earliest opportunity of 
sending to Rustam Shah Jahan, the leading merchant among 
them, the letter I carried from his brother in Isfahan. 4 He 
occupied a shop adjoining the main bazaar, and as I entered the 
room behind the outside booth, I found a number of persons 

1 See Browne, A Year Amongst the and Coste, Voyage en Perse, Ancienne, 
Persians, p. 286 ; Curzon, Persia, 2. 1. pi. 55, and photographed by Stolze, 
108 ; Ker Porter, Travels, 1. 698. Persepolis, 2. pi. 96 ; cf. also Perrot 

2 The earliest notice that I have and Chipiez, Histoire de VArt, 5. 754. 
seen of these ancient monuments is 3 The precise number at the time of 
inMasudi(x.D. 843), Les Prairies a" 1 Or, my visit was 42, according to the 
ed. Barbier de Meynard, 4. 79. They statistics I subsequently obtained at 
have been described by various writers, Teheran from the Secretary of the 
among them Ker Porter, Travels, 1. Society for the Amelioration of the 
698-706, and Curzon, Persia, 2. 95, n. Persian Zoroastrians. 

2, and have been drawn by Flandin * See pp. 274-275, above. 



THE ZOROASTRIANS OF SHIRAZ 337 

gathered there. Remembering my Isfahan experience and 
knowing that Shiraz was Islamitic to the extreme of fanati- 
cism (so much so, that some of my friends later expressed sur- 
prise that my body-servant did not meet with persecution there, 
as he was a Christian convert from Mohammedanism), I began 
by making commonplace observations and inquiries in Oriental 
style, until I should be surer of my ground, and only indirectly 
indicating my interest in the religion. In this case, however, 
I found there was not the slightest occasion for reserve, as my 
host Rustam had been prepared by his brother's letter, and 
he told me that all the persons who were present were Zoro- 
astrians, so that we could speak without hesitation on religious 
matters. 

From the conversation I learned that the Zoroastrian 
community in Shiraz keep up their religious observances and 
beliefs, in a general sort of way, but not so strictly as at Yezd 
and Kerman. They have no regular dastur, or High Priest, 
nor have they any fire-temple, whereas in antiquity there must 
have been at least one pyrseum at Shiraz, as is shown by the 
remains of an ancient Atash Kadah which a Mohammedan 
pointed out to me on the hill overlooking the city. No 
dakhmah, moreover, is kept up by the Zoroastrians of Shiraz, 
although this could hardly be expected in so small a community 
and it is their practice to inter the body in the earth, placing 
stones around it and over it. They possess no manuscripts 
of the Avesta, so far as I could learn, but they encouraged 
me in my hope of rinding copies at Yezd, the chief centre of 
the Persian Zoroastrians, and told me that I would there meet 
the Chief Priest of the Faith and be able to learn from him 
more about religious matters. Despite their lack of knowledge 
concerning their creed and it would be unreasonable, per- 
haps, to expect merchants and traders to possess technical 
information on theological points I was favorably impressed 
by these believers in Ormazd. They seemed honest and 
thrifty, and fairly prosperous, considering the fact that they 



338 SHIRAZ, THE HOME OF THE PERSIAN POETS 

have had to live for over a thousand years under restriction 
and persecution, and they appeared also to cultivate the 
sterling virtues which their prophet of old enjoined. This 
fact made me more anxious than ever to visit Yezd, so I began 
my preparations for departure on the third day, bidding adieu 
to them and to my Christian hosts at the English mission, as 
well as to other friends. 



CHAPTER XXII 
FROM SHIRAZ TO YEZD 

' From Shiraz to Yezd is seventy -four farsakhs.' 

IBN HAUKAL, Geography, tr. Ouseley, p. 111. 

IT was toward mid-day on the sixth of May, and the sun 
was scorching hot, when I set out from Shiraz on my way to 
the city of Yezd to visit the Zoroastrians in that ancient strong- 
hold of the faith. I was provided with a letter from the Per- 
sian governor of Shiraz, directing that certain privileges and 
attentions be extended to me on the journey, and was furnished 
with an authorization from the Director General of the Per- 
sian Customs and Post, enabling me to procure horses or other 
means of transport over portions of the route that had not yet 
been formally laid out. As I knew on my southward journey 
that I should have to travel for three days over the same route 
by which I had come from Pasargadae to Shiraz I had taken 
the precaution to pave the way with silver. The investment was 
expensive, but it proved to have been worth making, for it quick- 
ened the pace of the post-horses and hastened the movements o/ 
the men at the halting-places, and haste is a rare thing in Persia. 
By this expenditure, by liberal use of whip and spur, and by 
reducing the time of sleep at night to three or four hours, with 
cat-naps stolen at odd moments during the day, I was able to 
shorten the ordinary time of ten days between Shiraz and Yezd 
to five days and a quarter. 

Toward evening of the first day, as I retraced my route 
northward, I reached once more the ruins of Persepolis. The 
desolate terrace looked picturesque in the moonlight, but I did 

339 



340 FROM SHIRAZ TO YEZD 

not halt again to wander through its crumbled palaces and 
deserted halls. On the next day, however, I paid another 
visit to the sepulchres of the Achsemenian kings at Naksh-i 
Rustam and the Magian altars and rock-hewn tables on 
the cliff that overlooks them, and I reached Pasargadse and 
the tomb of Cyrus that same afternoon, spending the night 
once more at Meshad-i Murghab. After a hard march of 
seven farsakhs in six hours over rough hills and stony ways 
I rested for an hour at Deh-Bid, and then resumed the jour- 
ney and arrived, about five o'clock, at Khan-i Khorah, where 
the night was to be passed and I could at last strike eastward 
on the trail to Yezd. From my previous experience I remem- 
bered Khan-i Khorah as a desolate place at which I was held up 
for two hours and a half because the post-horses had roamed 
miles away over the plain to graze. The interval till their 
return I had spent in a wretched hovel surrounded by natives 
in a dazed condition from opium-smoking, the unfortunate 
effects of which common vice are only too frequently seen in 
Persia. My present impression was more pleasant, as I was 
conducted to a fairly comfortable little house, not far from 
the caravansarai, where I could spend the night. 

There was a small but pretty garden at the rear of this simple 
abode ; the fruit trees were in full blossom, and everything 
looked cheerful in the light of the setting sun. I hardly had 
time to dispose of my pack and arrange my camp-bed for the 
night, before the head man of the village paid me a visit. He 
had come for medical aid, he said; his wife was suffering 
from toothache. I prescribed as best I could from my limited 
stock of drugs, but I soon became convinced that the real 
patient was my visitor himself and that he hoped he might 
have some arac and tobacco added to the prescription. I 
yielded to his broad hints so far as the tobacco was concerned, 
for I had one or two cigarettes left in my case, but I omitted 
the spirituous part of the remedy, probably to my visitor's 
regret. 




FROM DKH-BID TO KHAN-I KHORAH 




OPIUM-SMOKERS 



FROM SHIR A Z TO ABARKUH 341 

The night was short, as Persian nights are in the spring, 
and when one is trying to make time, it is necessary to rise 
long before three o'clock in order to start from the caravansarai 
by daylight. In fact, I saw more sunrises in Persia than I 
ever expect to see again in all my life. Darkness was melting 
into dawn when I found myself again in the saddle, with a 
cavalcade of five horses and three footguards to accompany me 
over the barrier of mountains that lay between Khan-i Khorah 
and the sandy deserts of Abarkuh and Yezd. The scenery for 
a time was superb. Steep ascents, deep ravines, narrow 
gorges, and wild passes succeeded each other in great variety. 
One welcome excuse for rest and refreshment was found at the 
foot of a rocky height, where a crystal spring pulsed up with 
cool water, offering the last chance for a drink before crossing 
the desert to Abarkuh. Leaving the great ridge behind us 
and dismissing most of the guards, as they were no longer 
needed, for Persian highwaymen operate chiefly in the moun- 
tain passes, we entered upon the arid tract marked on the 
maps as the Sandy Desert, whose barren stretch as far as Yezd 
is broken only by the oases of Abarkuh and Deh-Shir and the 
mountains beyond the latter. 

The town of Abarkuh, or Abarguh, (Abarkuh) is evidently 
of great antiquity. The Persians in earlier and later times 
have commonly pronounced the name as Barkuh or Warkuh, 
understanding it to signify ' upon the mountain,' or 4 over the 
mountain,' especially when speaking from the standpoint of 
the desert towns that lie beyond it and are separated from it 
by hills. 1 The tenth-century geographer Istakhri (c. A.D. 950) 
describes the town as follows : 

< Abarkuh, or Warkuh, is a fortified city, densely populated, and 
of about one-third the size of Istakhr. The houses are lattice-worked, 

1 So my informant, Khodabakhsh Goeje, Bibl. Geog. Arab.} Istakhri, 1. 

Rais of Yezd ; see also the state- 126 (Abarkuh}, Mokaddasi, 3. 437 

ment of Yakut, p. 8, 'the Persians (Barkuh}, and Al-Hamadhani, 5. 203 

say Varkuh for Barkuh, "upon the (Abarkuiah) ; see also Schwarz, Iran 

mountain." ' Cf. likewise (in De im Mittelalter, p. 17. 



342 FROM SHIR A Z TO TEZD 

and most of the buildings, as well as the buildings at Yezd, are built 
with colonnades (or vaulted domes). 1 It is a barren place ; there are 
no trees or gardens around it except at a distance, but the soil is 
productive and the living is cheap.' 2 

On the right of the road as we approach from the southwest 
there is a large fortress-like ruin called Dakhmah-i Ddrab, 
4 Structure of Darab,' after the name of Darius Codomannus, 
the last of the Achsemenians. The term dakhmah, which is ap- 
plied to it, is used, as also in Turkish, for a structure in general, 
and is not to be confused with the usage of the word dakhmah in 
the technical sense of 'Tower of Silence.' The ruin looks like 
a deserted stronghold of considerable antiquity. 

On the left of the road, upon an elevation, stands the 
Dakhmah-i Gabrahd, ' Structure of the Gabars,' another ruined 
edifice of mud and sun-dried bricks, closely resembling the 
Atash Kadah, or 4 Shrine of Fire,' near Isfahan, which I have 
already described. 3 Adjoining this so-called Gabar structure 
stands another building, evidently an old temple, but, like the 
shrine, a crumbling ruin. The site of these ancient Iranian 
structures appears to be a historic one, as is shown by some 
allusions in Mohammedan writers. Ibn Haukal, for example, 
in the tenth century, says : 

'In the vicinity of Abarkuh are considerable heaps of ashes. 
The common people say that here was the fire of Mmrod (into which 
he caused Abraham to be thrown), but this is not true; the fact 

1 The Arabic expression mugtabekat nard, Diet. geog. de la Perse, p. 8 

al-bind, 'of netted-work buildings,' ('cintre'e'), and of Schwarz, Iran im 

seems to allude not to the crowding Mittelalter, p. 18 (' mit gewolbter 

together of the houses, but to the open Decke'). Such domed mud roofs are 

or trellis-work style of architecture common in Persia, as is shown also 

seen in the front of the building. in some photographs that I have of 

The second term (Arabic word azaj> Yezd and Kashan, but this explana- 

note i, p. 351) appears to allude to tion seems to me not so good, 
colonnades or arched galleries. See 2 Istakhri, ed. De Goeje, 1. 126. 

the pictures of Yezd in Malcolm, Five The Persian version adds to the 

Years in a Persian Town, pp. 134, Arabic a note on the export of fruits 

184, 216. But 'vaulted domes' is from Abarkuh. 
the rendering of Barbier de Mey- 8 See pp. 252-261, above. 



RUINS OF AN ANCIENT FIRE-TEMPLE 343 

is that Nimrod and the kings of Canaan dwelt in the land of 
Babylon.' 1 

This statement of Ibn Haukal is evidently based on older 
authorities, for he re-edited Istakhri, whom Yakut (c. A.D. 
1220) briefly cites. Yakut has a somewhat similar legend, 
but differs from this account in certain noteworthy details. 

1 At Abarkuh there is a large hill of ashes, which the inhabitants 
claim was the fire of Abraham, lighted by Bardah and Salamah. 2 
But in the Book of the Avesta, which is the book of the Magi, I 
have read that So'da (Sudabah), 3 the daughter of Tubba, wife of Kei 
Kaus, fell in love with his son Kei Khosru (Siavash). 4 She en- 
deavored to seduce him, but he rejected her, whereupon she told his 
father that he had tried to dishonor her, which was a lie. There- 
upon Kei Khosru built a large fire at Abarkuh for an ordeal and 
said : " If I am innocent the fire will not harm me ; if I am guilty, 
as claimed, the fire will devour me." At this he entered into the 
fire and came out of it unhurt, without having suffered any harm. 
In this way he dispelled the entire charge against him. The ashes 
of that fire are to be seen at Abarkuh in the shape of a large hill, 
and to-day it is called the Mountain of Abraham. But Abraham 
never saw the land of Fars, nor entered it ; he abode in Kutharabba, 
in the land of Babylon. I have read elsewhere, however, that 
Abraham came to Abarkuh and prohibited its inhabitants from em- 
ploying cows in farming ; consequently they do not employ cows in 
this way, although there are plenty in the country. Abu Bekr 
Mohammed, who is known as al-Harbi of Shiraz, . . . states : " I 
was in Abarkuh three times, but I never saw rain fall within the 
walls of the city, and the people said that this was owing to the 
prayers of Abraham."' 5 

!Ibn Haukal (c. A.D. 975), tr. .* Firdausi (tr. Mohl. 2. 164-195) 

Ouseley, p. 130. and the author of the Haft Iklim 

2 The Arabic seems to mean 'which narrate the story not of Khosru, but 
Bardah and Salamah lighted upon it of Siavash, which is apparently more 
(i.e. the hill) ' ; but Barbier de Mey- in accordance with the facts. 

nard, Diet. geog. de la Perse, p. 8, 6 Yakut, Geographisches Worter- 

renders * qu' Abraham alluma pour buck, ed. Wiistenfeld, 1. 86, Leipzig, 

Berdeh et Selamah.' 1866 ; cf . also the translation of Yakut 

3 Sudanah, as well as Sudabah, is by Barbier de Meynard, Diet. geog. 
found as a variant form of the name. de la Perse, pp. 8-9. For assistance 
For references, see p. 344, n. 1. with the Arabic I am indebted to 



344 FROM SHIRAZ TO YEZD 

This legend of the fire-ordeal, as preserved in Yakut, is suffi- 
cient to prove the sacred associations of the place and its right 
of title to having been the site of a pyrseum, even if Firdausi 
and Thaalibi, in their accounts, do not give the precise locality 
where the ordeal took place. 1 We may consider, therefore, 
that still another identification has been added to the list of 
sites of ancient fire -temples in Persia and that the ruined 
shrine at Abarkuh commemorates, in its location at least, the 
place where the honor of Siavash was vindicated. 

In this connection I may mention a legend connected with 
a Mohammedan saint at Abarkuh, as told by Hamdallah Mus- 
taufi (A.D. 1340). This author narrates that there was in the 
town the tomb of a saint of Islam, named Taus al-Haramein, 
which literally means ' Peacock of the two Sanctuaries ' (that 
is, of Mecca and Medina), and he reports that the walls of the 
building would not allow a roof to enclose it, for if a roof was 
erected over the tomb, some supernatural power always de- 
stroyed it. He likewise adds that no Jew could exist in Abar- 
kuh for more than forty days. 2 

When I entered the town, I went at once to the Reis of 
Abarkuh and presented my letters from the Governor of 
Shiraz and from the Director of the Post. The town is only a 
small place nowadays, and the Reis could not obtain horses for 
me, but he managed to procure four mules to make up a cara- 
van for crossing the desert, and arranged that the start should 
be made soon after midnight. This gave me time to rest in 
the afternoon, which I spent looking out from the window 
upon a little stream bordered with sycamore trees whose 
branches rang with the song of birds. When evening came, 
I fell asleep and did not wake until midnight. After some 

the kindness of my colleagues Yohan- Thaalibi, Histoire des Hois des 

nan and Gottheil. Perses, tr. Zotenberg, pp. 171-186. 

1 For the well-known account of 2 See Hamdallah Mustaufi, Nuzhat 

Sudabah and Siavash in Firdausi's al-Kulub (L. 174 gr), cited by Le 

Shah Ndmah, see the translation by Strange, JRAS. 1902, p. 519, n. 1. 
Mohl, 2. 153-195 ; and compare 




I -^ 





LETTER OF CONVOY FROM THE GOVERNOR OF SHIRAZ 



ACROSS THE DESERT TO DEH-SHIR 345 

delays we finally got our company of mules, muleteers, and 
guards under way and resumed the journey. The moon was 
flooding the sky with a soft Oriental light and the nightingale 
was singing in the tamarisk bush behind the mud walls as we 
threaded our course through the narrow streets. The out- 
skirts of the town were presently left behind, the roads began 
gradually to merge into sandy trails, and an hour later the 
swiftly rising sun had brushed aside the silvery veil of night 
and gleamed over the desert, upon which we had entered. 

From this point on, for fourteen farsakhs, or nearly fifty 
miles, the track led straight across an arid waste marked 
only by hoof-prints in the snowy sand and by skeletons of 
beasts of burden that had fallen by the way. The air was 
warm, but not excessively hot, and every now and then a 
breeze sprang up, sweeping powdery whirlwinds off into the 
distance to perish in the sand that gave them birth. Mirage 
after mirage arose to surprise the eye and give play to the 
fancy, thus relieving the tedium of the march. Once in a while 
the beaten track forked for a mile or more, but the two 
branches always rejoined, pointing again toward the oasis town 
of Deh-Shir, which, though miles away, was our nearest goal. 
There was little need for guides and none for guards, so we dis- 
missed a part of our company of attendants, but could not spare 
the muleteers, who found plenty to do in keeping our unruly 
animals in order. The pack-mule kept bolting from the track 
at the most unexpected moments, and had constantly to be re- 
captured and brought back to the train. My own beast was 
always ready for rebellion at the slightest provocation, and I 
narrowly escaped having my skull fractured by his heels. 
A halt had been made in the ankle-deep sand, and, as I was 
remounting, the badly girt saddle slipped and I was thrown 
under the beast's heels, with my foot caught in the stirrup. 
A shower of terrific kicks, resembling those seen in comic 
pictures of the Mississippi mule, filled the air. I was bruised, 
dragged, and torn, but I managed to shield my head from the 



346 FROM SHIRAZ TO YEZD 

vicious animal's hoofs, and he was finally reduced to subjec- 
tion. 

The sun was well on its way in the western sky when our 
long, heated march of nearly fourteen hours, without a drop 
of water to drink, ended at the green oasis of Deh-Shir. The 
lord of the town extended a kindly welcome in Oriental fashion 
and provided a hearty meal, served by his Persian servants, one 
of whom was a eunuch slave. He then conducted me to the 
terrace on the roof to show me the view over the oasis, the 
desert, and the hills beyond, expressing a doubt whether any 
land could compare in beauty with Iran. I appreciated his 
enthusiasm, even if I did not fully share in it. He inquired 
about my own country, the location of which I could best 
convey to his mind by describing it as a far-off land, many 
thousand f arsakhs distant, and explaining that eight days of the 
journey had been across 4 the black water,' whereupon he at 
once exclaimed 4 Yankl DunydJ 1 which means literally ' New 
World' in Persian, whereas the term America conveyed 
nothing to him. He commented upon my tanned face, which 
was almost as dark as his own, and talked about the life at 
Deh-Shir, which, it appeared to me, must be rather monotonous. 
He knew the surrounding country well, but the main fact that 
I elicited regarding antiquities was that some stone coffins with 
human remains had once been found in the vicinity. Our view 
over the landscape was enjoyed through a telescope, of which 
he was the proud possessor, having inherited it from a father 
or grandfather, to whom it had. been apparently presented 
many years ago by its original owner, an English officer, if I 
may judge from the inscription engraved upon the brass. As 
I was a foreigner, he expected me to be equipped with tobacco, 
and I was sorry not to have any American cigars to offer him 
as samples of 4 the weed ' in our country, but I tried in other 
ways to return his hospitality and to reciprocate his acts of 
courtesy. 

1 Pronounced ' Yankee Doonya.' 



DEH-SHIR AND TAFT 347 

When I was taking my leave and asking directions and 
information, he told me that the road through all his district 
was safe ; there had been bandits operating recently in the hills, 
he said, but he had attended to the last of these about a fort- 
night previously. The closing sentence was accompanied by a 
significant gesture with his hand drawn like a knife across his 
throat to indicate the punishment he had inflicted upon the 
thieves. Nevertheless, to make my safety assured, he accom- 
panied me himself, after the Persian manner, some distance 
beyond the town and gave a guide and guard to conduct me 
across the Shir Kuh Mountains. 

The road led over wild and jagged hills, and the rocks took 
on fantastic and picturesque shapes in the rich moonlight as 
our little caravan clambered up and down the rough ways to 
the pretty hamlet of Deh-Zeresh. After a few hours' sleep we 
pushed forward again over the hills to Aliabad, where our 
pack-mule was exchanged for two small donkeys, and we 
enjoyed a half hour's rest and breakfast under the trees near a 
little brook. We had not proceeded far after resuming the 
march, before one of the donkeys had a severe fall and received 
a cruel cut in his chest. I urged the muleteer to send the 
animal back to Aliabad, as I would pay for a substitute, 
but he gave no attention to the poor beast's suffering, for 
animals seem to have no rights in Persia, where a society for 
the prevention of cruelty would have an ample field for activity, 
and he seemed almost amused that I should consider the matter 
of an animal's pain. I was obliged, therefore, to postpone my 
sympathy until I could reach Yezd and there have the don- 
key's wound properly treated. 

As we proceeded on our way the signs of increasing civiliza- 
tion and comparatively fair prosperity grew more and more 
marked, and about an hour after noonday we reached Taft. 
This is a well-to-do suburb of Yezd and a place that counts 
among its population a considerable number of Zoroastrians, 
mostly gardeners. We halted here long enough to have one 



348 FROM SHIRAZ TO YEZD 

of the mules shod, as its shoes had been loosened and lost on 
the hard journey over the rocky hills and pebbly roads of the 
Shir Kuh. This gave the muleteer and the guides a chance to 
rest ; they had been walking forty miles a day, yet without 
apparent fatigue, for when called upon for a spurt, they were 
always ready to set off at the pace of a sprinter. These men 
have evidently kept up the tradition of the ancient Persian 
couriers, and I was told by English residents in Yezd some 
remarkable stories of the feats of endurance and speed these 
desert runners were able to accomplish. 

Most of the remainder of the afternoon was spent in 
reaching Yezd, ever tantalizingly near because of the clear- 
ness of the atmosphere, yet still remote because of the 
expanse of desert around it. This encircling tract of sand, 
which is nearly thirty miles broad and many more miles long, 
is bounded on the south and west, and partly on the north, by 
ranges of rugged hills, while a belt of sand-dunes on the east 
reaches almost to the. walls of Yezd, but is kept back by 
gardens, whose green tinge affords a welcome contrast to the 
brown and parched aspect of all around. 

Yezd is a city of considerable antiquity, since its name 
apparently occurs, under the form of 'lo-art^at (Isatichai), in 
Ptolemy's Greek geography, where it is named among the few 
towns of the desert of Carmania. 1 According to Persian 
tradition, moreover, it must have been known in Alexander's 
time, having been used by the victorious invader as a place 
of confinement for his prisoners of war. 2 The common view 
associates the name Yezd, or Yazd, with Yazdagard I (A.D. 399- 
420) , father of Bahram Gor, probably as the rebuilder rather 
than the original founder. 8 

1 Ptolemy, Geography, 6. 6. 2. This see Sykes, Ten Thousand Miles in 
name is not to be confused with Istakhr Persia, pp. 419-420. 

(Curzon, Persia, 2. 239). 8 For remarks on the name of Yezd 

2 On this latter point (drawn from and Kathah, see note i at the end of 
Hafiz) and for the common view re- this chapter, p. 351. 

ing Yezd (Yazd) and Yazdagard, 




VIEW OF YEZD 




A STREET IN YEZD, SHOWING A WIND-TOWER 



THE CITY OF YEZD 349 

During the earlier years of Mohammedan rule Yezd became 
a place of refuge and stronghold for the Zoroastrian Gabars, 
probably because of its remote situation in the desert, although 
it was never out of touch with the rest of Persia. The first 
European known to have visited it was Marco Polo in 1272, 
who calls it ' the good and noble city of Yasdi.' l The Italian 
friar Odoric of Pordenone, who came here about fifty years 
later than Marco, speaks of the town as ' Geth,' ' Gest,' or 
4 lest,' 2 and Josafa Barbaro, the Venetian (1474), writes the 
name as 4 Ies' or 4 Jex.' 3 

Notwithstanding the earlier and later importance of the 
city, Yezd has little to offer in the way of sight-seeing, and it 
certainly cannot lay any claim to natural beauty. One rides 
for hours through narrow winding streets, with nothing to see 
but walls of clay, the backs of houses, a streak of sky (which 
blazes as soon as summer begins), and a glimpse of high wind- 
towers rising from the roofs of the better dwellings. These 
lofty air-shafts (Idd-girs, ' wind-collectors ') resemble square 
chimneys with slat-openings above to catch the slightest breeze. 
They are rendered necessary by the long continuance of the 
heat during the summer months, and are characteristic of 
Yezd. 

The area of the city is considerable ; Odoric spoke of the 
town as 4 walled and of V myles in circuite,' and the same is 
practically true to-day, except that the circuit may be some- 
what larger and the walls have fallen in some places. The 
fort within the city was erected or reconstructed in the year 
1137, and is built of sun-dried bricks and mud, but neither 
this nor the citadel of the Governor inside its walls would 
afford any material defence at the present time. 4 In several 

1 Marco Polo, ed. Yule, 1. 88. 1. 381 ; Sykes, Ten Thousand Miles 

2 Odoric de Pordenone, ed. Cordier, in Persia, p. 421. The Arabic allu- 
p. 451. sions in note i at the end of this chap- 

8 Josafa Barbaro, 49. 59, 73, 82. ter show that the citadel must have 

* On this point see Curzon, Persia, existed before 1137 and that it was 
2. 240 ; Landor, Across Coveted Lands, probably rebuilt in that year. 



350 FROM SHIRAZ TO YEZD 

parts of the town there are public squares, the one near the 
Governor's palace being somewhat effective, 1 and among the 
features of Yezd are numerous arches over the narrow streets ; 
but the only public building that has any claim to considera- 
tion is the Friday Mosque (Masjid-i Jum'-ah). It was built 
in 1119 by Sultan Allah ad-Daulah Garshasp, who thus won 
for Yezd the cherished title Ddr al-'-Ibddat, 'Abode of Wor- 
ship,' a dignity which was rendered still further assured in 
after centuries by the pious munificence of some of the later 
rulers, traces of whose donations, especially two beautifully 
carved wooden doors, are still to be seen. 2 

The population of Yezd is estimated at between thirty and 
forty thousand, or at nearly sixty thousand if the villages in 
the district be included in the count. A large part of the 
population is engaged in silk-weaving, which is one of the 
chief industries of this region. But this was a matter of minor 
interest to me at the moment, nor did I pay much attention 
to the manner in which Yezd meets the difficult problem 
which it has to face in securing an adequate water-supply, 
nor again was I especially interested in the bazaars, or in 
the trade and commerce of the city, except that I found 
it very convenient that Yezd had a branch of the Imperial 
Bank, from which I could draw upon my letter of credit. 
My real interest in Yezd was to see the Zoroastrians, and I 
shall devote the next chapter to the visit that I paid to this 
interesting community. 

1 For a picture, see Malcolm, Five gether with quotations and a descrip- 
Years in a Persian Town, p. 184. tion of the mosaic dome near the fort, 

2 For notes upon this subject, to- see Sykes, Ten Thousand Miles, p. 421. 



NOTES 351 



I. NOTE ON YEZD AND KATHAH 

There is good reason for claiming that an old name for Yezd or a collateral 
title for the main quarter of the city was Kata, judging from the article Kathah 
in Istakhri and other Perso-Arabic geographers. Istakhri (ed. De Goeje, 
Bibl. Gfeog. Arab. 1. 125) writes as follows: 'One of the most famous cities in 
the district of Istakhr on the confines of Khorasan is Kathah. It is the chief 
place (haumah) of Yezd and Abarkuh . . . ,' and he continues : ' Kathah, the 
chief town of Yezd, is a city located on the edge of the desert. It has good 
and healthy air from the desert and possesses also the comforts that belong 
to large cities. Its districts are noted for their fertility, and living is cheap. 
The houses are generally built of clay and have colonnades (or, 'vaulted 
domes' azaj, plur. from azaj, see p. 342, n. 1, above). It has a fortified 
citadel, with two iron gates, one of which is called the 'Gate of Izad' (it 
would be tempting to compare Ptolemy's 'Icra-Hxcu, as a possible corruption 
from 'lo-aT-re^xea, if not for the Persian variant Bdb-i Andar, 'Inner Gate'); 
the other is called 'Gate of the Mosque' (Bab al-Masjid), on account of 
its proximity to the chief mosque. The mosque of the city is located in the 
suburbs. The water for the city comes through kandts, but there is also a 
river which comes from the vicinity of the Kalah, or citadel (i.e. the Kal'at 
al-Majus, or Kal'at-i Zard), near a village where there is a lead-mine. It is 
very delightfully located and has fertile and extensive districts. The town and 
its districts abound in fruit, so much so that it is exported to Isfahan 
and other places. The mountains have many trees and plants, which are 
also exported. Outside the city there is a suburb which has fine houses 
and bazaars. For the most part the inhabitants are men of education and 
letters.' Ibn Haukal (ed. De Goeje, 2. 181, cf. also 2. 196) repeats this state- 
ment verbatim and adds : ' As regards the district of Istakhr, the division of 
Yezd is the largest division in it, and in this there are the following cities : 
Kathah, which is the citadel, and Maibud, Nairi, and Al-Fahraj, and in all the 
divisions there is none that has four pulpits (minbars') except this one.' In his 
list of fortified places (taken from Istakhri, 1. 112) Ibn Haukal (2. 187) 
mentions Kathah directly after Istakhr (Persepolis), saying, ' The city of 
Kathah has a fortress.' Yakut has two separate paragraphs devoted to 
Kathah and Yezd. In the first (ed. Wiistenfeld, 4. 239) he says : ' Kathah, 
a place of Pars ; it is the chief city (haumah) of the district (kurah) of 
Yezd and belongs to the district (kuraJi) of Istakhr.' In the second 
paragraph (ed. Wiistenfeld, 4. 1017) he states : * Yezd, a place midway 
between Nisapur, and Shiraz, and Isfahan ; it is reckoned to the province of 
Pars and belongs to the district of Istakhr ; it (Yezd) is the name of the region 
and its citadel is called Kathah ; between it and Shiraz the distance is seventy 
farsakhs.' Compare also the French translation by Barbier de Meynard, Diet, 
geog. p. 475 (Kathah), p. 611 (Yezd) ; likewise Schwarz, Iran im Mittel- 
alter, p. 19. (For help in translating the Arabic passages I am indebted to 
the kindness of my colleagues Gottheil and Yohannan.) With the name Kathah 



352 FROM SHIRAZ TO YEZD 

(Old Iran. *Kata) we may compare the first part of the Latin name Cetrora 
for the corresponding place in the geographical list of the Tabula Peuting- 
eriana, as Toinaschek (Sb. Akad. Wiss. zu Wien. 102. 165, Vienna, 1883) 
derives that name from the Old Iranian word kata, used in the Avesta to 
designate an * excavation,' house dug to receive corpses, and the presumably 
Iranian element ravara, found in the Mod. Pers. Eudh-ravar (cf. Ptolemy's 
'Podpa, Geog. 6. 5. 2). This suggestion is plausible in the main, since it 
is reasonable to suppose that Istakhri, who was a native of Stakhr, would 
have been well acquainted with the territory around Yezd. The details 
which he gives regarding Kathah and the river that flows from a village 
near a citadel where there is a lead-mine, would answer in general to the 
region about the modern village Kattu near the 'Yellow Castle' (Kal l ah-i 
Zard}, as described by Browne, A Year Amongst the Persians, p. 358. [I find 
the identity of Kathah and Yezd accepted also by Le Strange, Lands of the 
Early Caliphate, p. 285, Cambridge, 1905 ; but Khodabakhsh Bahrain Kais writes 
me that the name Kathah seems to be absolutely unknown in Yezd, although 
there is a large village called Kahtu or Kathu about 20 farsakhs south of the 
city. Proof-sheet addition.] 



II. NOTE ON Two OLD ITINERARIES FROM SHIRAZ TO YEZD 

I add two old itineraries, the one by an Oriental, the other by an Occidental, 
covering in part at least the route followed in this chapter. The first is by the 
Arab geographer Istakhri in the tenth century and reads as follows (ed. De Goeje, 
1. 129-130) : 'Route from Shiraz to Kathah, the chief town of Yezd, along the 
Khorasan route : From Shiraz to the village of Zarkan (Zargan) 6 farsakhs ; 
from Zarkan to the city of Istakbr 6 /; from Istakhr to the village of Bir (v.l. 
Bin, Plr, Giz) 4 /; from Blr to Kahmand (v.l. Kihandah, Kihandaz; Ouseley 
reads Kahndaz) 8 /; from Kahmand to the village of Bid (Deh-Bld, 'Willow 
Village ') 8 /; from the village of Bid to the city of Abarkuh 12 /; from Ab- 
arkuh to the village of Al-Asad ('Lion Village,' v.l. Deh-Shir) 13 /; from the 
village of Asad to the village of Al-Juz (or Jauz, v.l. Al-Khur, Deh-i Khvar) 6/ ; 
and from the village of Al-Juz to the village Kalah al-Majus 6 /; and from 
Kalah al-Majus to the city of Kathah, the chief place of Yezd, 5 /. ' The second 
is the memorandum of the route of Josafa Barbaro, in the fifteenth century, 
from Persepolis to Yezd (ed. Hakluyt, 49. 81-82), and it reads as follows: 
'Fromthense, iij daies io r ney, yo w come to a towne called Dehebeth (Deh-Bid), 
wheare they vse tillaige and making of fustians. Twoo daies io r ney further 
ye coine to a place called Vargari (or Vargan), which in tyme past hath been a 
great and a faire towne ; but at this pfit it maketh not aboue m 1 houses, in 
the which they also vse tillaige and making of fustians, as is aforesaid. 
Foure daies io r ney thense ye come to a towne called Deiser (Deh-Shir), and iij 
daies io r ney further an other towne called Tafte (miswritten as Taste), 
from whense following that waie another daies io r ney ye come to Jex, of the 
which I haue made sufficient mencon before.' 



CHAPTER XXIII 
THE ZOROASTRIANS OF YEZD 

' From Yezd's eternal Mansion of the Fire.' 

MOORE, Lalla Rookh. 

SITUATED amid a sea of sand which threatens to ingulf it, 
Yezd is a symbolic home for the isolated band of Zoroastrians 
that still survives the surging waves of Islam that swept over 
Persia with the Mohammedan conquest twelve hundred years 
ago. Although exposed to persecution and often in danger 
from storms of fanaticism, this isolated religious community, 
encouraged by the buoyant hope characteristic of its faith, has 
been able to keep the sacred flame of Ormazd alive and to 
preserve the ancient doctrines and religious rites of its creed. 

When the Arab hosts unfurled the green banner with the 
crescent and swept over the land of Iran with cry of Allah, 
shout of Mohammed, proclamation of the Koran, fire, sword, 
slaughter, enforced conversion, or compulsory banishment, a 
mighty change came over Persia. The battle-grounds of 
Kadisia and Nihavand decided not Iran's fate alone, but Iran's 
faith. Ahura Mazda, Zarathushtra, and the Avesta ceased 
almost to be known, the temple consecrated to fire became a 
sacrifice to its own flame, and the gasp of the dying Magian's 
voice was drowned by the call of the Muezzin to prayer on the 
top of the minaretted mosque. 

In a way the Moslem creed was easy of acceptance for Persia, 

since Mohammed himself had adopted elements from Zoroas- 

trianism to unite with Jewish and Christian tenets in making 

up his religion. The Persian, therefore, under show of reason 

2 A 353 



354 THE ZOKOASTRIANS OF YEZD 

or exercise of force, could be led to exchange Ormazd for Allah, 
to acknowledge Mohammed, instead of Zoroaster, as the true 
prophet of later days, and to accept the Koran as the inspired 
word of God that supplanted the Avesta. The conqueror's 
sword, inscribed with holy texts in arabesques, contributed its 
share, no doubt, to making all this possible, but many a Gabar 
stubbornly refused to give up his belief, and consequently 
sealed his faith with his blood. The few that sought religious 
liberty by accepting exile in India became the ancestors of the 
modern Parsis of Bombay, so often spoken of already; but 
the rest of the scanty handful that escaped the perils of the 
Mohammedan conquest found a desert-home at Yezd and in the 
remote city of Kerman, not to mention the straggling few that 
are found elsewhere in Persia, to prove the exception to the now 
universal rule of Islam in Iran. 

Almost immediately after my arrival at Yezd I inquired for 
the home of Kalantar Dinyar Bahram, the head of the Zoroas- 
trian community, which numbers between 8000 and 8500 in 
the city and its environs, 1 but it took me some time to find his 
house. For nearly two hours my tired mules and donkeys 
threaded their way through dusty, crooked lanes, across camel- 
filled squares, and in and out of closing bazaars, until we 
reached the Kalantar's door just as the sun was going down. 
The dwelling was unpretentious on the outside, as all Persian 
houses are. Several servants answered the summons of my 
man, who announced the arrival of a farangi, and I was then 
ushered into a large, oblong room carpeted with fine Persian 
rugs. The walls of the apartment were almost without deco- 
ration, and the furnishing was confined chiefly to divans and 
cushions, as in many Oriental dwellings ; but on one side there 
were arranged in Occidental manner a table and some chairs, 
made and upholstered after European models. The front of 

1 These are the figures given me at ration of the Zoroastrians in Persia. 
Teheran by Mr. Ardeshir Reporter, See also p. 336, n. 3, above; p. 376, 
Agent of the Society for the Amelio- n. 1, and p. 425, below. 




SCENE IN YEZD 




THE RESERVOIR IN THE MEIDAN AT YEZD 



A ZOROASTRIAN WELCOME AT YEZD 355 

the room seemed almost open to the air, because of the broad 
doorways and deep windows that ran from floor to ceiling and 
looked out upon a covered veranda and a court which enclosed 
a pretty garden with roses and potted plants. My Gabar host 
entered the room a few minutes later. 

He was a man somewhat over fifty years of age, with a 
roundish face and grizzled beard, and was dressed in a robe of 
grayish cloth with a large white cotton sash about his waist. 
Upon his head he wore the low rolled turban which is character- 
istic of the Persian Zoroastrians ; I had seen the same style of 
head-gear worn by an Iranian priest from Kerman when I was 
in Bombay. With genuine courtesy and manifest cordiality 
my host extended a welcome, and turned aside with a light 
touch the apologies I offered for my dusty appearance and for 
entering his room wearing riding-leggings as one has to do 
often in Persia. In the best Fdrsl phrases that I could com- 
mand I explained the purpose of my visit. In Eastern fashion 
he immediately placed his house and his all at my disposal, 
and this I found to be no empty phrase of courtesy in his case, 
even though I could not accept the generous invitation to 
lodge under his roof, because I had already promised to be the 
guest of the English missionaries. 

As soon as the Kalantar learned in more detail the reason 
for my coming to Yezd, he sent for a member of the community 
named Khodabakhsh Bahram Rais, who had studied in Bombay 
and spoke English fluently, and who was known in Yezd as 
' Master ' because of his attainments. The style of dress of this 
scholar was similar to the Kalantar's, even in the waistband 
and turban, and his features were of the same general cast, 
although somewhat sharper. The nose, as in the case of all 
the Persian Zoroastrians that I met, was rather prominent, 
but well shaped. In manner he was modest and courtly, and 
his face lighted up when he recognized the name he had heard 
from common friends in Bombay, where my Zoroastrian in- 
terests were known. He held a hurried consultation with the 



356 THE ZOROASTRIANS OF YEZD 

Kalantar, and they at once proposed a plan for a conference 
on the morrow with the High Priest and with the spiritual 
and secular leaders of the Zoroastrian community, setting the 
time in Persian fashion at so many hours 'after sunrise.' 
Gifts of flowers were brought in and presented to me as a 
sign of welcome, and the hospitality of supper was extended 
in Zoroastrian style. 

At this meal the host himself declined to take a seat at the 
table, but moved about, standing now at the doorway and 
again withdrawing to give directions, but returning to see 
them carried out. He explained that this was regarded 
among his people as the true manner of hospitality in olden 
times, when the master of the house was supposed to be ever 
ready to serve his guests in person, and he thought that I 
would best like to have the time-honored custom observed. 
The number of dishes was perhaps ancient Median in its 
variety, rather than early Persian in other words, the abun- 
dance of Astyages and not the frugality of his grandson Cyrus, 
if we may accept the picture in Xenophon's Greek romance as 
accurate. A hearty broth as first course was followed by lamb, 
vegetables, and some dishes characteristic of Yezd, with sweet- 
meats and tea for dessert and some mild wine such as ' the house 
of the Magian ' produced in the days of Hafiz. To converse 
at table was, I knew, contrary to the Avestan code, but I 
preferred not to observe this prescription, even in the house 
of a Zoroastrian, as I wished to use every possible moment to 
learn more concerning the interesting people among whom I 
had come. We talked about matters of home life among the 
Zoroastrians, the size of their community, their relations 
with Kerman, and the communication they had with their co- 
religionists in India, until it was time for me to leave for the 
English Mission, where I found a hearty welcome awaiting me. 

At an early hour the next morning I returned again to the 
house of my Zoroastrian host. The Anjuman, or synod of leading 
men in the Gabar community, was assembled to the number of 



MEETING WITH THE ANJUMAN, OE SYNOD 35T 

eighteen. The Chief Priest, Dastar-i Dasturdn, who was named 
Namdar, happened to be absent in India at the time, but the 
Acting High Priest, Tir Andaz, who was his father-in-law, was 
at home and entered the assembly a few minutes later. He 
was a tall, handsome man, dressed in robes of pure white, 
and his flowing beard of snow lent the dignity of age to his 
kindly face. A brownish turban set off his dark, intelligent 
eyes, which had the gleam of youth and were in keeping 
with his manly frame, erect bearing, and clear voice. 

The formal reception in Oriental manner now began, and I 
was reminded of the description in the Zartusht Namah of the 
ceremonies when Zoroaster first appeared before his patron 
Vishtaspa. Settees and chairs were placed in a large open hall 
that faced upon the garden court. They were arranged in the 
form of a widespread V, in much the same manner as in the 
council of Ormazd described in the old Iranian Bundahishn. 1 
I was formally conducted to a seat at the apex of this V. My 
host took the place on the right, the High Priest sat on the 
left ; the other members of the assembly were arranged in order 
of seniority or rank. When all were seated there was a mo- 
ment's pause. Then those sitting on the right turned toward 
me and made a solemn bow, to which I responded ; the same 
salutation was formally repeated on the left. A servant next 
entered with a tray of confectionery, a ewer of rose-water, and 
a hand-mirror. From the hospitality of the Parsis in India, I 
was familiar with the rose-water and sugar candy, but I had 
not previously seen the mirror used in ceremonies, although 
I was told it was an old Zardushtian custom in receiving 
a guest. My momentary embarrassment was relieved when 
the mirror was handed to the High Priest. He looked gravely 
into it, slowly stroked his white beard, on which he poured a 
few drops of rose-water, and then with perfect dignity passed 

1 See my article in Archiv fur Eeli- speaks of this as the Parsi manner of 
gionswissenschaft, 1. 364. I have also sitting at meals, in contrast to the Jew- 
been told that the Talmud somewhere ish fashion. 



358 THE ZOROASTRIANS OF YEZD 

the glass to the next, who did likewise, and so did the others. 
The sugared bonbons, for which the Zoroastrians of Yezd are 
renowned, proved very refreshing and served to satisfy that 
craving for sweets which is felt by travellers in hot and 
dry climates. Meanwhile a number of the company regaled 
themselves with snuff, as there seems to be no objection to the 
use of tobacco in that manner, but only to its being smoked, as 
that is regarded as a defilement of the fire. 

The formalities finished, the real conference began, and for 
three or more hours I asked and answered questions relating 
to Zoroaster and his faith, and concerning the condition of his 
followers in Persia. Two manuscripts of the Avesta and some 
fragments were first shown me. One of these was a fine large 
copy of the Vendidad Sadah, seen by Professor E. G. Browne, 
when he visited Yezd in 1888; the other was a text of the 
Yasna. The copy of the Vendidad Sadah was much the older 
of the two, and was said to date back about three hundred 
years. The Yasna manuscript belonged to the middle of the 
last century. The third text, incomplete, was a good tran- 
script of the Vishtasp Yasht, which is a comparatively late com- 
pilation devoted to the praise of Zoroaster's patron and other 
worthies of the religion. These were all the manuscripts that 
could be produced at the moment, and the best-informed mem- 
bers of the assembly stated that all their more important manu- 
scripts had been sent to India for safe-keeping or for use, and 
they feared that the chances of obtaining hitherto unknown 
copies were growing yearly less. 1 I urged upon them the 
importance of making a careful search, especially among the 
older families, who might possibly have texts that had not 
found their way to Bombay, and I have since corresponded 

1 A number of these manuscripts the manuscripts I had seen at Yezd 
which are now in Bombay had already and also in Mr. John Tyler's posses- 
been used by Professor Geldner in the sion at Teheran, as the Secretary of the 
preparation of his edition of the Avesta. Panchayat had requested from me a 
I communicated afterward to the Parsi report regarding any copies of Avestan 
Panchayat in Bombay the facts about texts I could find. 



MANUSCRIPTS OF THE A VEST A 359 

with them on the subject ; but I am hardly more sanguine 
about the results of the search than was Westergaard, who 
visited Yezd and Kerman in 1843. l The members of the 
assemblage naturally ascribed the loss of their texts largely 
to the persecutions that followed after the Moslem conquest, 
an instance of which I gathered from an oral tradition current 
among them. It is worth repeating. 

About a century and a half after the Arab conquest, or more ac- 
curately in the year A.D. 820, there was a Mohammedan governor 
of Khorasan, named Tahir, who was the founder of the Taharid 
dynasty and was called 'the Ambi-dextrous ' (Zu'l-Yaminein). 
He was a bigoted tyrant, and his fanaticism against the Zoroas- 
trians and their scriptures knew no bounds. A Musulman who 
was originally descended from a Zoroastrian family made an at- 
tempt to reform him and laid before him a copy of the book of 
good counsel, Andarz-i Buzurg-Mihr, named from the precepts 
given by Buzurg-Mihr, the prime minister of Anushirvan the 
Just, and he asked the governor for permission to translate it 
into Arabic for his royal master's edification. 2 Tahir exclaimed, 
4 Do books of the Magians still exist? ' On receiving an affirm- 
ative answer, he issued an edict that every Zoroastrian should 
bring to him a man (about fourteen pounds) of Zoroastrian and 
Parsi books, in order that all these books might be burned, and 
he concluded his mandate with the order that any one who 
disobeyed should be put to death. As my informant added, 
it may well be imagined how many Zoroastrians thus lost their 
lives, and what a number of valuable works were lost to the 
world through this catastrophe. A variation of the story, but 
told of Tahir's son, named Abdullah (A.D. 828-840), and 
applied to the romance of Vdmik and 'Adhrd, which is described 
in its title as 'a pleasing story (khub hikdyat) compiled by 

1 See Westergaard, Zendavesta, 2 This work corresponds to the Pah- 
preface, p. 21, n. 4, and p. 11, n. 3, lavi treatise Pandnamak-i Vazhorg- 
Copenhagen, 1852-1854; see likewise Mitro-i Bukhtakan, which has survived, 
his letter to Dr. Wilson, quoted by See on this point West, Grundr. iran. 
Karaka, History of the Parsis, 1. 60. Philol. 2. 113. 



360 THE ZOROASTRIANS OF YEZD 

sages and dedicated to King Anushirvan ' (A.D. 531-579), is 
given by the Persian biographer Daulatshah in his literary 
notices. 1 The story as it exists to-day among the Zoroastrians 
is an interesting illustration of their pertinacity in keeping up 
the tradition regarding the loss of much of their literature 
after the Mohammedan conquest as well as during the invasion 
of * Alexander the Accursed.' 

Inquiries regarding legends of Zoroaster did not result in 
bringing out anything particularly new, but it was interesting 
to obtain their views on some of the debated questions in con- 
nection with the prophet's life. Zoroaster, they believe, came 
from Rei, the ancient ruined city of Ragha near Teheran, long 
associated with his mother's name. 2 They knew nothing of 
the tradition that connects him with Urumiah. 3 They associ- 
ate his home, or rather his father's house, which is said in the 
Vendidad to have been located on the Drejya, Darejya, or 
Daraj, with the region about the river Karaj on the road from 
Teheran to Kazvin. The village, they said, corresponds to the 
modern Kalak near the Karaj River which flows from the 
mountain Paitizbara, as they interpret the words paiti zbarahi 
in the Avestan text. 4 The resemblance between the letters D 
and JTin Avestan Darejya, Drejya, Phi. Dareji, Pers. Daraj, if 
written in the ancient script, does make this ingenious compari- 
son seem plausible for a moment, especially as the river Karaj 
itself, a photograph of which I took three weeks later when I 

1 See Daulatshah, Tadhkirat ash- and Customs, p. 3, Bombay, 1893, and 
Shu'ara, ed. Browne, p. 30, London, this treatise has been translated into 
1901, and compare Browne, Literary Persian by Master Khodabakhsh. The 
History of Persia, pp. 12, 346-347. same interpretation appeared to be 

2 See p. 430, below, and compare found in a lithographed work from 
my Zoroaster, pp. 17, 85, 192, 202. which they quoted, and which was a 

8 See pp. 87, 103, above. compilation by Mirza Tath-ali-khan 

4 See Vd. 19. 3, 11. The view that Zanganahi (so far as I could catch the 

the text contains an allusion to a moun- name). The comparison of the Daraj 

tain called ' Paitizbara ' (paiti zbarahi") with Karaj is due to this latter writer, 

from which the Darejya flows, is found There are some incidental references 

in an essay in English by Ervad She- to the Karaj in Yakut, pp. 65, 478, 

riarji Bharacha, Zoroastrian Religion 488 ; see also p. 443, below. 




THE ZOROASTRIAN ANJUMAN AT YEZD 



QUESTIONS CONCERNING ZOROASTER 361 

crossed it, also shows precipitous banks that would answer to the 
conditions supposed to be required by the phrase paiti zbarahi 
in the Vendidad ; l but in spite of this the identification seems 
fanciful, and I have given reasons elsewhere for believing that 
the river Darejya, Drejya of the Avesta, is the modern Daryai 
in Azarbaijan. 2 I may add in passing that a number of persons 
in the assembly knew that Zoroaster's name was associated by 
tradition with the city of Balkh in eastern Iran. 

For Zoroaster's name, which appears in the Avesta as Zara- 
thushtra and in Modern Persian as Zartusht or Zardusht, which 
is believed in reality to mean some sort of a camel (Av. ushtra, 
see p. 89, above) they offered nearly a dozen fantastical interpre- 
tations or attempted etymologies. Dastur Tir Andaz, after the 
Oriental manner, suggested that the name, if divided as Zar- 
tusht^ might be explained as ' pure gold,' or ' washed gold,' as if 
the latter element were connected with shustan, 4 to wash.' 
Another member of the company proposed 4 enemy of gold,' as 
if the final member of Zar-dusht were dushman, 'foe.' Finally 
my host turned to the lithographed book that he held in his 
hand, and which seemed to be a compendium of the various 
Persian and Arabic writers who mention Zoroaster and are 
already known to Western scholars. 3 The work contained no 
less than nine different explanations, part of them cited from 
Persian lexicographical works, and I subsequently learned that 
it was handed down by Farzanah Bahrain ibn Farhad, a disciple 
of Azar Keivan, who lived in the time of Akbar the Great, 
about A.D. 1600. 4 The value of the book was most highly 

1 1 have reproduced the photograph shasp was a learned and well-known 

in Chap. 28, below. For paiti zbarahi, Persian priest who believed in a 

see Bartholomae, Air. Wb. p. 1699. universal religion. After spending 

2 See my Zoroaster, pp. 194-195. twenty-eight years of his life in medi- 

3 For the main sources, see my tation he came to India and settled at 
Zoroaster, pp. 280-286. Patna, where he became known as a 

4 According to the Parsi Prakash, teacher of a universal creed. He wrote 
ed. Bamanji Bahramji Patel, p. 10, the Makashifat-i Azar Keivan and died 
Bombay, 1888, the above-mentioned at Patna in 1614, at the age of eighty- 
Dastur Azar Keivan bin Azar Go- five. For this information from the 



362 THE ZOROASTRIANS OF YEZD 

thought of by my best-informed critic, Khodabakhsh Ra'is, 
who referred to it as an imaginative work and branded the 
etymologies as 4 fanciful and invented by the disciples of the 
aforesaid Azar Keivan, who was a half-Brahman, half-Zoroas- 
trian, a believer in metempsychosis.' Scholars will certainly 
agree with the estimate as to the philological value of the inter- 
pretations, but I give the list as I noted it. 

1. afanda-i avval, ' first created being.' 

2. nafs-i kull, ' universal soul/ 

3. nafs-i natikah, ' spirit of speech.' 

4. l akl-i falak-i 'utarid, ' genius of the heaven of the planet 
Mercury.' 

5. nur-i mujarrad, ' incorporeal light/ 

6. 'akl-ifa"al, 'active genius/ 

7. rabbu 'n-nau'-i insan, 'lord of all mankind/ 

8. rast-gu, 'truth-speaker/ 

9. nur-i khuda, or nur-i yazddn, ' light of God/ 

From the same historical compilation the reader cited a passage 
to the effect that the Mohammedans believed that there were 
several Zoroasters a view which I had heard propounded 
also by some of the Zoroastrians in India and that the Zar- 
dusht of Vishtasp's time was the ninth in order, the first of 
them being Hoshang; l but this view, according to my host, 
was due to a mistaken reading of a verse in the Shah Namah. 

From questions relating to Zoroaster we turned to religion 
and philosophy. The discussion led to the problem of dualism, 
the relation of Ormazd (Ahura Mazda, 4 Lord Wisdom') and 
the archangels and angels (Amesha Spentas and Yazatas) to 

Parsi Prakash I am indebted to ray Bharucha, op. cit. p. 121) Zartosht is 
pupil and friend, Ervad ManeckjiNus- the thirteenth in the line of prophets, 
servanji Dhalla, of Karachi, India, a Such is the view held also by some of 
student at Columbia University. For the theosophists among the Modern 
a note on Farzanah Bahrain ibn-Far- Parsis of India, certain of whom re- 
had, see Shehriarji Bharucha, The gard him as the seventh of the name. 
Dasatir, in Zartoshti, 3. 122, Bom- See Bilimoria, Zoroastrianism in the 
bay, 1905. Light of Theosophy, p. 4, note, Bom- 
1 In the Dasatir (see Shehriarji bay, 1896. 



DISCUSSIONS REGARDING THE RELIGION 363 

Ahriman (Angra Mainyu, 4 Evil Spirit, ') and the arch-fiends 
and fiends (Daevas and Drujes), who war against the soul of 
man. I found that the most enlightened of these Zoroastrians 
look upon Ahura Mazda as comprising within himself the 
conflicting powers of good and evil, designated respectively as 
Spenta Mainyu, 'Holy Spirit,' and Angra Mainyu, ' Evil Spirit,' 
and that their views in this respect, and possibly under the 
influence of Bombay, would agree with the monotheistic tenets 
upheld by the Parsis of India to-day, who stoutly deny the 
allegation that Zoroastrianism teaches pure dualism. 1 They 
believe also in the resurrection of the dead, or are acquainted, 
at least, with this doctrine, which their faith has taught since 
early times ; and my informant promptly gave me the tech- 
nical term (Pahlavi ristdkhez, Mod. Pers. ristdkhlz) for the 
'rising of the dead.' The Messianic doctrine of a Saoshyant, 
or Savior, appeared likewise to be well known. 

When hearing the High Priest recite passages from the 
Avesta and when listening to a Mobed as well as a layman read 
from the sacred texts lying before us, I was struck by certain 
peculiarities of pronunciation that are worthy of note. For 
some of the striking features I was prepared through a previ- 
ous study of the variations in the Iranian manuscripts of the 
Avesta, used by Geldner for his great edition of the Avesta, and 
through my observation of the pronunciation of the Parsi 
priests in India ; 2 but some of the peculiarities and cer- 
tain phonetic inconsistencies in reproducing the words were 
quite unexpected. What I noticed most was the fact that the 
Avestan letters th, ph, dh, gh, and generally kh, which are 
presumed historically to have been spirants, as in English kith, 
burthen (for burden), and German hoch, were pronounced as 
ordinary t, d, g, k, or occasionally as aspirates t\ d\ g\ k', (t h , 

1 On the whole subject of dualism, 2 Many of the phonetic features are 

see the views expressed in my article common in the ordinary pronunciation 

in Orundr. iran. Philol. 2.626-631, of the Indian Parsis, except among 

647-649, 663. the trained scholars. 



364 THE ZOROASTRIANS OF YEZD 

d h , g h , k h ') : for example, atha, ' so,' sounded as aid or at* a, 
at h d\ veret h rag h na, 'Victory.' The consonant t was given 
everywhere as d ; for example, cvat, 'as many as,' was pro- 
nounced like c a wad. The secondary nasal nh (nTi) arising in 
Avestan from an original sibilant was pronounced like nk 
(vank-e-osh, vank-hi-osh, or vank-i-ash, for vahhe'ush, 'of good/ 
and ank-i-ush for anhSush, 'of the world'). The voiced sibilant 
z was pronounced like the English z, and the Avestan letter for 
zh could not be distinguished from our/ (or from/,,/*), while 
the previously mentioned th occasionally interchanged with , as 
in the Avestan manuscripts (serish for thrish, ' thrice '), thus 
coming near to the earlier spirant character of the sound th than 
does the pronunciation t or t' in vogue among the priests as in- 
dicated above. The vowels a, 0, w, were frequently confused 
with each other, and i was shaded in the direction of e (yeheshta, 
'best/ for vahbhta), while certain of the diphthongs were 
merged into simple vowels (ao in mraot, 'he spoke/ pronounced 
as u, m a rud). The anaptyctic and epenthetic vowels were 
clearly marked: thus, pa-i~ti, 'against/ 

A few illustrations of the general characteristics of the pro- 
nunciation will suffice. The name of the prophet Zoroaster, 
in the nominative form Zarathushtro, was pronounced as 
Zarat^ushtrU) Zarat^oshtru^ or even Zarat^ashtru. The 
opening lines of the well-known Profession of Faith, naisml 
daevo fravardne mazdayasnd zarathushtrish vidaevo ahura- 
tkaesho, ' I abjure being a Demon-Worshipper, I profess 
myself a Worshipper of Mazda, a foe to the demons, and a 
believer in the faith of Ahura/ were sounded like ' ndismi diva 
fravardne mazdayasnu vldivu ahura-d-klshu.' The sacred 
formula of the Ahuna Vairya sounded on their lips quite dif- 
ferent from the pronunciation generally given to it in the Occi- 
dent, at least as indicated in the accepted philological works. 
This will be clear from a comparative transcript, first in the 
ordinary transliteration with which we are familiar, and then 
in the transliteration reproduced from the memoranda I made 



PRONUNCIATION OF AVESTAN AT YEZD 365 

of the Yezd pronunciation, supplemented by notes from 
Master Khodabakhsh. 

AHUNA VAIRYA STANZA AS ORDINARILY WRITTEN 

yathd ahu vairyo athd ratush ashdtcit hacd 

vanheush dazdd mananho shyaothananam ahMush mazddi 

khshathremcd aliurdi a yim dregubyo daddt vdstdrem. 1 

AHUNA VAIRYA STANZA AS PRONOUNCED AT YEZD 

(with the variant pronunciations in parentheses) 

yatd (yat h d) alii vaireyu atd (at h a) ratosh (ratdsh) ashdd a cld hacd 
vank-e-osh (vanke-hi-osh, vanh-i-ash) dazdd manankahu she-yu-tananume 

anke-hi-osh (arik-i-ash) mazdde 
kashatramcd (khashatremcd) ahorde (ahdrde or ahurde) d yem dare- 

gdbe-yu (dargdbyu) dadad vds-e-tdram (vdwstdrem). 2 

For a fuller collection of material to illustrate the pronun- 
ciation I must refer to a monograph on the subject which I 
hope soon to publish in one of the Oriental journals. 

While on the subject of pronunciation and the reading of the 
sacred texts, I may add an observation which will not, however, 
surprise specialists; I refer to the fact that the Acting High 
Priest and also the more scholarly members of the assembly 
were unaware that a great part of the Younger Avesta is com- 
posed in metre. The idea of verse and verse-structure appeared 
wholly new to them, when I read for them a portion of the Horn 
Yasht metrically in the manner that is familiar to students in 
the West. In all such matters it is manifest that ages of 
persecution and of neglect of their sacred lore have not been 
without a detrimental influence upon their technical knowledge; 
on the other hand, certain points in their pronunciation appear 
to deserve the consideration of linguistic scholars, because the 
Persian Zoroastrians are not affected by any philological bias 
and have remained practically free from the Indian influences 

1 For the sake of parallelism I have 2 The a is sometimes labialized to 

here retained, with trifling modifica- aw (Eng. law}. 
tions, the older transliteration of Justi. 



366 THE ZOROASTRIANS OF YEZD 

that may have affected, in some respects, the pronunciation of the 
Parsis of Bombay. 1 

By this time it was considerably past mid-day, and nearly an 
hour more was spent in examining the manuscripts and in 
photographing specimens of the text. A rare privilege was 
now accorded me ; I was invited by Tir Andaz to visit his fire- 
temple early that afternoon after I had enjoyed the repast 
spread by our host. I was glad to accept at once this oppor- 
tunity to become acquainted with a place of worship used by 
the Persian Zoroastrians. It was the temple of the Atash-i 
Varahrdn, or Atash Bahrdm, ' Fire of Victory,' situated in the 
Parsi quarter and located next to the house of Dastur Namdar, 
the priest who was absent in India at the moment. It is the 
chief Zoroastrian sanctuary of Yezd, although there are three 
other fire-shrines or chapels, designated either as Dar-i Mihr or 
Adaridn, besides one such minor place of worship in every 
Zoroastriau village in the vicinity of the city. 2 

Upon reaching the temple I found it to be a simple, unpre- 
tentious building. From its exterior and from the entrance it 
would hardly have been possible to recognize it as a temple at 
all. Mohammedanism allows no rivals to its beautiful mosques 
with turquoise domes, arabesque arches, and slender tessellated 
minarets. The splendor of the ancient temple of Anaitis at 
Ecbatana, from which, as I have described above, conquerors 
carried off untold wealth in gold and silver plate, the grand 
ruins of Kangavar and the gorgeous display at the Shrine of 
Fire in Shiz, under the Sasanian kings, belong to ages long 
since dead. 3 

Before reaching the main room of the sanctuary at Yezd it 

1 It is only the younger generation 2 The name Dar-i Mihr, ' Shrine of 

of Zoroastrian students at Yezd that Mihr' (used also in India) contains a 

has coine into close contact with the reminiscence of the ancient Mithraic 

Zoroastrians of India, through the in- worship, but is now used (like Adar- 

fluence of Master Khodabakhsh and a idn, ' pyraea') merely as a designation 

few other scholars who have been in for a small chapel or shrine of fire. 

Bombay. s See pp. 131-143, above. 



A VISIT TO THE FIRE-TEMPLE 367 

was necessary to pass through several corridors and an ante- 
chamber, all of which help to render the shrine safer from 
desecration. On one side of the last passageway I observed a 
pile of short logs, one or two feet long and several inches thick, 
that were used as fuel for the holy flame ; x it appeared to be 'well- 
dried and well-examined wood,' as the Avesta enjoins. 2 From 
the anteroom I entered the large oblong chamber, or chapel, 
adjoining the sanctum sanctorum in which the fire was kept. 
My ear caught at once the voice of the white-robed priests who 
were chanting in the presence of the sacred element a hymn of 
praise sung by Zoroaster of old. It was a glorification of 
Verethraghna, the Angel of Victory, in the Bahram Yasht, and 
I felt a thrill as I heard the Avestan verses verethraghnem 
ahuradhdtem yazamaide, 4 we worship the Angel of Victory, 
created by Ahura' ring out from behind the walled recess 
where the fire was hidden. The door was open and I stood 
within a few feet of the fire, so as to listen, but I made no 
attempt to see the flame, as I knew such a step would be 
regarded as a profanation and might bar the way to other privi- 
leges which I wished to enjoy. It seemed an unusual experience 
thus to be standing in a fire-temple in Zoroaster's own land 
and listening to the priests of his hereditary line chanting verses 
from the sacred texts as had been done for nearly three thousand 
years. The voice of the zot, or officiating priest, was high, nasal, 
and resonant, and his intonation was so rapid that he had to 
pause at times to catch his breath ; while his assistant, the raspl, 
chanted in a lower key or accompanied his recitation in a nasal 
minor key with great rapidity of utterance. 3 Each of the cele- 
brants wore over his mouth the paitiddna, a small white veil 
prescribed by the Avesta to be worn over the lips when before 



1 Cf. Vd. 3. 1. heard in Bombay and Udvada, and I 

2 Cf. Vd. 14. 2 ; 18. 27, 71. observed the same peculiarities in pro- 

3 The intonation of both the priests nunciation that I had observed in the 
was loud and resonant and more swift conference of the forenoon. 

than that of the Parsi dasturs I had 



368 THE ZOROASTRIANS OF YEZD 

the fire, in order to prevent the breath and spittle from defiling 
the hallowed flame. 

I almost fell into a revery as I listened to this monotonous 
chanting of the Yasht ; but the hymn was soon ended, and the 
veiled priests came out from the presence of the fire and were 
kind enough to allow me to take their photograph, although 
the light was too dim to secure a good picture. 

While speaking of pictures I may mention a so-called por- 
trait of Zoroaster hanging on the wall of this main chamber. 
I had heard of it a number of years before, and when writing 
my book on the Prophet of Ancient Iran I had expressed a 
keen desire to see it. 1 My conference with Tir Andaz in the 
forenoon, when he gave me the meagre information that he had 
about its possible remote connection with Balkh, had prepared 
me for disappointment as to its value, but I did not expect to 
find it of so little importance. The picture is merely a modern 
colored print, apparently a cheap Parsi chromolithograph from 
India, perhaps not twenty years old, and of no historic interest. 
It is a variety of the familiar representation based on the Tak-i 
Bostan sculpture ; 2 but the staff is not fluted, as in the sculp- 
ture, the top is capped with a symbolic flame, as in other modern 
representations current among the Parsis in India, and the lower 
end of the staff rests upon the ground. This colored picture was 
the only decoration I noticed on the bare, whitewashed walls. 

At the rear of the chamber there was a gallery used on 
occasions when a considerable number of the Zardushtian com- 
munity come together, as at the Gahanbar season, the Farvadin 
festival, on some commemorative day, or at some special cele- 
bration. The gatherings on such occasions are the nearest 
approach that the Zoroastrians have to the assembling of a 
congregation in church, for they have nothing that corresponds 
precisely to our general Sunday worship. 

The Acting High Priest now opened a door leading into a 
small side-chamber to the right of the sanctum where the fire 

1 See my Zoroaster, pp. 288-289. 2 See pp. 216-218, above. 



THE ZOROASTRIAN RITUAL 369 

was kept. It was a room arranged as an Izashnah G-dh, a place 
set apart for the performance of religious ceremonies and 
priestly rites. The floor was built of stone and was cemented 
and marked off into little channels (pdvi) or grooves (hash), 
to enclose the space within which the priest sat while conduct- 
ing the ritual, as I had witnessed in the halls adjoining the 
Parsi temples in India. 1 A lambskin, used apparently as a 
seat, was lying on the floor, and there were small, low stone 
stools such as are generally employed in the Izashnah Gah, be- 
sides a number of sacrificial utensils. Among the latter were 
the cups for holding consecrated water, milk, and the juice of 
the ^Jra-plant (Av. haoma), from which the sacred drink was 
prepared in ancient times, as nowadays, and partaken of by the 
priest as a part of the ceremony. 

The haoma, as is well known, corresponds to the soma of 
Vedic India, which grows on the mountains, 2 and the two 
branches which the priest gave me came from the mountain 
heights some distance from Yezd. In addition to this and the 
urvard hadhdnaepatd, or pomegranate, 3 there is still another plant 
employed in the sacrifice, and it has been used in the Magian 
ritual since time immemorial. It is the barsom (Av. baresmari), 
the twigs or sprays of which are tied in a bundle at a certain 
point in the sacrifice, corresponding in a distant manner to the 
barhis, or straw, strewn as a seat for the divinities in the Vedic 
ceremonies of old. In Yezd the tamarisk bush is used to form 
this bundle, and it is bound with a slender strip of bark from 

1 1 refer to the so-called urvls-gdh 3d ed., pp. 392-409 ; cf. likewise my 

connected with the fire-temples at note in JAOS. 22. 321. 
Udvada, Navsari, and Bombay. For 2 g ee YS. 10. 3, and Rig Veda 5. 85. 

a photograph and a description of the 2 ; 10. 34.1. 

latter, together with a representation 3 The Zoroastrians of Yezd, like the 

of the various implements and uten- Indian Parsis, agree in regarding the 

sils employed in the sacrifice, see pomegranate as the representative of 

Darmesteter, Le ZA. 1. in trod. p. 72 the Avestan urvard hadhdnaepatd ; on 

(pi. 4), and compare the interesting the latter, compare Haug, Essays on 

notes descriptive of some Parsi cere- the Parsis, pp. 251, 399, and West, 

monies, by Haug, Essays on the Parsis, SBE. 37. 186. 

2B 



370 THE ZOROASTRIANS OF YEZD 

the mulberry tree, probably in exactly the same manner as it 
was in Zoroaster's day. 1 Brass rods are sometimes substituted 
for the twigs, as is done by the Parsis in India, but at Yezd 
this substitution is made only in winter, when it is impossible 
to procure the branches, or at some particular time when it is 
impracticable to obtain them. It was the use of these very 
branches, perhaps, that the Prophet Ezekiel denounced as an 
abomination to God when he saw in a vision 'about five and 
twenty men, with their backs toward the temple of the Lord, 
and their faces toward the east, and they worshipped the sun 
toward the east, . . . and, lo, they put the branch to their nose.' 2 

I saw the large tamarisk bush from which the sprays were 
cut for use in the barsom ceremony ; it was of a light green 
color, twelve or fifteen feet high, and stood in the garden 
adjoining the rear of the temple. A high wall shut in the 
garden at the back; a gallery ran part of the way around 
the enclosure ; a flight of steep steps led down from this to the 
ground, where there were blossoming rose bushes, sweet-scented 
shrubs and plants, a pomegranate tree, and the tamarisk bush. 
Tir Andaz cut off from this three handsome sprigs, each nearly 
two feet long, and presented them to me. They were slender 
and delicate, covered with downy fibrous leaves, and look 
graceful even in the dried form in which I now have them. 3 

Besides the sacred plants, perfumes (baodhi), bread-offerings 
(draonali, myazda), consecrated water, the haoma^ and milk, the 
Avesta frequently refers to the cow (g&o) in connection with 
the Yasna ceremony. Like their Parsi brethren in India, the 
Zoroastrians of Persia interpret the Avestan words gao jlvya, 

1 The Avestan words employed in 2 Ezekiel 8. 16, 17. 

connection with the bardsman indicate 8 My friend Mr. Percy Bodenstab, 

that the twigs were originally spread of Yonkers, has made a drawing of 

(star-, frastzrdta-}, then gathered into the sprays (here reproduced) in a re- 

a bundle and bound (yah-, aiwydsta-, duced size ; to convey a clearer idea it 

aiwy&whana-} ; see the references un- would be necessary to reclothe the 

der each of these words in Bartholo- branches with the softest green color 

mae, Air. Wb. pp. 98, 947, 1290, 1595. imaginable. 



\\ . 










SPRAYS OF THE BARSOM PLANT 



SACRIFICIAL CEREMONIES 371 

lit. ' living cow,' as goat's milk (Pers. shir), and similarly em- 
ploy an egg and melted butter to represent the gao hudhdh, 
lit. 'beneficent cow,' in the ceremony. The faithful of 
both communities agree in regarding the true Zoroastrian sacri- 
fice to be a bloodless sacrifice, an offering of 'good thoughts, 
good words, good deeds,' accompanied by praise and thanks- 
giving, with appropriate ceremonies. Such was the sacrifice 
offered by Zoroaster himself in the Yashts, after the manner of 
Ahura Mazda, 1 although the A vesta does allude to the sacrifice 
of animals, once, for example, in the Yasna, and several times in 
the Yashts, which represent Vishtaspa and the heroes of old as 
sacrificing thousands of animals, some of which must have been 
slain as a blood- offering. 2 

A possible survival of the ancient custom of animal sacrifice 
may survive at Yezd, down to the present, in the celebration of 
the Jashn-i Mihrgan, ' Sacrifice to Mithra,' although the views on 
this subject may differ. 3 This festival falls on the day of 
Mihr, in the month of Mihr (February-March), and is an impor- 
tant one among the Persian Zoroastrians, as they prolong it for 
five days, till the day of Bahram, or Verethraghna. According 
to the account I received, it commemorates the victory gained 
by Feridun (Avestan Thraetaona) over the Babylonian tyrant 
Zohak (Avestan Azhi Dahdka), whose cruel rule oppressed Iran 
for a thousand years. ' The Persian Zoroastrians used to be- 
lieve, and some of them still believe,' as my authority informed 

1 See Yt. 5. 17, 104 ; 9. 25 ; 17. 44 who, it should be noticed, attributes the 
(rendering gava each time as 'milk'). origin of the custom to Mohammedan 

2 See Ys. 11. 4 ; Yt. 5. 21, 25, 33, influence after the Arab conquest, like 
108 ; Yt. 9. 25 ; compare also the de- the sacrifices at the feast l ld-i kur- 
scription of the Magian sacrifice given 6aw, referred to above, p. 162, n. 1. 
by Herodotus, History, 1. 132. Ob- The opinion of the Parsis in India 
serve likewise that on the eve of battle would also be in favor of his view. See 
(Yt. 5. 68) Jamaspa himself offers an Modi, Meher ane Jashne Meherangan 
animal sacrifice. (Mithra and the Feast of Mithras), 

8 The notes which I present on the Bombay, 1889 ; cf. also Marquart, Un- 
Jashn-i Mihrgan are given on the au- tersuchungen zur GeschichtevonEran, 
thority of Khodabakhsh Bahram Rai's, 2. 132-136, Leipzig, 1905. 



372 THE ZOROASTRIANS OF YEZD 

me, ' that at this festival Feridun sacrificed sheep and bade his 
subjects to follow his example in this respect, and to eat, drink, 
and be merry, because of the overthrow of their arch-enemy. 
It was accounted meritorious, therefore, to celebrate the occa- 
sion joyfully and to sacrifice a sheep or a goat in every house, 
or, if the family were poor, to kill a chicken. The priests 
themselves at first used to kill the animals, but the people after- 
ward did this at home, sprinkling some of the blood on the 
door-posts and over the lintel, and cooking the rest of the 
blood with suet and onions, as a dish to be eaten with unleav- 
ened bread. 1 Since it was regarded not merely as a sacrifice but 
as a burnt-offering unto Mihr-i Irdn-ddvar, ' Mithra, Judge of 
Iran,' the flesh of the sheep and goats, when roasted, was car- 
ried to the fire-temple, prayers were said over it by the priests, 
to whom a share of the flesh was given, a portion was set 
aside for distribution among the poor, and the remainder 
was taken home to be eaten by the family and their friends.' 
Such is the account I received from my informant, who added, 
* this custom is now dying out ; the people are becoming wiser 
and saner, and outgrowing this cruel practice and bloody rite, 
which the Parsis of India do not recognize and like which they 
have nothing.' 

After leaving the fire-temple I asked if I might visit the 
Barashnum Gah, a place set apart for the performance of the ab- 
lution for nine nights, as I shall describe in the next chapter. 
Since it was situated in another street I had an opportunity, 
both when going and returning, to see more of the Parsi quar- 
ter of the town and make further observations as to the com- 
munity and its general condition. As there are about eight 
thousand Gabars in Yezd, they occupy a not inconsiderable sec- 
tion of the city. It is known as the Mahallah-i Pusht-i Khdn-i 
All, or Mahallah-i Pusht-i Khdnah-i All, 'the Quarter in the 
Bear of Khan Ali, or of Ali's House,' and I subsequently learned 

1 It is interesting to note the resem- custom and the observances of the 
"blance between this old-time Persian Jewish Passover. 



THE GABAR QUARTER OF YEZD 373 

that they have a tradition current among them as to the origin 
of this name. The common belief is that the designation by 
Ali's name is due to a device resorted to by the worship- 
pers of Mazda in order to escape persecution at the hands of 
their Mohammedan enemies after the Arab conquest. They 
pretended, it is said, that AH, the cousin and son-in-law of 
Mohammed, had a house in this part of Yezd and that he set- 
tled the Zoroastrians here, in order to shield them from perse- 
cution; and that they were Ali's cowherds. In support of this 
claim they cleverly urged the plea that the name G-abr-dn, 'Infi- 
dels,' by which they were stigmatized, and the modern pronun- 
ciation of which among the Parsis of Yezd would be Gravr-iin or 
G-avr-dn, really meant G-av-rdn, ' cow-keeper,' and that as Gabars 
the Zoroastrians were therefore worthy of Moslem protection. 
As I know through Khodabakhsh Bahram Ra'is, the better edu- 
cated among them regard this explanation of the name of the quar- 
ter as a mere fiction, a piece of popular etymology, and they 
suggest a more probable interpretation. The name Ali, they say, 
is not an uncommon one among the Persians, and this was prob- 
ably the name of a land-owner, or wealthy Khan (Pers. Khdri), 
who had a caravan sarai outside the olcf part of Yezd, near 
where the Parsi quarter now is, and the Zoroastrians settled 
there, 4 back of Khdn Ali ' (not khdnah, 4 house ' ), so that 
the designation has nothing to do with the house of Ali, the 
successor of Mohammed. 

Some details regarding the general condition of the so-called 
Gabars in Yezd and its environs may be of interest. A large 
proportion of the Zoroastrians who live outside of the city itself, 
especially in the neighborhood of the flourishing town of Taft, 
are occupied in gardening and the cultivation of the soil. Ac- 
cording to the Avesta, as I have already stated, 1 agriculture is 
one of the noblest of all employments, because he who sows 
grain, sows righteousness, and one of the most joyous spots on 
earth is the place where one of the faithful sows grain and 

1 See p. 246, above. 



374 THE ZOROASTRIANS OF YEZD 

grass and fruit-bearing trees, or where he waters ground that 
is too dry and dries ground that is too wet. 1 

The Zoroastrians who dwell within the city are largely oc- 
cupied in trading. 2 This privilege was not accorded them until 
about fifty years ago, and they are even now subject to certain 
restrictions and exactions to which no Mohammedan would be 
liable. They are not allowed, for instance, to sell food in the ba- 
zaars, inasmuch as that would be an abomination in the eyes of 
the Moslems, who regard them as unbelievers and therefore 
unclean. Until 1882 they were oppressed by the jazla tax, a 
poll tax imposed upon them as non-believers, and this gave an 
opportunity for grinding them down by extortionate assessments 
and trading- tolls. The jazia was finally repealed by Shah 
Nasr ad-Din, who issued a firman to that effect, September 27, 
1882. It was largely owing to influences brought to bear upon 
him by the Parsis of Bombay that the Shah was led to make 
this liberal-minded move. They worked through the agency of 
the Society for the Amelioration of the Zoroastrians in Persia, 
which they had founded with an endowed fund in 1854, send- 
ing at the same time a representative to Iran to look after the 
interests of their co-religionists. 3 Up to the time of the Shah's 
firman, a Zoroastrian was not allowed to build an upper story 
on his house, or, in fact, erect a dwelling whose height ex- 
ceeded the upstretched arm of a Musulman when standing on 
the ground. 4 Even within a year after the firman was issued, 
a Zoroastrian in one of the neighboring villages is said to have 
had to flee for his life because he had ventured to go beyond 
the traditional limits and add an upper room to his abode, 



1 See Vd. 3. 31 and Vd. 3. 4. 8 For an account of the efforts for 

2 The Zoroastrians in general appear the abolition of this tax, see Dosabhai 
to have an especial aptitude for busi- Framji Karaka, History of the Parsis, 
ness, and they appear rather to accept 1. 72-82, London, 1884 ; cf. also p. 397, 
than reject the designation Jews of below. 

the East ' that is sometimes applied to 4 The comparative scarcity of upper 

them because of their commercial stories on the houses in the Gabar 

activity. quarter is still noticeable. 






I 

3 g 



RESTRICTIONS ON THE ZOROASTRIANS 375 

and another Gabar, who was mistaken for him, was killed by 
the enraged Musulmans. 1 

As regards their dress, moreover, the Zoroastrians have al- 
ways been obliged to adopt a style that would distinguish them 
from the Mohammedans, and it is only within the last ten years 
that they could wear any color except yellow, gray, or brown, 
and the wearing of white stockings was long interdicted. The 
use of spectacles and eye-glasses, and the privilege of carrying an 
umbrella, have been allowed only within the same decade, and 
even now the Gabars are not permitted to ride in the streets or 
to make use of the public baths (hamdiri):, but the latter pro- 
hibition, as they told me, is no longer a hardship, because they 
have built a bathing-establishment for their own use. A score 
of petty annoyances that they have to undergo might be 
cited in addition to the more serious disqualifications; but 
enough have been given to show the disadvantages under which 
they labor and the persecutions to which they are exposed. 

In 1898 the present Shah, Muzaffar ad-Din, sought to relieve 
their condition further by issuing a firman revoking the formal 
disabilities from which they suffered. While imperfectly ob- 
served, this decree has contributed, in spirit at least, to bet- 
tering their position. The spread of Babist doctrines, which 
favor religious liberty and toleration, has possibly contributed 
also by lessening intolerance on the part of the Mohammedans. 
The presence of Europeans has likewise had a salutary effect 
and aided considerably in the general advance. But the most 
has been done by the Bombay Society for the Amelioration of 
the Zoroastrians in Persia, whose funds have helped the Gabars 
and whose reform measures have tended to their general good, 
so that their numbers have increased considerably within the 

1 For this point and the next, see able to incorporate one or two refer- 
Malcolm, Five Years in a Persian ences, and I would recommend to the 
Town, pp. 46, 49, London and New reader's attention Mr. Malcolm's re- 
York, 1905. This interesting book on marks on the restrictions in general 
life at Yezd appeared after the present upon the Gabars (pp. 44-53). 
chapter was written, but I have been 



376 THE ZOROASTRIANS OF YEZD 

last fifty years. 1 Nevertheless, they still do not feel themselves 
free from oppression, and they constantly have to avoid trouble 
and persecution by yielding to Moslem prejudice. In fact, their 
lives are in danger whenever the fanatical spirit of Islam 
breaks out, as was the case about a month after I was in Yezd. 
A general Musulman rising then took place against the Babis, 
a large number of whom belonging to the Behai branch 
are found at Yezd. These Babis were massacred by scores, 
and even hundreds, or were subjected to shocking outrages 
and cruel indignities. The Zoroastrians feared that they 
would suffer the same fate, and I was informed on the authority 
of one who had witnessed the horrors that such might have 
been the case if the fanatical wave had not been broken in its 
course by the prompt and energetic intervention of the Euro- 
peans in telegraphic communication with the authorities in 
power at Teheran. 

The organization of the Zoroastrian community at Yezd has 
already been indicated in a general way. The spiritual guid- 
ance is in the hands of the priesthood (dasturs, mobeds, and 
herbeds), but the authority which they exercise is greatly lim- 
ited by the fact that those who do not wish for any reason to 
accept it can simply throw it off and act in accordance with 
the rule of the Moslems around them. 2 In civic matters the 
community is under the leadership of a synod, the Anjuman 
(Av. hanjamana, 4 assembly, convention'), headed by a kaldntar, 
or mayor, the present incumbent of that office being Kalantar 
Dinyar Bahram, whose hospitality I have described, and whose 



1 In 1854 the number of Zoroastrians Ardeshir Reporter, Secretary of the 

in the vicinity of Yezd was given at Society for the Amelioration of the 

6658 souls (Karaka, History of the Zoroastrians). 

Parsis, 1. 55) ; in 1882 as about 6483 2 For the relations between the spir- 

(Houtum-Schindler, Die Parsen in itual and temporal powers in ancient 

Persien, in ZDMG. 26. 54) ; in 1903 times, see Wilhelm, Kingship and 

as between 8000 and 8500, including Priesthood in Ancient Eran, pp. 1-21, 

the environs of Yezd (these last figures Bombay, 1892 (translated from his Ger- 

being given to me in Teheran by Mr. man treatise in ZDMG. 40. 102-110). 



ORGANIZATION OF THE ZOROASTRIAN COMMUNITY 377 

official duties often take him to Kerman, Anar, and other 
towns in this region where there are Zoroastrians. 

With the Kalantar's young son Bahram I formed a friend- 
ship in the short time of my stay, for he acted as my guide 
round the city and through the mazes of the bazaar. He was 
a bright, intelligent fellow, straightforward and honest, manly 
in his bearing, and agreeable in his manners. I could picture 
from him what might have been the type of youth in Zoroaster's 
day, since the blood of the ancient faith flowed in his veins by 
direct descent. I liked his naturalness and lack of affectation, 
and certain of his characteristics were charmingly naive, for 
when I took his photograph he instinctively plucked a rose to 
hold in his hand (for a true Persian portrait would be artis- 
tically incomplete without a rose), and in the other hand he 
held up to view his European watch. I could understand his 
pride in this respect, since a Zoroastrian would not have been 
allowed some years ago to carry a watch or even to wear a ring. 

Benevolence is a Zoroastrian characteristic, and the Avesta 
inculcates the virtue of generosity. Many of the Parsis of 
Yezd live up to this doctrine so far as their limited means will 
allow. As an instance of this I may cite the following 
example. When the English Christian Mission at Yezd was 
in need of quarters for its hospital a branch of their work 
with which the Parsis especially sympathized a prosperous 
Gabar merchant, named Gudarz Mihrban, came forward and 
donated to the cause a large caravansarai and its property, 
including a house that adjoined it. The structure of this erst- 
while halting-place for caravans lent itself in a remarkable 
manner to the uses to which it was now to be put : the central 
court that once was filled with camels, asses, and pack-mules 
was turned into a pretty garden ; and the old-time lodgings 
of the camel-drivers and muleteers were transformed into 
chambers and wards for the Good Samaritan work. 



CHAPTER XXIV 
ZOROASTRIAN RELIGIOUS CUSTOMS AT YEZD 

* Ev'n them who kept thy truth so pure of old.' 

MILTON, Sonnet on the Massacre in Piemont. 

IN the preceding chapter I have touched upon the religious 
rites of the Zoroastrians as connected with the temple ; in the 
present chapter I shall speak of the religious aspect of their 
home life, the ceremonies and customs observed in connection 
with birth, bringing-up, marriage, and death. 1 

In respect to ceremonies connected with birth, the Iranian 
Parsis differ from their co-religionists in Bombay chiefly 
in the scarcity of such observances. A few Zoroastrians in 
Persia, like their Parsi brethren in India, call in the services 
of an astrologer (nujumi) on the occasion of a birth in their 
family, and the wiseacre makes up the horoscope of the infant's 
nativity. The custom, if resorted to, has the advantage at 
least of preserving a record of the child's day of birth, for most 
Zoroastrians, as I was informed, do not know the age of their 
children, or even the day, month, and year of their own birth. 
As a rule the astrologer is a Mohammedan ; a Parsi astrologer 
is far to seek, and if found, he practises his calling on the lines 
of the Moslem astrologer, 2 but with less skill and cunning in 
making his superstitious forecast. Charms and amulets are 
commonly worn by children, and it is probable that this custom 
is very ancient, antedating Islam by ages. 3 In order to avert 

1 1 have been able to add largely to the 2 The same is true in India, where the 
notes taken at the time through corre- method adopted is that of the Hindu 
spondence with Master Khodabakhsh astrologer (joshi). 
Bahram Ra'is, whose kindness I wish 8 For the usage of talismans in early 
again to acknowledge. times, including the Avesta (Yt. 

378 



YOUTH AND EDUCATION 379 

the influence of the evil eye or to cure a child of some disease, 
a parent will occasionally hire the mobeds, or Zoroastrian priests, 
to read from the Yasna, the Yashts, or the Khordah Avesta ; 1 
and when women are childless, they will sometimes pay to have 
the Vendidad Sadah recited by the priests, in order that the 
curse of sterility may be removed. With reference to birthday 
anniversaries, it is needless to say that they are not observed 
with any particular attention, for the obvious reason already 
given. This is in contrast to what must have been the custom 
in ancient days, for Herodotus says : 4 The Persians honor their 
birthday above all other days, and on this day they prepare a 
feast more abundant than usual ; the rich serve up on such 
an occasion an ox, a horse, a camel, and an ass, roasted 
whole in ovens, and the poor set on the table smaller cattle.' 2 
The custom now is practically forgotten, although a few 
Zoroastrians who have been in Bombay, where birthdays are 
festively celebrated, do observe the natal anniversary (sdl-giriTi) 
of their children by dressing the child in new clothes, prepar- 
ing savory dishes, and inviting their friends and relatives to 
the feast. 

The present education of the Zoroastrian youth is meagre as 
compared with what it might be, but all such statements are 
relative, of course, and depend upon the standard that is used. 3 
Until half a century ago the Zoroastrians were even worse off 
in the matter of education than the rest of the Persians, because 
they either had no schools or were not allowed to have them. 
Happily in the year 1857 certain moneys from the Persian 
Zoroastrian Amelioration Fund were devoted to establishing 

14.34-40), see Jivanji Modi, Charms 1-19) is regarded by the Parsis in 

or Amulets for Some Diseases of the India as especially efficacious in this 

Eye, in the Journal of the Anthropo- respect. 

logical Society of Bombay, 3. 338-340 2 Herodotus, History, 1. 133. 

(1894), and Nirang-i Jashan-i Burzi- 3 For the status of education in pre- 

garan, ibid. 5. 398-405(1900), and An Mohammedan times see Modi, Edu- 

Avesta Amulet, ibid. 6. 418-425 (July cation among the Ancient Iranians 

and October, 1900). (reprinted from The Parsi, vol. 1, 

lr The Ardabahisht Yasht (Yt. 3. nos. 2-9), Bombay, 1905. 



380 ZOROASTRIAN RELIGIOUS CUSTOMS AT YEZD 

and maintaining Parsi schools in the districts of Yezd and Ker- 
man. 1 This progressive move was brought about by the energy 
of the Bombay Parsis and has been carried on with varying 
success ever since. It is to be hoped that even more ample 
funds will be subscribed for this purpose and still greater edu- 
cational facilities afforded the Zoroastrians in Iran. The value 
which such increased advantages would have for them is recog- 
nized by those among their number who have studied in Bombay, 
and it is an encouraging sign to note that members of the 
priestly class are now going to India for advanced education 
and bringing back the seed of knowledge to sow once again in 
the Zoroastrian soil which originally produced it. 

When the Zoroastrian boy or girl passes from childhood into 
the period of full youth, it is customary for each to assume per- 
sonal responsibility in religious matters and be initiated into 
the community of the faithful. This confirmation of faith, if 
I may so call it, takes place between the ages of seven and 
fifteen. In contrast to the formal ceremonies performed by the 
Indian Parsis when the young novitiate puts on the sacred 
thread and shirt, there is practically no religious ritual per- 
formed in Persia. 2 The Iranian boy or girl simply puts on the 

1 See Karaka, History of the Parsis, anabdata) . The wearing of the shirt 
1. 83-89, and Malcolm, Five Years in is also alluded to (by implication) in 
a Persian Town, p. 47, and for a favor- the old Parsi Patits, or formulas of 
able estimate of the Parsi school at confession, see for example Patit Adar- 
Yezd compare Landor, Across Coveted bat 19, in Spiegel, Avesta Ubersetzt, 3. 
Lands, 1. 388-389. 213, Leipzig, 1863 = tr. Bleeck, p. 157, 

2 The modefn name for the sacred London, 1864. For the usage of the 
cincture, or thread, is kosti, kusti, or shirt and girdle in India and the 
kushtl ; the Avestan designation was ceremonies connected with assuming 
aiwydnhana, 'girdle' (Yt. 1. 17; Ys. them, see Dastur Jamaspji Minoche- 
9. 26 ; Vijirkart-i Denig, 12, 18, 20 ; herji, Navjot Ceremony, Bombay, 1887 ; 
Nirangistan, 95). The consecrated Modi, Eeligious System of the Parsis, 
shirt is now called sudrah, sedrah, or in Parliament of Religions, 2. 912, 
sadarah ; its Avestan designation is Chicago, 1893 ; Sheriarji Bharucha, 
not known, but it is presumed to cor- Zoroastrian Religion and Customs, 
respond to vastra, 'garment,' and it pp. 35-36, Bombay, 1893; Darme- 
is alluded to, together with the kusti, steter, Le ZA. 2. 243, n. 13 ; 251, n. 64. 
in Nir. 85-96 ; Vd. 18. 54 (anaiwydsta, 



INITIATION INTO THE FAITH 381 

kushtl, kusti, or kosti, 1 the sacred thread of the religion, as 
soon as he or she can recite the so-called 4 Four Avestas'; that 
is, the tfrosh Bdj, 'prayer to the angel Sraosha,' Kushti-bastan, 
Hying the thread,' Pa Ndm-i Stdyishn, lit. 4 in the name of 
praise,' and Birasdd, 4 may it come.' 2 No priest is invited to 
the house to conduct this rite, as in India, nor are the Zoroas- 
trian Scriptures recited. Instead of that, the kusti is merely put 
on at home without formality, although occasionally the youth 
goes with the kusti in hand to the house of the person who has 
taught him the four Avestan formulas, and there puts on the 
thread in his presence, making, at the same time, a gift of a sugar 
loaf to his preceptor. The wearing of the consecrated shirt, 
sudrah, sedrah, or sadarah (as the word is variously written), 
and formal investiture, which is scrupulously observed in Bom- 
bay, as prescribed by the religion, is not common at Yezd; but 
some of the Indianized Zoroastrians in Persia imitate their 
Parsi brethren of Bombay in keeping up this orthodox 
custom. 

After assuming religious responsibility by putting on the 
kusti, the person is qualified for receiving the rites of the faith, 
and should, in theory at least, be subject to its regulations, 
especially to the rules of purification that are inculcated by the 
A vesta. Throughout the religion, Zoroaster enjoins purity of 
body as well as of soul, and the Avesta prescribes an elaborate 
set of lustrations and ablutions to remove any defilement that 
may have been incurred by contact with unclean matter. The 
greatest pollution comes from touching anything that is dead, 
since death is Ahriman's greatest triumph over Ormazd's 
creation. For that reason the code of the Vendidad gives 
elaborate rules for an * Ablution of Nine Nights ' (barashnum 
nu-shaba, no-shva), to be gone through with the most scrupu- 

1 In India kusti is the current pro- Darmesteter, Le ZA. 2. 685-688 ; 
nunciation, and I have adopted that Spiegel, Avesta Ubersetzt, 3. 4-7, Leip- 
form. zig, 1863 = tr. Bleeck, 3. 4-5, London, 

2 For a translation of the first two 1864. 
(Nlrang Kusti and Srosh Bdj), see 



382 ZOROASTRIAN RELIGIOUS CUSTOMS AT YEZD 

lous care in order to restore the ceremonial purity that may 
have been lost by contact with the dead. It consists in a 
series of sprinklings with bull's urine (a supposed disinfectant) 1 
and consecrated water, accompanied by an endless amount of 
ritual that is thought to aid in exorcising the spirit of con- 
tamination. 2 

Originally, as in the Avesta, the rite appears to have been 
used only in cases of great defilement, when a man had actually 
touched a corpse, or when a woman had given birth to a 
dead child, or it was employed to insure the absolute cere- 
monial purity of a priest who is to perform a sacrifice ; later 
its usage appears to have become more extended, so that it was 
resorted to as a means of securing purity in general, and it is 
so used by the Indian Parsis, both in the abridged form of a 
mere ablution and in its full form of a lustration covering nine 
nights. According to the Persian treatise entitled Sad Dar, 3 
4 it is strictly incumbent on mankind, on man and woman, to 
perform the Barashnum ceremony ' at least once in a lifetime 
in order to purify the soul for entrance into heaven, for other- 



1 The urine of beef was thought to logues, that his hero, Menippus, was 
possess great purifying and medicinal washed in the Tigris at Babylon for 
qualities and is called gaomaeza, twenty-nine days ' by the Magi, the 
'beef's urine,' in the Avesta (Vd. 9. disciples and successors of Zoroaster.' 
14 ; 19. 21, 22) and gomez in Pahlavi It is probable that this was the way in 
and Modern Persian. When conse- which the Magian ' Zabratas,' or 
crated by special prayers it is called * Zaratos,' according to Porphyrius 
riirang and is generally spoken of by ( Vita Pythagorae, 12) cleansed Pythag- 
that name. See the treatise of Wil- oras of all the sins he had committed 
helm, On the Use of Beefs Urine in his life. For the Greek texts of 
according to the Precepts of the Avesta, these passages, see my Zoroaster, pp. 
Bombay, 1889. 237, 242 ; Nauck, Porphyrn. Ojwscula 

2 See .Vd. 9. 1-46 ; 8. 36-72 ; com- Tria, p. 18, Leipzig, 1860 ; and cf. 
pare also the discussion of the Bara- Kleuker, Zend-Avesta, Anhang, vol. 2, 
shnuin ceremony by West, SEE. pt. 3, pp. 104, 117, Riga, 1770-1783. 
18. 431-454, and the notes by Darme- 8 See Sad Dar, 36. 1-8, tr. West, SEE. 
steter, Le ZA. 2. 159-172, and SEE. 24. 296-298, and compare Darmesteter, 
4. 122-134. The Greek writer Lucian SEE. 4. 123. The treatise Sad Dar, 
alludes to this prescription when he in its oldest form, dates back to the 
says, iii one of his humorous dia- time of the Arab conquest. 




Two PICTURES OF KHODABAKHSH BAHRAM RAIS 
(The Kalantar's young son in the background) 




PARSI WOMEN AND GIRLS 
(1 and 2, English ; 3 and 4, Armenian) 



THE BARASHNUM GAH AT YEZD 383 

wise the natal impurity contracted by being in the womb and 
sucking at the mother's breast will not be removed. 1 

When I visited the Barashrmm Gah at Yezd, which is located 
within easy walking-distance of the fire-temple I found a prim- 
itive mud-walled structure, circular in form and differing 
greatly from the somewhat elaborate rectangular enclosure 
I had seen at Udvada in India. There were a few heaps of 
stones placed at certain intervals around the edge of the 
circuit, intended to serve as standing-places for the person 
undergoing this purification, since, according to the Vendidad, 
he must move from spot to spot within lines specially drawn 
in the sand and be sprinkled by the priest with the drops of 
bull's urine and water. The liquid is poured from a ladle 
fastened to the end of a stick ' nine knots ' long, or long enough 
to allow the officiating priest to stand outside the circles that 
have been drawn. I observed a number of niches in the wall, 
precisely like the ordinary tdkchahs, or sunken recesses, in the 
walls of a Persian room ; they were used, I was told, as recep- 
tacles for food and drink to be taken by the person undergoing 
purification. The whole place looked dilapidated and neg- 
lected, 2 and from what I observed, as well as from conversation 
on the subject, I infer that less attention is given in Yezd to 
keeping up this ceremony than I should have expected in the 
centre of Zoroastrian orthodoxy in Iran. From the remarks 
of Khodabakhsh Rai's I conclude that certain of the more 
advanced Zoroastrians are opposed to insisting upon the 
importance of keeping up this rite, especially in the case 
of women, although the Avesta prescribes it under certain 

1 Sad Dar, 37. 1-6. According to in India, see the account of the initia- 

Khodabakhsh Rai's, the cleansing from tory ceremony called navar, by Modi, 

the natal impurity is technically Zoroastrian Priesthood, in Zartoshti, 

called sustan-i sar-i .sir, lit. ' washing 1. 94, Bombay, 1903. 

of the head of milk,' milk being re- 2 This was perhaps to be expected 

garded as blood turned white and there- from its nature as a place of isolation, 

fore impure, since blood defiles. For but I could not help contrasting it with 

the comparative frequency with which the Barashnum Gah at Udvada. 
the Barashnum ceremony is employed 



384 ZOROASTR1AN RELIGIOUS CUSTOMS AT YEZD 

circumstances for women as well as men. 1 I inquired further 
about the subject from Bahram, the son of Kalantar Dinyar, 
and I found that the youth himself, for example, had never 
undergone the ceremony, a fact which shows that it is not 
regarded as obligatory for removing the original taint incurred 
at birth, whatever may be the custom with regard to priests. 

I shall now turn to the subject of marriage among the fol- 
lowers of Zoroaster. The age at which the Iranian Zoroastrians 
marry is usually between twenty-five and thirty in the case of 
men, and between fourteen and nineteen in the case of women. 
Yet boys sometimes marry at the age of fifteen, and girls are 
given in marriage when only twelve. Instances have also oc- 
curred where a widower of sixty has married a girl of fifteen or 
has taken a widow of twenty for a wife. I was also informed 
that in the case of families closely connected by ties of friend- 
ship, it has happened that a two-year-old daughter in one family 
has been betrothed by her parents to a three-year-old son in 
the other household; but such contracts are looked upon by 
the community with disfavor. 2 

Parents, as a rule, arrange the marriages of their children, 
since a son cannot take a wife, nor a daughter be married, with- 
out the consent of the parents or guardians, and the mother has 
as much to say as the father regarding the choice that is to be 
made. When the consent of all concerned has been obtained, 
a formal betrothal of the young couple takes place; gifts of or- 
naments and money are frequently exchanged; and the marriage 
banns are then publicly announced by the High Priest, without 
whose consent a Parsi cannot marry, unless he disregards the 

1 See Vd. 9. 21, and compare Sad but now are forbidden. See Karaka, 
Dar, 36. 1 (quoted above) and connect History of the Parsis, 1. 171-172. 
with it the custom of segregation in For valuable statistics of the mar- 
the Armesht Gah, Vd. 5. 45-62, or in riage ages of the Parsis in India, by 
the Dashtdnistdn, Vd. 16. 1-18 ; cf. Bamanji Behramji Patel, see the inter- 
Darmesteter, Le ZA. 2. x-xv. esting chapter on marriage, in Mile. 

2 The same is true in India, where D. Menant, Les Parsis, pp. 154- 
such infant marriages formerly took 155, Paris, 1898. 

place occasionally among the Parsis, 



MARRIAGE CEREMONIES 385 

authority of the priest and marries according to the law of 
Islam. 

The ceremonies connected with the marriage itself differ from 
those in use among the Parsis of India, especially in being 
much simpler. In India the bride sits by the side of the bride- 
groom, and after the marriage witness for each has formally 
given sanction to the union, the two priests stand before the 
young couple, recite an address of prayer, admonition, and 
benediction, a part of which is in Sanskrit and accompanied by 
rites borrowed from the Hindus. 1 In Persia the bride does not 
take part formally in the ceremony except by renewing her con- 
sent to the bridegroom's representative, who comes again for 
that purpose to her house on the marriage day, before the wed- 
ding begins. In Yezd only men are regularly present at the 
matrimonial service, but the bride and her female relatives and 
friends are usually near enough to hear the texts recited, and 
sometimes they stand on the roof of the house to watch the 
proceedings. When the male relatives and friends of both fami- 
lies are assembled, the priest (not two, as in India) takes a 
seat, and the bridegroom and a representative of the bride, her 
father, or some one who has helped to arrange the match, are 
seated on his right, the groom's party being on the other side. 
Sugar candy is placed in the hand of the bridegroom by the 
bride's marriage witness and is formally accepted, and after a 
few minor ceremonies the officiating priest recites the marriage 
address, Andarz-i Gf-avdh, lit. 'Admonition of the Witness,' 
composed partly in the Dari dialect, the language current 
among the Iranian Zoroastrians, and partly in Pazend and 
Zend. 2 It contains wholesome advice and admonitions and 

1 In India this address is called Pai- Avesta (lithographed), pp. 435-450, 
vand-Ndmah or Ashirvdd. See Modi, Bombay, 1900. Mr. Khodabakhsh 
Marriage Customs of the Parsis, pp. Rai's (who compares gavdh with Skt. 
34-39, Bombay, 1900 ; Karaka, History mvdha} has given an interlinear ver- 
of the Parsis, 1. 189-192. sion of the Dari words in Modern 

2 The text of the Andarz-i Gavdh Persian, 
may be found in the Persian Khordah 

2c 



386 ZOROASTRIAN RELIGIOUS CUSTOMS AT YEZD 

a benediction invoking blessings upon the two that are now 
married and upon all mankind. When the Andarz is finished, 
gifts of Kashmir hats and presents of sugar loaves are freely 
exchanged by both parties, good cheer is the order of the 
hour, and the guests partake of the viands provided for the oc- 
casion. When the supper is finished, they accompany the 
bride, who is veiled from head to foot in a robe of green silk, 
to the bridegroom's house, where she enters upon her new 
life. 

In the home the wife occupies a freer position than the 
women in the Mohammedan household, and despite the Moslem 
influence, which would tend to make the Zoroastrian regard his 
wife as his inferior, she enjoys more of the old Persian law of 
equality ' and sometimes even gets the better of her husband,' 
as the statement was frankly made to me. 1 The women whom 
I saw appeared to be dignified without reserve, and modest 
without diffidence, although, of course, they do not enjoy the 
greater advantages of the Parsi women of Bombay, whose op- 
portunities have been many. 

In their domestic relations the Zoroastrians of Yezd are 
monogamists, as a rule, but bigamy and even polygamy, 
which they attribute to the influence of their Mohammedan 
surroundings, were not uncommon in former times. The sen- 
timent of the Zoroastrian community, as was evident from 
two particular instances which they cited, is distinctly against 
dual marriages, even in cases where the first wife has borne no 
child to the husband, and for a woman to commit bigamy 
would mean death. The cause of the difficulty of enforcing 
the standards of the Zoroastrian faith and preventing infringe- 
ments of the marriage law is found in the force of the Moslem 
example that prevails around them. A Zoroastrian who is 
unwilling to abide by priestly regulations in such matters 

1 On the law of equality of the wife trian Women in Remote Antiquity, as 
in Ancient Iran, see Darab Dastur illustrated in the Avesta, pp. 35-42, 
Peshotan Sanjana, Position of Zoroas- Bombay, 1892. 




If 

Q 




1 

SI 

3 ft 

I - 

N .2 



DOMESTIC RELATIONS 387 

simply rejects them, or, if he chooses, goes over to Islam. For 
that reason, as I learned, cases of breach of promise and divorce 
occur even in the circle of the faithful, and certain other vices, 
which I need not mention, also exist ; but the main body of 
the Zoroastrian community is making earnest efforts to eradi- 
cate these evils from their midst and to uphold the high ideals 
set by their religion. 

The funeral rites of the Zoroastrians in Yezd are practically 
a continuation of the ancient customs of the Avesta and 
substantially the same as those of the Parsis in India, but with 
minor points of difference due either to local conditions or 
to present circumstances as contrasted with the past. Since 
nothing in detail has been written in English regarding these 
mortuary ceremonies of the Modern ' Gabars ' in Iran, I shall 
devote some space to the subject, making occasional compari- 
sons with the more familiar Parsi customs in India and the 
ancient rites of the Avesta. 1 

When a person is at the point of death, a mobed, or priest, is 
usually summoned to administer the last rites. He recites the 
prayer of repentance for sins (patit pashimdnl, 'penitential 
office ') and performs the ceremony of extreme unction by pour- 
ing on the lips of the dying some drops of consecrated bull's 
urine (gomez, Av. gaomaeza). When the person is dead, the 
priest goes to the fire-shrine or to the temple and performs 
the srosh-drun ceremony for the repose of the soul of the 
departed. In Persian this ceremony is called ravdn barsm. 

1 For detailed information on the my pupil, Mr. Maneckji Nusservanji 

Zoroastrian funeral rites at Yezd I am Dhalla. In comparing the rites of the 

again indebted to the kindness of Parsis in India, which I knew also by 

Khodabakhsh Bahrain Rai's. I have experience, I have referred to Modi, 

supplemented my material by notes Funeral Ceremonies of the Parsees, 

from an interesting article, written in Bombay, 1892 (reprinted from Journ. 

Gujarati, by Dastur Khudayar She- Anthropolog. Soc. of Bombay, 1891) ; 

heryar, A Zoroastrian Death in Per- Karaka, History of the Parsis, 1. 192- 

sia, in Zartoshti, 1. 169-181 (Bombay, 213 ; Mile. D. Menant, Les Parsis, 

1904). For a translation of the main pp. 179-235, Paris, 1898. 
points of this Gujarati article, I thank 



388 ZOROASTRIAN RELIGIOUS CUSTOMS AT YEZD 

Shortly after death has taken place the corpse, which 
henceforth must not be touched, except by those who have to 
do so, is placed at one side in the room and washed. This task 
is performed by a murdah-shur, 4 corpse-washer,' or pdk-shur, 
* clean- washer,' accompanied by an assistant, since, according to 
the Avesta, 1 one is never allowed to be alone with a dead body. 
The two corpse- washers also constantly hold a Jcusti- thread 
between them to signify their joint action (paivand, lit. 
4 union, connection ') in the work. The person who washes 
the body wears on his hand a fleecy glove of wool (pashm^), 
over which, as he rubs the body, his assistant pours beef's 
urine ( gd-mez) 2 from a brass bowl with a long-handled spoon. 
Water is never used for this purpose. It is a rule, moreover, that 
men should wash the corpse of a man, and women that of a 
woman, after which they clothe the body in a clean, but worn- 
out, white garment, including the sacred thread (k-ustl'), but 
they do not put anything new on the body, as the Avesta for- 
bids such an act. 3 When their task is completed, they wash 
themselves thoroughly in order to remove the defilement caused 
by contact with the dead. 

The ceremony of the sag-did, ' glance of a dog,' is now per- 
formed for the first time. This ancient rite, which dates back 
to the period of the Avesta, consists in making a dog look at 
the dead body, since its gaze is believed to have a peculiar 
efficacy for driving away the nasu, or spirit of defilement. 
Various explanations have been suggested to account for the 
origin of this custom, from the mythological idea of the four- 
eyed dogs of Yama in the Veda, all the way down to the 
rationalistic theories that it is due to the instinct which the dog 
has for discerning whether any life remains in the body, or 
owes its origin to the time when the corpse was devoured 
by dogs as well as birds. 4 No special dogs are employed in 

1 Vd. 8. 14. Cf. Vd. 5. 61 ; 8. 23-25 ; see also 

2 Av. gaomaeza, Vd. 9. 14 ; 19. 21, Sad Dar, 12. 1-2, and cf. Vd. 6. 51. 

22. For the role of the dog in connec- 



FUNERAL RITES 

Yezd for performing the sag-did ; l the ordinary street-dog is 
called in for the purpose. Morsels of bread are strewn around 
the corpse, or, according to the older usage, laid on the bosom 
of the dead, and the dog eats these. 2 

The corpse-bearers now come to remove the body, first to a 
sort of mortuary chapel or charnel-house (zdd-o-marg, lit. 
4 birth and death '), as I shall explain, and then to the dakhmah. 
These carriers (nasu-kashas), or pall-bearers {plsh-gdhdn, lit. 
4 before the bier'), convey the body on an iron bier (gdhdri).* 
The men who perform this office do not as a rule belong to a 
special class, but any one who may volunteer or be hired for 
the purpose may do this service ; some tend, however, to make 
it a part of their livelihood. The number of bearers varies from 
twelve, sixteen, twenty-four, to thirty-two, according to the 
weight of the body and the distance to be traversed, but in no 
case can they be less than two, since the Avesta forbids that 
one man alone should carry a corpse. 4 They always hold a 
kusti between them to keep up the mystic union in their task, 
and must be prepared afterward to wash their persons and 
their clothes thoroughly before again associating with other 
people. 

A procession (pdddsJi) is now formed to conduct the body 
to the temporary receiving-vault. A man walks in front, hold- 
ing a fire-vase in which incense is burning. He is followed by 
the relatives and friends; and after them the dead is carried, 



tion with death, see Modi, Funeral khdndhiahs, ' shoulder-men,' and are 

Ceremonies, pp. 8-10, and Bloomfield, not to be confused in either case with 

Cerberus, the Dog of Hades, pp. 27-31, the bearers who carry the corpse in- 

Chicago, 1905. For the dog in the side the dakhmah ; cf. Modi, op. cit. 

Avesta, see Hovelacque, Le Chien p. 12. 

dans V Avesta, Paris, 1876, and Kuka, 4 Vd. 3. 14-21 ; 8. 10. For the 

The Dog in the Vendidad, in Zar- numbers twelve to thirty-two, see 

toshti, 1. 271-280. Dastur Khudayar Sheheryar, op. cit. 

1 See p. 78, above. p. 172. Anquetil Duperron (Zend- 

2 This information I have directly Avesta, 2. 584, Paris, 1771) said that 
from Khodabakhsh Rai's. forty was the number at Kerman. 

8 In India they are generally called 



390 ZOROASTRIAN RELIGIOUS CUSTOMS AT YEZD 

lying on the iron bier and followed by the mobeds and some 
more members of the family of the deceased. 1 

They all march in solemn manner to the zdd-o-marg, OYparsish- 
khdnah {pursish-khdnafr), the charnel-house, which serves as a 
receiving-vault before the body .is finally carried to the dakh- 
mah. 2 

The use of this mortuary building can be traced directly 
back to the kata of the Avesta, since the Vendidad commands 
that 'in every house and in every village they shall erect 
three katas for the dead.' 3 The present charnel-house is of 
larger dimensions evidently than the small structures described 
in the Vendidad, for there is a separate room for the women, 
a compartment for the corpse, and one for the bier, besides the 
room where the relatives and friends gather. The building, 
as now constructed, has two doors, through one of which the 
corpse is brought in and through the other of which it is carried 
out, symbolizing the idea of birth and death (zdd-o^marg) given 
in a Persian couplet: 

' What is the world ? It is simply a halting-place, with two gates. 
By the one ye enter; by the other ye depart.' 4 

But the custom of carrying the corpse out by a way different 
from the entrance by which it came in, appears to be as old as 

1 In former times it was customary 8 See Vd. 5. 10-14. Anquetil Duper- 
in the villages and outlying districts to ron (op. cit. 2. 583) alludes to the 'zad 
have the procession led also by some marg' in India, and its use is said still 
one blowing a horn, beating a drum, or to linger in the Gujarat provinces, but in 
making doleful music, but this custom general the Parsis have abandoned it 
has almost died out. For the sub- and convey the body within twenty-four 
stance of these two or three particular hours directly to the dakhmah, as ne- 
paragraphs I am indebted to Dastur cessitated by the hot climate. A par- 
Khudayar Sheheryar's article already tial survival of the zad-o-inarg among 
referred to. the Indian Parsis is the nasd-khdnah, 

2 The name zad-o-marg, * birth and ' dead house,' where the bier and 
death,' is explained directly below ; other funeral equipments are kept by 
the designation pursish-khanah, lit. the nasd-sdldrs. See also Modi, Fu- 
1 house of inquiry,' is from the cus- neral Ceremonies, p. 7, n. 9, and Darme- 
tom of coming to inquire (pursi- steter, SBE. 4. 53, n.2 ; 07, n. 1. 
raftan) and condole with the family 4 The Persian text is given by Khu- 
of the deceased. dayar Sheheryar, op. cit. p. 171. 



A MORTUARY CHAPEL 391 

the Avesta, since the Vendidad alludes to making a ' breach ' 
in the wall to take out the body. 1 

When the corpse is brought within the zad-o-marg, it is 
removed from the iron bier and laid on a raised platform of 
mud paved with stone, about nine feet long and four feet wide, 
and the bier is carried into a separate room. 2 The friends 
gather to pay their last respects to the dead ; this is called pursl- 
raftan, lit. 'coming to ask,' or sez (so/). 3 The mobeds then 
begin the G-dhdn Srdyishn, or chanting of the Gatha Ahunavaiti 
(Ys. 28-34), which treats of various subjects relating to piety, 
faith, and the future life. When the recitation is half over, the 
priests cease chanting, 4 the sag-did is performed once more, 
and the body brought from the room where it was laid, and 
replaced upon the iron bier, to which it is fastened by a thick 
kusti, and a white sheet is thrown over the whole. 

The procession is then re-formed and leaves the building, 
reciting verses from the Avesta in memory of the dead (iris- 
tanftm). 5 After going a short distance a halt is made, the 
women and those who do not desire to follow the body all the way 
to the dakhmah, pay their parting tribute to the dead, and then 
the near relatives and friends go on in procession. As the 
main dakhmah at Yezd is about nine miles from the city, many 
in the procession ride on horses or donkeys, but the priests go 
the entire distance on foot. It sometimes happens also, espe- 
cially in the case of hamlets and villages far remote from the 

1 Vd. 8. 10. Compare the Pahlavi Sheheryar, op. cit. p. 172), and the same 
commentary, on this passage and the is true in India (cf. Modi, op. cit. p. 14). 
notes by Darmesteter, Le ZA. 2. 121, &See Sheheryar, op. eft. p. 173. I am 
n. 15, and SEE. 4. 97, n. 6. not quite certain which particular verses 

2 For these special details I have are referred to, but (if not Ys. 26. 7) I 
combined my notes from Khodabakhsh presume that the passage may be 
Rai's with material from Khudayar Ys. 16. 7, 'we praise those bright 
Sheheryar. abodes of Righteousness in which 

3 See note above and compare the dwell in happiness the souls of the 
sejdo of the Indian Parsis ; cf. Modi, dead (iristonSw), which are the 
op. cit. p. 15. spirits (fravashayo) of the righteous. 

4 The pause is made at the words tat_ We praise the best world (Paradise), 
moi m&dyai, Ys. 31. 5 (so Khudayar holy, brilliant, and all-glorious.' 



392 ZOROASTRIAN RELIGIOUS CUSTOMS AT YEZD 

dakhmah, that the corpse is placed upon a cow or a donkey, 
instead of being carried 1 a procedure which would be in no 
way striking in Persia, because there corpse-caravans convey 
dead bodies for long distances to Mohammedan shrines. 

Upon reaching the dakhmah, the sag-did is repeated for the 
third and last time, the final preparations made, and the corpse 
is given over into the charge of two other men whose special 
office it is to convey it within the tower. They are called 
nasd-sdldrs, 'chief of the dead,' a designation also applied to 
them in India. 2 They must be men well advanced in years and 
of a high moral character. Owing to the nature of their occu- 
pation they are obliged to live apart from men and are not 
allowed to mingle with other members of the community or 
to enter a house where religious rites are being performed or 
any special festival is being celebrated, and if possible they 
must also refrain from tilling the earth. They are obliged, 
moreover, to wash themselves thoroughly after having carried 
a corpse into the tower, and if they should ever wish to resign 
from their office, they must undergo the ceremony of the 'Nine 
Nights Washing' (barashnum no-shva, no-sJiaba), as already 
described. 

As soon as they take the corpse into their charge at the door 
of the dakhmah they make the paivand-bond, by holding a 
kusti between them, and they recite the Srosh Baj. One of 
them next takes a piece of metal or an iron key and, beginning 
near the left ear of the corpse, draws three furrows (kash) 
around it, reciting the Ahuna Vairya formula, or Avestan 
paternoster, as he draws each circle. 3 The two nasa-salars now 
take up the body and carry it into the dakhmah, laying it down 
with the head toward the south, and removing the clothes, in 

1 A reference to this custom is even in this respect, see Modi, op. cit. pp. 

found in the Pahlavi commentary on 12-18. 

the Vendidad (Phi. Vd. 3. 14) . 8 For the material in this and the next 

2 My statements in this paragraph five paragraphs I am indebted to the 

are based on the authority of Khoda- Gujarati article by Dastur Khudayar 

bakhsh Rai's. For the Indian customs Sheheryar, op. cit. pp. 169-181. 



AT THE DAKHMAH, OR TOWER OF SILENCE 393 

accordance with the ordinances of the Avesta. 1 They then 
recite in Persian the following prayer to the angels (yazatas) : 

'0 victorious and holy Srosh, we have removed N or M 2 from 
the earth, Spendarmad, and have committed him (or her) to the 
stone Ayokhshust. 3 Angel Srosh, we turn our backs upon him 
(her), but do thou turn thy face towards him (her); into thy 
keeping we have given him (her); do thou take his (her) hand.' 

[To the corpse.] 'Do not thou, N or M, be afraid, do not tremble, 
because this place is thousands of years old; it is the resting-place of 
our fathers and mothers and our ancestors.' 

[To the Angels.] '0 Srosh, Mihr, and Eashn the Just, we have 
delivered him (her) into your keeping; take his (her) hand, and lead 
him (her) to the abode of our forefathers and the righteous and the 
pure. So be it, in accordance with the will of the Angels and Arch- 
angels (izad u amshaspandari) ; so be it; so, verily, let it be.' 4 

The corpse-carriers withdraw from the tower after this, and 
the body is left exposed for the birds of prey to devour the 
flesh. When the bones have been denuded and become dry 
they are usually laid in a separate place in the dakhmah and 
turn to dust. 5 In no instance nowadays is the corpse torn by 
dogs or wild beasts as it was in ancient days, nor in any case 
is burial in the ground lawful, since it was prohibited by 
Zoroaster. If a dakhmah is not accessible, the body may be 
disposed of after the manner known as sang-chin, 'heap of 

1 The removal of the clothes is im- metal, 1 see my article in JA08. (Pro- 
plied in the Avestan phrase rao6&-aiwi- ceedings, p. Iviii), 1890. 

varena, ' clothed with the light of * See Khudayar Sheheryar, op. cit. 

heaven' (Vd. 6. 51). In the Avesta p. 174. 

(Vd. 19. 1; Yt. 22. 7) and throughout 6 See what I have said below (p. 439) 

the Zoroastrian Scriptures the southern regarding the dakhmah at Rei near Te- 

region is auspicious, the northern re- heran. Compare likewise the descrip- 

gion the abode of Ahriman and the tion of the dakhmah at the village of 

demons. Shah Ali near Yezd, which was exam- 

2 Here the name is to be inserted and ined by Westergaard in 1843 (JRAS. 8. 
the rest of the prayer made to conform 352); furthermore the account of the 
to it. dakhmah near Isfahan, which Chardin 

3 This expression, * stone Ayokh- in the seventeenth century described as 
shust' (Pers. sang Ayokhshasf), is not round with a central pit ( Voyages, 3. 
quite clear; but it appears to answer 131). In the Bombay towers this 
to Avestan ayokhshusta, ' molten central well is called the bhanddr. 



394 ZOROASTRIAN RELIGIOUS CUSTOMS AT YEZD 

stones.' Under these circumstances the corpse is carried to 
some remote place in the hills or mountains, is then piled 
around with stones and covered with a slab, but not interred. 1 

It is customary, after the body has been exposed upon the dakh- 
mah, for the friends and relatives to partake of some refresh- 
ment after their long march ; the simple meal consists of bread, 
cheese, potatoes, or eggs, as the case may be, with some wine, 
but no meat or melted butter is eaten. Prayers are again 
offered for the dead and sympathy is expressed for those in 
affliction, and then all those present perform the ^ws^-ceremouy 
and return home. 2 

According to the ancient and present Zoroastrian belief, the 
soul hovers near the earth for three days after death, before 
departing to the other world. During that time the family 
observe certain rules, pray for the dead, abstain from eating 
meat and from any act that might cause distress to the soul or 
detain it longer on earth. On the spot where the body lay 
before it was removed from the house, it is usual to place three 
bricks in the form of a little arch and to thrust an open pair of 
scissors into the ground to drive away any evil spirits that may 
be lurking near. The ceremony of the Yasna is performed 
each morning, between sunrise and noon (Havani Grdli), the 
Srosh Yasht is recited in the evening after sunset (Aiwi- 
sruthrima Graft), and the Vendidad ritual is celebrated at mid- 
night, unless for some reason the corpse must remain in the 
house over the first night, in which case the Vendidad observ- 
ance is omitted. 

On the afternoon (Uzayeirintt Gdh) of the third day the 
priest is invited to recite some texts from the Avesta, 3 with 
accompanying ceremonies, and some food is prepared for the 

1 This is practically the method pur- eral description here given presents all 
sued at Shiraz (see p. 337, above) and the more important details, 
wherever there are not enough believ- 8 Selections from the Khordah. 
ers to justify a dakhmah. Avesta (e.g. Khurshed Nyaish, Uzeirin 

2 Minor variations in the funeral Gah, and Patit Pashimani) are used 
rites are found, of course, but the gen- for this purpose. 



RITES AFTER DEATH 395 

evening. At midnight, in the presence of the assembled com- 
pany, religious rites in honor of the deceased (yasht-i sedush) and 
the dedication of a white muslin garment takes place at this 
vigil (shab-girih), so that the soul may not be naked in paradise. 1 
At dawn ( Ushahin Grdh), when the soul is believed to be 
crossing the Bridge of Judgment (Av. Ohinvat Peretu, Mod. 
Pers. Ohinvad PuT) the ceremonies of the cliahdrom, ' fourth 
day,' are carried on. These rites are believed to be efficacious 
in facilitating the difficult passage of the soul over the Bridge. 
Invocations are made to the angels Sraosha, Rashnu, Arshtat, 
Rama Hvastra, and to the Fravashis, and prayers are offered for 
the dead. After the completion of this requiem mass all those 
who are present, with the exception of the priest, partake of a 
slight repast of food previously consecrated, and the priest, with 
a rosary of beads, asks each of the mourners how many prayers 
he will offer in memory of the deceased, and after announcing 
the number and reciting some texts of absolution and benedic- 
tion, dismisses the assembly. 2 

1 The designation shab-girih appar- made only in cases where the person 
ently means * night-watch, vigil for the dies absolutely childless ; but in former 
dead, wake,' and is then applied to the times it was done even if daughters 
garment that is dedicated to the de- were left, but no son. Formerly only 
ceased and thus answers to shiydv a boy was eligible, but now even a girl 
among the Indian Parsis. This cus- may be chosen, although such a choice 
torn, with others that are still kept up is rare. The age of the person adopted 
by the Zoroastrians, is alluded to in is generally over fifteen years, but now- 
Sad Dar, 87. 1-11, see West, SEE. adays an infant may be nominated to 
24. 350-352. the office, its father acting for it dur- 

2 In performing the rites of the third ing the years of its minority. As is 
night and the fourth day (chaharoni) , natural, the one chosen to serve in the 
when the soul is crossing the Bridge, office of pul-guzar is selected from the 
the offices of a son and heir are particu- nearest relatives and acts like an ex- 
larly important. For that reason, in ecutor to an estate, dividing the prop- 
the case of an adult of fifteen years or erty among those who are of kin to the 
more who has died without leaving a deceased, and distributing a large part 
child, it is appropriate to appoint an of it in charity, especially in funds for 
adopted son who assists in the crossing the annual celebration of the Gahanbar 
of the Bridge and is therefore called pul- festivals. (This note is from memo- 
guzar, ' bridge-crossing.' Nowadays randa furnished me by Khodabakhsh 
the appointment of an adopted heir is Ra'is.) 



396 ZOROASTRIAN RELIGIOUS CUSTOMS AT YEZD 

Some additional rites are observed on the tenth day (dahah), 
the thirtieth (siruzah or slruzhah), the return of the day (ruzah 
or razhah) each month, and again on the anniversary (sZ or 
sar-i sal) of the death, and these observances are to be kept 
up as long as possible, besides remembering the dead during 
the Fravardigan Festival of ten or eighteen days at the end of 
the Parsi year. 1 There are likewise several other observances 
which, though not directly connected with death, have neverthe- 
less a bearing on the subject of the repose of the soul hereafter. 
They are enjoined by the priests and performed by the orthodox, 
but by no means without exception. Among them are the reci- 
tation of a thousand prayers of repentance (hazdrah-i patit) to 
gain absolution from sins ; likewise a thousand prayers to 
fire and water (Jiazdrah-i dtash-nydish, hazdrah-i db-nydish) in 
atonement for sins committed against those elements ; or again, 
an invocation of the earth (yasht-i bin-i Sipanddrmiz or Spanddr- 
mad) to pardon any defilement of the ground, which may have 
been committed in life, even by going barefoot ; and lastly the 
consecration of several fires (dtash-i mas kartvun or dtash buzurg 
kardan, lit. 4 aggrandizing the fire ') as an act of merit and 
atonement. Mention may also be made of the so-called rite of 
sahm-astah, lit. 4 dread of bones,' offered by a widow who marries 
a second time and desires to appease the soul of her former 
husband ; and finally, of prayers recited near the dakhmah for 
the repose of the dead (yasht-i daur-i damah), together with 
the ' Nine Nights Ablution ' (barashnum no-shva, no-shaba) 
already described. 2 

It is interesting to note that, according to information I 
received, a recent convert (Jadld) from Zoroastrianism to Islam 
sometimes still maintains his old-time orthodoxy sufficiently 

1 The Fravardigan Festival is a per- the preceding paragraph I am indebted 
petuation of the Avestan fravashi-wor- to notes given me by Khodabakhsh 
ship, or commemoration of the souls Ka'is. I am not certain as to the pre- 
of the departed, somewhat like our cise meaning of the words daur-i 
All Saints' Day. damah, although he explains damah as 

2 For the subject-matter of this and dakhmah. 



CEREMONIES OF PURIFICATION 397 

to desire his funeral services to be performed according to 
the Avestan ritual. It has likewise happened that the body 
of such a convert after being buried has been stolen by 
night from the grave and carried under cover of darkness 
to the dakhmah. No precise investigation of the matter 
is afterwards made. The Parsis give out that angels 
have come down from heaven and borne the dead on high to 
the throne of God, and the Moslems believe that the angels 
have come and taken the body to Najaf in Arabia, to lie by the 
side of the holy Ali, Shdh-i Najaf, 4 King of Najaf.' 

I shall conclude this sketch with some notes regarding the 
the dakhmahs in the vicinity of Yezd. There are two such 
towers in the hills west of the city, and both are in use. One 
of these is old and is called Dakhmah-i Jamshld^ ' Tower of 
Jamshid ' ; the other, which is round in shape like the Indian 
Towers of Silence, is situated opposite to this and is of com- 
paratively recent date. It was erected by Manakji Limji 
Hoshang Hantaria, who came from Bombay to Persia in the 
beginning of the year 1854 as representative of the Persian 
Zoroastrian Amelioration Fund. 1 Still farther to the west of 
Yezd, at a distance of twelve or fifteen miles in the direction 
of Taft, there are the ruins of an ancient tower, called 
Dakhmah-i Kuhnah, 4 Old Dakhmah,' which is now used only 
as a place for exposing the bodies of still-born or abortive 
children and of persons who have died by suicide or in some 
violent manner. 

There are several other dakhmahs located at various points 
north of Yezd. One of these is situated on a hill called Zarch 
Kuh, near the village of Ilahabad, about ten miles from the 
city ; it was built in memory of a rich childless merchant, 
Khosru-i Mihrban-i Rustam, by his adopted son, Ardeshir Mihr- 
ban Irani, a philanthropic Zoroastrian of Yezd. Still farther 

1 An interesting account of this aka, History of the Parsis, 1.72, where 

agent of the Indian Parsis and of his his name is recorded as Manakji 

efforts in behalf of his oppressed co- Limji Antaria. 
religionists in Persia is given by Kar- 



398 ZOROASTRIAN RELIGIOUS CUSTOMS AT YEZD 

north, near the village of Sharafabad, are several other Towers 
of Silence, which are referred to in the next chapter. l 

Most of my details regarding the dakhmahs I received 
directly from Khodabakhsh Rais, from whom I also learned 
that the Parsis had a tradition, handed down from antiquity, 
that when Persia was under the rule of Zoroastrian kings 
and the country was rich and prosperous, each worshipper 
of Ormazd built for himself during his lifetime a dakhmah to 
be used at his death, and that individual dakhmahs of this 
kind were called dakhmah-i tan bah tan, 4 dakhmah for a single 
body.' 

It is furthermore reported that the large structures, like 
those in use at present, were originally called dakhmah-i 
lashkarl, ' dakhmah for soldiers,' because they were put up for 
the corpses of those slain in battle, and their use became 
general after the Arab invasion, as the Zoroastrians were no 
longer able to keep up their religious rites with all the former 
detail and were obliged to resort, therefore, to the common 
large towers and discontinue the practice of building individ- 
ual receptacles for the bodies. The tradition of individual 
dakhmahs is certainly interesting because of its bearing on the 
Vendidad, where dakhmahs are alluded to as if very numerous. 
It is worth adding that Khodabakhsh is of the opinion that the 
original dakhmahs were built with mud walls, like those of the 
old one near Sharafabad. 

From all that has preceded, it will be manifest how closely 
the Zoroastrians of Yezd still follow the injunctions of the 
Vendidad. A further illustration may be gathered from the 
following amusing incident in daily life. The cook of my 
English hosts at Yezd was a Gabar, and on one occasion, as 
they told me, he had made some wine and purchased an earthen 
jar in which to store it, but he first filled the vessel with water 
and let it stand over night before placing the wine in it. 
A mouse accidentally fell into the jar and was drowned. The 

1 See p. 403, below. 



> 



TOWERS OF SILENCE NEAR YEZD 399 

receptacle was henceforth unclean in the eyes of the Zoroastrian, 
because it had been polluted by contact with dead matter, and 
was therefore unfit for use. 1 The man's business sense, how- 
ever, and his regard for thrift since the Avesta prohibits 
wasting anything 2 would not allow him to throw the jar 
away, so he sold the vessel at a reduced price to an Armenian, 
who had no scruples against using it. 

This combination of thrift and practical sense, united with a 
tenacious adherence to the faith of their forefathers, is a char- 
acteristic of the Zoroastrians of Yezd. The impression which 
I gained of them was very favorable, on the whole, considering 
the conditions under which they live in Persia, as contrasted 
with the advantageous environment of their Parsi brethren in 
India. From the latter they have much to learn in the way of 
progress, enterprise, and intellectual activity, and they have 
little to offer in return, even in the way of religious customs 
and observances or gifts of ancient manuscripts relating to the 
faith. 

Nevertheless, so far as my limited observations allowed me 
to judge, there are some of their customs and certain of their 
methods of conducting religious ceremonies that deserve 
further study from the specialist, as such observances may 
actually be nearer the ancient forms, and therefore historically 
valuable, even if it be no longer practicable or desirable to 
follow them. 3 In any case the Parsis of India are thoroughly 
justified in taking an active interest, as they have done, in 
their Zoroastrian kinsmen in Iran, whose motto, wherever 
they be, whether in Yezd, Kerman, Teheran, or elsewhere, is 

1 The laws which underlie the Zoro- the Zoroastrians in Persia, nor-do I on 
astrian Vendidad are largely sanitary the other hand forget the presence of 
in their origin and these rudimentary Hindu, Mohammedan, and European 
attempts at sanitation take on a new influences on the Parsis in India ; it is 
complexion when viewed in the light a task for some thoroughly versed 
of modern hygienic theories. scholar to estimate the relative extent 

2 Cf. Vd. 5. 60. and proportion of these outside influ- 

3 I am not unmindful of the exist- ences in each case, 
ence of Mohammedan influence upon 



400 ZOROASTRIAN RELIGIOUS CUSTOMS AT YEZD 

*/ the same as their own 4 good thoughts, good words, good 
deeds ' ; and they will likewise do well to promote, as hereto- 
fore, all causes that may tend to improve the condition of 
these Persian brethren and to enable them to live up to 
the standards of the ancient creed which they both possess 
in common. 



> 



CHAPTER XXV 
FROM YEZD TO TEHERAN 

* They have ridden the low moon out of the sky, 
Their hoofs drum up the dawn.' 

KIPLING, The Ballad of East and West, 39. 

IT was the morning of May 13 when I bade adieu to Yezd 
and its Zoroastrian community and to my English hosts, and 
started on the journey northward to Teheran. The distance 
to be covered was about 375 miles over a trail through plains 
alternating with deserts which now and then encroach on the 
track if the hills on either side do not hold them back. Some 
day the journey will be made in seven hours by an Occidental 
express-train, but it took me seven days to accomplish the 
weary march, most of it on the back of animals only less tired 
than myself. 

As I mounted my horse at the door of the Mission and rode 
out through the gates of Yezd into the desert, I was warned 
that, if a heavy sand-storm should break, I was to take my 
bearings by means of the compass and head toward the nearest 
haven of refuge, as the path might be wholly obliterated by 
the sand. There was happily no occasion to necessitate this 
measure, and as the horses were good^ I enjoyed 4 chaparing ' at 
a brisk gallop for a number of miles. 

Safar, all this while, kept up a spirited conversation with 
the little postilion (jhdgird-cKdp&r), a bright lad who did 
not allow the horses to lag, but kept whipping them up from 
time to time with a thin metal chain that served as a whip, 
so that we reached before long the vicinity of the Gabar dakh- 
mah, which crowned a high sand-dune in the distance. Here 
2o 401 



402 FROM YEZD TO TEHERAN 

I halted for an instant to take a photograph of our youthful 
postilion holding in his hand the chain, which he sold me as 
a memento at a price (thanks to Safar's dealings) far below 
the figure which my farangl extravagance might have offered 
him. 1 

The hamlet of Hojatabad, about twelve miles from Yezd, 
was the first station for changing horses, and I rested in its spa- 
cious caravansarai for about an hour, from one to two in the 
afternoon. My luncheon consisted of raw eggs (tukhmahd 
na pukhtaJi) ; these formed my staple food when 4 on the road ' 
in Persia, because I always found them good and nutritious, and 
I could save time, when hastening to make long stages of 
twelve or thirteen hours in the saddle, by not even waiting to 
have them boiled. For dessert on this occasion I had some 
sharbat, which was sickishly sweet in taste, but was served in 
an antique brass saucer engraved with a tracery so artistic in 
design that I bought the dish as a curio and had Safar wrap up 
the sticky receptacle and place it in his capacious saddle-bags. 
Sundown found us in Maibud, which Yakut and other early 
Oriental geographers, who wrote before the thirteenth century, 
locate at a distance of ' ten f arsakhs from the borders of Yezd 
and the same distance from Akdah.' 2 Like most of the an- 
cient farsakh measurements, these numbers have remained un- 
changed and are still given as the respective distances to these 
places when reckoning the pay by farsakhs for post-horses. 
Even if we do not go back to the Persian and Arabic geog- 
raphers, we have more or less precise records of the route 
dating from the time when Marco Polo traversed a part of it 

1 At the time of the purchase I whip with leather thong and wooden 

thought that this thin metal chain handle, one of which I had purchased 

might be tfce modern representative near the Tomb of Cyrus, and that 

of the ancient aspahe aStra, * horse- the chain represents rather the sraoSo- 

goad,' of the Avesta (Vd. 4. 19; 6. Carana, as seen also in the chain 

6 ; 14. 2, etc.), but I have since become whips at Modern Merv in Turkis- 

convinced that the aspahe aStrd is tan. 

represented rather by the ordinary 2 Yakut, p. 655 ; cf . also p. 404. 




SHOEING A DONKEY 




THE ZOROASTRIAN DAKHMAH AT RBI NEAR TEHKRAN 



LONG MARCHES 403 

in the latter part of the thirteenth century. 1 The Italian friar 
Odoric of Pordenone rode over it from Kashan to Yezd early 
in the fourteenth century (about 1325), 2 and in the latter part 
of the fifteenth century (1474) Josafa Barbaro, the Venetian 
envoy to the court of Usun Cassan, describes Kashan and 
Kum, two of the most important towns on the line. 3 

The second day's march, as my diary shows, was a plodding 
ride of fourteen hours, with two brief breaks before the goal, 
fifty-six miles distant, was reached. A mid-day halt on this 
journey was made for an hour at Akdah, or Agdah, which is 
described by Yakut as 4 a town on the borders of the desert of 
Yezd.' 4 Somewhere in the hills in this vicinity there is said to 
be a shrine sacred to the memory of Banu-i Fars, or Khatun 
Banu, the mother, or more probably the daughter, of the last 
Sasanian monarch Yazdagard, with whose death the line of 
Zoroastrian rulers in Persia came to an end. 5 

In this same region, in the Zoroastrian village of Sharafabad, 
in the district of Ardakan, there is an old mud-walled Tower 
of Silence (dakhmah), and the story goes that seven charitable 
sisters built seven different dakhmahs at various points on the 
plain of Ardakan, and the sites of these structures are indi- 
cated by mounds of earth which are still pointed out by the 
aged Parsis of Sharafabad. 6 There is also a modern Zoroastrian 
dakhmah between Sharafabad and Mazra-i Kalantari, in Arda- 
kan; it was erected by Manakji Limji Hoshang Hantaria, the 



1 See Marco Polo, ed. Yule, 1. 88 ; that was to quench her thirst, and the 
cf . also Sykes, Ten Thousand Miles in consequent traditional sacrifice of 
Persia, p. 155. cows on the spot by Zoroastrians (now 

2 See Odoric de Pordenone, ed. Cor- discontinued), is recorded by Karaka, 
dier, p. 41, Paris, 1891. History of the Parsis, 1. 85-87 ; Sykes, 

3 See Josafa Barbaro, Travels in Ten Thousand Miles in Persia, p. 156. 
Persia, 49. 73. 6 For this information regarding 

4 Yakut, pp. 404, 555. The older the dakhmahs I wish again to thank 
form of the name is generally given as Khodabakhsh Bahrain Rai's of Yezd. 
' Ukdah in the Arab geographies. On the Zoroastrian village of Shara- 

6 The legend of her flight and the fabad, see Sykes, Ten Thousand Miles 

cow which kicked over the pail of milk in Persia, p. 156, n. 1. 



404 FROM YEZD TO TEHERAN 

same person that built the new tower near the dakhmah of 
Jamshid at Yezd. 1 

The survivals of Zoroastrianism noticeable throughout this 
entire district were evidently observed centuries ago by Josafa 
Barbaro when he halted at a town called ' Guerde,' the identity 
of which is not clear, although its general location is undisputed. 
He says : 

'Thense [from Yezd] ye go to Meruth, a little towne, and twoo 
daies io r ney further is a towne called Guerde, in the which there 
dwell certein men called ABRAINI, which in myne opinion either 
be descended of Abraham orells haue Abrahams faith, and they weare 
longe heare.' 

The association of Abraham with Zoroaster by the Mohamme- 
dans is a familiar fact, 2 and the identity of the two religious 
leaders is assumed by many of the Mohammedan Se'ids in this 
district, who are really converts from Zoroastrianism to Islam 
and regard the Parsis as their kinsmen. 3 

My journey continued for some distance along the line of 
the Persian telegraph, whose posts and wire became welcome 
company as a reminder of civilization and an assurance of 
safety in case of accident. The feeling of security is marked 
when one is within reach of the wire, not because a station is 
near, for they are miles apart, but because, if anything hap- 
pens, the traveller can get assistance by simply cutting the 
wire, as some one will be sent from the nearest station to dis- 
cover the cause of the broken current. The element of civili- 
zation also comes in strongly, because one meets with hospitality 
from the few European telegraph-employees along this benighted 
route, and I felt grateful that afternoon for an hour's rest and 
a cup of tea at the temporary camp they had set up, some miles 
south of Nu-Gumbaz. Darkness was falling and a storm ap- 
proaching when I reached the chdpdr-khdnah at Nu-Gumbaz 
a dreary and desolate place. I felt too tired to wait for any- 

1 See p. 397, above. 8 For this latter point, see Sykes, 

2 See p. 343, above ; p. 438, below. Ten Thousand Miles, p. 156. 



ZOROASTRIAN SURVIVALS AT NAIN 405 

thing to be cooked for supper, but I enjoyed a hearty meal of 
thirteen raw eggs (adding the extra one to the dozen for good 
measure) and then threw myself on my camp-cot for a short 
night's rest. 

My foot was in the stirrup again at 3.45 A.M., and after a 
stretch of six short farsakhs, or eighteen miles, 1 part of them 
over plain and desert, we filed slowly into the town of Nam, 
called ' Nairn ' by Josafa Barbaro, who found it ; evill enhabited, 
not exceading V c houses.' 2 A century earlier (1340) the Per- 
sian geographer Mustaufi described it as surrounded by 'a 
rampart 4000 paces in circumference,' 3 and Yakut, a hundred 
years before him, spoke of the theological reputation of Na'in, 
as having produced a number of eminent students of the Koran 
and scholars versed in Mohammedan lore. 4 I was struck by the 
evident antiquity of the place, and I find Nam mentioned by the 
Arab geographers in the ninth and tenth centuries, 6 but I have 
not yet been able to trace its history back to the Sasanian pe- 
riod, although the designation of the ancient citadel as Kal'-ah-i 
G-abar, 4 Gabar Castle,' and the retention of Zoroastrian names 
in some of the local designations point to a very early age. 6 S 

All went well on the journey until Neistanak (Barbaro's 
4 Naistan ') was reached on the same day, two hours after noon. 
There I found that the outgoing post, north and south, had 
taken all the horses of the chapar-khdnah, and that not even a 

1 The farsakh, or ancient parasang, De Goeje, 3. 51 (mere mention) ; Ista- 
a variable measure derived from con- khri, 1. 100, 135, 136, 155, 202, 229, 
venient stages in the day's march of a 231, 232 ; Ibn Haukal, 2. 182, 203, 
caravan, differs considerably in differ- 204, 289, 291, 296. 

ent parts of Persia, especially accord- 6 On this latter point see Sykeg, op. 

ing to the nature of the country to be cit. p. 157. It is even possible that the 

traversed. In the region of Yezd the Kal'ah-i Gabar, ' Castle of the Gabars,* 

farsakhs are short. may represent Marco Polo's Cola 

2 Josafa Barbaro, Travels, ed. Hak- Ataperistan, ' Castle of the Fire-Wor- 
luyt, 49. 82. shippers,' whence one of the three Magi 

8 See Barbier de Meynard, Diet. is said to have come to worship the 

geog. de la Perse, p. 561. infant Christ, as I have pointed out in 

* See Yakut, p. 561. JAOS. 26. 79-81, but Kashan has a 

6 See, for example, Mokaddasi, ed. stronger claim, as I shall state below. 



406 FROM YEZD TO TEHERAN 

mule could be hired in the neighborhood. There was conse- 
quently nothing to do but wait until one of the post-relays 
returned, and we were thus delayed until midnight. We 
started at 1.00 A.M., but the horses were so tired that we could 
not urge them off a walk, and since no fresh relay was found 
at the next station, we were obliged to use the same jaded 
animals all the way to Ardistan. This single day's journey of 
forty miles occupied sixteen hours ! l 

Ardistan is a flourishing town, abounding in streams and 
orchards, and counting some twelve thousand inhabitants, a 
number notably larger than the estimate given in the fifteenth 
century by Josafa Barbaro, who calls 4 Hardistan a little towne 
that maketh a V c howses,' 2 but more in accordance with the 
statement made in the tenth century by Mokaddasi (Makdasi). 
The latter says : ' Ardistan is larger than the other cities in the 
region of the desert. It is well populated, and has fine bazaars 
and numerous mosques. There are many sages and learned 
men residing there. The region of Ardistan abounds in 
white flour, whence it derives its name (arc?, 'white flour,' 
and stdn, 'place').' 3 

Historically the place is of considerable interest to the student 
of Zoroastrianism, as may be gathered from information in the 
older Mohammedan writers. Ibn Rostah (c. 900) speaks of Ar- 
distan as a fine city, and says that Anushirvan (i. e. the Sasanian 
king Chosroes I, A.D. 531-579) was born there. 4 Istakhri (951) 
states, ' Ardistan is a walled city, every quarter of which is 

1 Two days was the time occupied ed. De Goeje, Bibl. Geog. Arab. 3. 390. 
"by Josafa Barbaro in the fifteenth cen- The etymology is not correct; see be- 
tury, for he says : 'From thense [i.e. low, p. 407. In contradistinction to 
from Ne'istanak] two other daies io r - Mokaddasi's praise of the people of 
ney is Hardistan, a little towne that Ardistan I may cite a Persian writer of 
maketh a V c howses ' (ed. Hakluyt, 49. the seventeenth century, Sadik Isfahan! 
83). (p. 62), who reports that * the people of 

2 See preceding note, and for the this place are, it is said, prone to exces- 
present estimate of 12,000, cf. Sykes, sive anger and violence.' 

op. cit. p. 167. 4 Ibn Rostah, ed. De Goeje, 7. 153, 

8 Freely rendered from Mokaddasi, 275. 



ARDISTAN AND ITS ANCIENT REMAINS 407 

fortified, and there are in it some ancient traces of such Magi- 
ans as Anushirvan and Khosru, and it has a large and wonder- 
ful underground aqueduct (kandt) ; its people are men of culture 
and letters, and they have a knowledge of the traditions of 
Islam.' 1 Mustaufi (1340) says that ' Isfendiar (the son of 
Zoroaster's patron, King Vishtasp) built at Ardistan a fire- 
temple which enjoyed great fame during the ages of idolatry 
and attracted many pilgrims.' 2 Yakut expands Istakhri's 
account by speaking of the vaulted roofs of Ardistan and its 
beautiful gardens, and naming the distinguished men that 
the city had produced. 3 The name Ardistan argues for the an- 
tiquity of the place, because ardastdna, or more precisely arda- 
stdna dthangaina, was a designation for a stone construction in 
Achyemenian times and was applied, for example, to the win- 
dows in the Palace of Darius at Persepolis. 4 Traces of the 
wonderful water-courses, referred to by Istakhri, were still 
plainly to be seen, as I remember, when I walked through the 
town between rows of gardens in full bloom, to which Yakut 
alludes. Careful investigation would probably give more infor- 
mation also regarding the fire-temple which Isfendiar is said 
to have founded here, and Ardistan should be added to the list 
of places already recommended as inviting sites at which to 
pursue archaeological investigations. 

My attention at the moment had unfortunately to be de- 
voted more to the question of transportation than to archaeology, 
so I was obliged to direct my steps to the telegraph station, 
which was situated in an attractive garden, filled with blossom- 
ing fruit trees, and was in charge of a Persian telegraph opera- 
tor. Although he was not able to procure a post-relay for me, 
he kindly placed his own horse at my disposal and provided 
donkeys to make up my caravan so that I might start at niid- 

1 Istakhri, ed. De Goeje, Bibl. Geog. p. 22, n. 1 ; cf. also Le Strange, JRAS. 
Arab. 1. 202, n. 1. 1902, p. 243. 

2 Mustaufi, Nuzhat al-Kulub, cited 3 Yakut, pp. 22-23. 
by Barbier de Meynard, Diet. geog. 4 See p. 315, above. 



408 FROM YEZD TO TEHERAN 

night, besides wiring a message in advance to have horses meet 
me at Khalatabad on the morrow. As I returned to the post- 
quarters, a messenger, looking like a mounted arsenal, galloped 
in on a fine horse, with a rifle across his shoulder and two 
enormous holster-pistols on either side of his saddle. I thought 
that he might possibly allow me to hire his horse to add to my 
limited cavalcade for the next stage of the journey, but his 
errand was urgent, and he dashed onward over the plain, leaving 
us to watch the sunset and retire. 

After four hours' sleep I gave the signal at 1.15 A.M. for my 
caravan of donkeys and the horse to start, and the company of 
half a dozen beasts of transport moved slowly out on the plain 
and gradually into the desert. The gait was a snail's pace, but 
I had long before become accustomed to taking naps in the sad- 
dle, waking from time to time to watch the progress of the 
stars, particularly glorious in Persia, and to count the long hours, 
till the brightening hues of dawn ushered in the day with a 
burst of sunlight. Forward through the sand trudged our 
little cavalcade until, about the time when one would ordinarily 
be breakfasting in America, we were approaching the small 
ttown of Moghar, on the edge of the desert, having been six 
hours already in the saddle. The outskirts of the settlement 
were rich in grain-fields, and the gleaners were already at work 
with the sickle. The Persian sickle curves far more beyond the 
semicircle than does our own, so that the blade resembles an 
immense steel hook. In using it, the gleaner squats on the 
ground in Oriental fashion, gathers an armful of the tall barley, 
deftly cuts it, and lays it down to be bound afterwards into a 
sheaf. 

The stretch of desert from Moghar to Khalatabad seemed 
interminable at our slow pace. The sun beat down pitilessly, 
and I could feel its scorching rays penetrating through the 
white cotton covering that I wore over iny hat. The sand was 
dazzling and in many places was encrusted with a coating of 
salt that looked like ice or snow. From time to time we 



THROUGH THE DESERT TO KHALATABAD 409 

encountered a row of sand hillocks that looked as if some gigan- 
tic mole had been burrowing beneath the surface of the earth. 
These high-heaped mounds, which are a familiar sight in the 
arid districts of Persia, are made from the sand thrown up 
around the mouth of the kandts, or deep wells, dug at intervals 
and joined by a succession of underground tunnels which carry 
the water sometimes several miles. Wherever moisture had 
gathered and had been evaporated by the sun, the sand was 
baked into huge cakes like clay, across the cracks of which 
myriads of lizards darted, while every suggestion of humidity 
on the desert gave rise to mirages so deceptive that it was often 
impossible, a few feet away, to tell whether we were looking 
upon a pool or not. The unceasing stretch of sand made the 
stage seem endlessly long. Now and then I varied the monot- 
ony by giving the horse to Safar to ride and taking one of the 
donkeys ; my particular choice was a beast that was easy to 
ride, but had a provoking habit of lying down in the sand at 
the most unexpected moments. 

Khalatabad was reached at last after a ride of fourteen hours, 
and we found to our joy that the chdpdr horses had arrived. 
After an hour's rest I vaulted again into the saddle for a ride 
which proved to be one of the keenest pleasures I experienced 
in Persia. The horse assigned me was a splendid animal, one 
of the three truly fine mounts I had ridden since leaving Uru- 
miah (and I had ridden more than fifty), but it was evident 
that he was not used to a Western bit, for he took the iron at 
once between his teeth and ran away with me, galloping for 
nearly three miles before I could bring him down. The plain 
was as smooth as a cinder path, his mettle was superb, and the 
speed delightful after the crawling pace at which we had pro- 
ceeded all day. Again and again he became unmanageable and 
ran, leaving the rest so far behind that I had to gallop back 
again for fear of losing the servants and my baggage. After 
three hours of this exhilarating sport, which the horse enjoyed 
as much as I, he suddenly bolted, being frightened by a flock 



410 FROM YEZD TO TEHERAN 

of sheep running in confusion before an approaching thunder- 
storm ; but I was in too .good training to be unseated, and we 
galloped at full speed beneath the bald khdnah of the post-house 
at Abuza'idabad at 7.30 P.M., just as the rain began to pour 
down. The day had been one of almost incessant riding for 
seventeen hours, and the journey, which ordinarily takes four 
days, had been made in one. My mettlesome steed, however, 
paid for his speedy gait, for I found his high spirits gone the 
next morning, and sheer fatigue made him quite docile by the 
time we caught sight of the blue domes of the mosques at 
Kashan that rose with a glimpse of color over the brown plain. 

Kashan, pronounced in Persian as Kdslidn, is flanked by 
mountains and hills on the south, west, and northwest, but on 
the other sides it lies open to the plain. The city looks low 
and level as one approaches it, but the outline is broken by the 
characteristic vaulted mud roofs of the houses, the domes of 
the mosques and madrasahs, and by a lofty minaret, over one 
hundred feet high (seen in the left of the picture) ,* which looks 
in the distance as if a modern factory with a high chimney had 
been set up to give occupation to the inhabitants, who number 
over seventy thousand. 

The history of Kashan, like that of its rival city Kuin, is 
wrapped in obscurity, but Firdausi assumes that it was in 
existence in the time of Kei Khosru, the legendary king who 
is supposed to have reigned about eight centuries before the 
Christian era, for the great warrior Kamus is frequently alluded 
to as the hero ' of Kashan.' 2 Persian history also records that 
Kashan and Kum furnished about twenty thousand men to the 
army that fought in vain against the Caliph Omar, at the time 

1 This is the leaning tower of Zein from Mme. Dieulafoy's work), and the 

ad-Din, a picture of which is given valuable account of the province of 

by Mme. J. Dieulafoy, La Perse, Kashan by Houtum-Schindler, East- 

p. 198, Paris, 1887; see also Landor, ern Persian Irak, pp. 109-118, Lon- 

Across Coveted Lands, 1. 263, and com- don, 1897. 

pare the description of Kashan by 2 Firdausi, Shah Ndmah, ed. Vul- 

Curzon, Persia, 2. 12-16 (who repro- lers-Landauer, 2. 870, 918, etc., and 

duces the picture of the leaning tower transl. Mohl, 3. 1, 58, 97, etc. 




DOMED ROOFS OF KASHAN 




KASHAN 



KASHAN 411 

of the Arab conquest. 1 A common report assigns the building 
of the city to Zobeidah, the wife of Harun al-Rashid (A.D. 800), 
but this is incorrect, as it was in the case of Tabriz, already 
cited, 2 although Zobeidah may have rebuilt the city. 

Mustaufi's brief but interesting account of the city of Kashan, 
which was written about the year A.D. 1340 and based on older 
sources, is in substance as follows : ' This city was built by 
Zobeidah, the wife of Harun al-Rashid. The heat is intense 
here in summer, but the winters are very pleasant. The water, 
which is not very abundant, is brought from a reservoir at Fin, 
which is fed by the Kuh Rud. 3 Rain water is also collected in 
cisterns by the inhabitants. The people of Kashan are intelligent 
and well informed, and belong to the Shiite sect of the Moham- 
medan faith ; but the inhabitants of the eighteen villages in the 
vicinity, on the contrary, are Sunnites. The melons and figs of 
Kashan are held in esteem.' 4 The European traveller Josafa 
Barbaro, a century later, speaks of it as the ' well enhabited citie 
called Cassan, wheare for the moste parte they make sylkes and 
fustian in so great quantitie that he who wolde bestowe x ml 
ducates in a daie may find enough of that merchaundise to bestow 
it on. It is about iij myles in compasse, walled, and w'houtf oo rth 
hath faire and large subvrbes.' 5 Barbaro resided in Kashan for a 
time and was visited here by his fellow-countryman Contarini, 
the Italian traveller, who arrived at 'Cassan' on October 25, 
1474, and calls it a finer city than Kum ( 4 Como '). 6 

All modern travellers speak of the heat of the place and 
re-iterate the statement that the town is noted for three things : 

1 See Ouseley, Travels in Persia, supplied with water, were once a 
3. 3, n. 3, and 3. 100. Ouseley cites favorite place of resort for the Persian 
as his authority ' The Book of Con- kings, including Shah Abbas and Fath 
quests,' a chronicle history (tarikti) Ali Shah ; but Fin is now deserted, 
by Ibn Aasim of Kufah, who flourished See Curzon, Persia, 2. 12 ; Landor, 
in the eighth century A.D. Across Coveted Lands, I. 265-266. 

2 See p. 39, above. 4 Mustaufi, cited by Barbier de Mey- 
8 Fin is located on the mountain nard, Diet. geog. p. 434, n. 1. 

slopes about five miles southwest of 5 Josafa Barbaro, Travels, 49. 73. 

Kashan. Its garden and groves, well 6 Contarini, Travels, 49. 129. 



412 FROM YEZD TO TEHERAN 

its manufacture of porcelain tiles, brasswork, and silk, its black 
scorpions, and the cowardice of its inhabitants. The latter 
stigma is constantly put upon the Kashanis by the other Per- 
sians, 1 but I was glad not to have any occasion to judge whether 
it is true. The silk, however, I found good, judging from the 
samples brought me by a vender, who followed me from the 
bazaar to the telegraph quarters and showed the same enter- 
prising spirit and business energy for which the Kashanis were 
praised hundreds of years ago. 2 As for the scorpions, I saw 
some formidable specimens, as the manager of the telegraph 
station was making a collection of them. 

I regretted that during my short halt at Kashan I did not 
know of a Western legend which connects this city with the 
Three Kings of the Orient who went to Jerusalem to worship 
the infant Christ. It is a known fact that a majority of the 
Church Fathers agree in regarding Persia as the native country 
of the Wise Men, without expressly locating their place of 
origin. 3 The Italian traveller Marco Polo (1272) and the 
Venetian envoy Odoric of Pordenone, who traversed this 
route, the latter about 1320, record traditions that definitely 
attach the Three Wise Men to certain cities. Odoric expressly 
says that Kashan, or ' Cassan,' as he calls it, was 4 the city of 
the Three Kings ' and that these worshippers set out from 
there to Jerusalem, which they reached by divine aid in thir- 
teen days. 4 The passage in the account of Odoric de Por- 
denone reads as follows : 

' DE LA CITE DE CASSAN. De ceste cite m'en alay vers la grant 
Inde par mer. Si vins par maintes journees a une cite des trois roys 
qui firent offrande a Jhesu Crist nouvel ne. Et appelle on ceste cite 
de Cassan, cite royal de grant honneur mais Tartre Font moult de- 
struite. De cette cite de Cassan jusques en Jherusalem a plus de L 

1 On this point, see Sykes, Ten 8 See my article in JA OS. 26. 
Thousand Miles in Persia, p. 168. 79-83. 

2 See Barbier de Meynard, Diet. 4 See Odoric de Pordenone, ed. Cor- 
geog. p. 434, n. 1. dier, p. 41, Paris, 1891. 




KUM 




PERSIAN GLEANERS 



A LEGEND OF THE THREE WISE MEN 413 

journe'es dont on puet clerement appercevoir que ly troy Roy qui 
de ceste cite de Cassan furent en xm journees amene en Jherusalem 
par vertu divine et non humaine. En ceste cite a moult grand habon- 
dance de tous biens, de pain, et de vin et de toutes autres choses.' 

According to a legend given by Marco Polo, two of the kings 
came from 4 Saba ' l and 4 Ava,' both of which places are located 
about fifty miles southwest of Teheran, and the third is said to 
have come from ' a place three days' journey ' from Avah, and 
Marco Polo states that he ' found a village there which goes by 
the name of Gala Ataperistan (i.e. Kal'ah-i Atashparastdri), 
which is as much as to say 4 The Castle of the Fire- Worship- 
pers.' And the name is rightly applied, for the people there do 
worship fire.' In an article entitled The Magi in Marco Polo I 
have given various reasons for identifying the so-called ' Castle 
of the Fire-Worshippers ' with Kashan, which Odoric mentions, 
or a village in its vicinity, the only rival to the claim being the 
town of Nam, whose Gabar Castle has already been mentioned 
above, 2 but I may refer to the discussion in the article without 
repeating it here. 3 I should like to have had time to visit 
Gabarabad, a deserted town on the Isfahan road about twenty 
miles distant from Kashan, as its name (lit . ' Gabar Town ') 
shows that there was once a settlement of fire-worshippers in 
its vicinity, and the ruins of a magnificent caravansarai are 
still to be seen there. 4 In Kashan to-day there are some 
Zoroastrians, for the statistics which I gathered in Persia 
show that about forty-five of them do business in this city. 6 

Most of my short stay in Kashan was occupied in purchasing 
articles which I needed, and in securing some sort of a vehicle 
to convey me to Kum. I had been riding uninterruptedly for 
weeks and was very tired and anxious to procure some means 
of conveyance on wheels. With the aid of the manager of the 

1 On 'Saba' (= Savah) cf. also quart, Untersuchungen zur Geschichte 
Sykes, Ten Thousand Miles, p. 264. von Eran, 2. 1-19, Leipzig, 1905. 

2 See p. 405, above. 4 Cf. Bishop, Journeys in Persia, 1. 
8 See my article, The Magi in Marco 232. 

Polo, JAOS. 26. 79-83, and cf. Mar- 5 See p. 425, below. 



414 FROM YEZD TO TEHERAN 

telegraph station I was able to hire a lumbering cart ; it had no 
springs, but it could at least be driven, and I was able to lie 
down as we travelled. Four horses were hitched to it, and 
away we jolted for hours over ruts and hollows, till we caught 
sight of Kum towards nightfall and were passing through its 
bazaars and shops about as they were closing. 

Kum is little less than a city of mosques, minarets, madra- 
sahs, and corpses, for, next to Meshad, it is the most famous 
burial-place in Persia. Its special sanctity is due to the fact 
that it is the proud possessor of the shrine of Fatima, sister of 
Imam Riza, the eighth Imam. She was interred here in 
A.D. 816, arid honors are accorded her that Mohammedanism 
rarely vouchsafes to a woman. Kings have chosen the city as 
a final resting-place for their bones, and the great Kajar mon- 
arch Path Ali Shah is among the Persian monarchs buried 
there. Burial near Fatima's shrine is in fact almost equivalent 
to a viseed passport for heaven, although Kum cannot quite 
rival Kerbela and Meshad in this respect. 1 

The city did not attract or interest me enough, however, to 
make me wish to stay longer than was necessary, and I was 
content to remember it by two photographs which I purchased 
afterwards. The mahmdn-khdnah, or hostlery, at which I put up 
showed signs that we were approaching a more civilized region. 
The building had a veranda, a large sleeping-room furnished 
with a table and a bed, and there was a kitchen, where I found 
a saucepan and tried the experiment of making an omelet, 
which proved quite a success. My most vivid impression of 
the mahmdn-khdnah, however, is associated with my first taste 
of arac. I needed spirits for medicinal purposes, and the ser- 
vants could procure only arac in the bazaar. I should not like 
to repeat the dose of this particular brand, for the taste re- 
sembled what I could imagine a combination of gin, whisky, 
and furniture-varnish might be ! 

1 For valuable details concerning Eastern Persian Irak, pp. 66-77, 
Kum, consult Houtura-Schindler, London, 1897. 



THE CITY OF KUM 415 

To transport me to Teheran I found that a so-called ' dili- 
gence ' could be procured as a substitute for the wretched cart 
I had used all day. I felt happy when I heard at midnight the 
announcement that it was ' ready ' (hdzir), but in spite of this 
assurance another hour was wasted before the horses were 
hitched and the rumbling vehicle rolled to the door. It was 
a ramshackle conveyance, threatening each moment to fall to 
pieces, but deserving, I suppose, the respect due to advanced 
age. Despite the discomfort of my cramped position and the 
distressingly raw rainy weather, I managed to fall asleep, and 
I slept during most of the journey of twenty hours, until, about 
nine in the evening, after seven days' travelling from Yezd, we 
arrived at Teheran. A note of the itinerary of the journey 
may be of some interest. 

ITINERARY FROM YEZD TO TEHERAN 

WEDNESDAY, MAY 13, 1903 

FARSAKHS MILES 
10.10 A.M. leave YEZD 

Pass Gabar Dakhmah 
1.20 P.M. arr. 






_ 

Hoiatabad 4 

2.20 P.M. Iv. j 

5.20 P.M. arr. Maibud 6 

Time : 7 hrs. 10 min. (6 hrs. 10 min. in saddle 

+ Ihr.halt) 
Distance : 10 farsakhs = about 36 miles 

THURSDAY, MAY 14 
4.00 A.M. Iv. Maibud 

^JjaftanCorChiaftah) 6 18 

Poor horses ; very slow progress 
10.15 A.M. arr.") . . , . 
11.30A.M. Iv. }Akdah(orAgdah) 4 12 

Frequent mirages 

Halt in afternoon for nearly an hour at construc- 

tion-camp of Persian Telegraph Company 
Road mostly over a brown plain 

6.10 P.M. arr. Nu-Gumbaz 9 27 

Time : 14 hrs. 10 min. (11 hrs. 40 min. in saddle 

-f 2 hrs. 30 min. halts) 
Distance : 19 farsakhs = 57 miles 



416 FROM YEZD TO TEHERAN 

FRIDAY, MAT 16 

FAKSAKHS MILES 
3.45 A.M. Iv. Nu-Gumbaz 

Plain and desert ; frequent mirages 

8.00A.M. MT.J 

9.20 A.M. Iv. J 

Remains of an old town in the vicinity 
1.55 P.M. arr. Neistanak 6 19 

Delayed here because of lack of horses 

Time : 10 hrs. 10 min. (8 hrs. 50 min. in saddle 
-f 1 hr. 20 min. halts) 

Distance : 12 farsakhs = 37 miles 

SATURDAY, MAY 16 
1.00 A.M. Iv. Neistanak 

Moonlight ; horses very slow 
8.20A.M. arr- | Jokand 7 24 

11.45 A.M. IV. J 

No fresh horses ; 3 hrs. delay ; proceed with 

same horses 

5.00 P.M. arr. Ardistan 4 15 

Time : 16 hrs. (12 hrs. 35 min. in saddle + 3 hrs. 

25 min. halt) 
Distance : 11 farsakhs = 39 miles 

SUNDAY, MAY 17 
1.15 A.M. Iv. Ardistan 

No post relay ; make up a caravan of 1 horse 

and 4 donkeys 
7.15 A.M. arr. I _. 

9.30 A.M. iv. } M e har 

Desert 
3.10P.M. a,rr. 6 16 



4.50 P.M. Iv. j 

Excellent post-horses waiting ; fast time ; slight 

thunderstorm 

7.30 P.M. arr. Abuzaidabad (or Abuzadabad) 6 17 

Time : 18 hrs. 15 min. (14 hrs. 20 min. in saddle 

-f 3 hrs. 55 min. halts) 
Distance : 16 farsakhs = about 51 miles 

MONDAY, MAY 18 

4.30 A.M. Iv. Abuzaidabad 

7.30A.M. arr.l 6 

9.10 A.M. Iv. } Ka * han 

Hire a cart (gari) instead of post-horses 
11.10 A.M. Kasimabad (?) 
2.20 P.M. Shurab 




A VIEW OF TEHERAN 




A PERSIAN GARDEN 



ITINERARY 417 

FARSAKHS MILES 

5.15 P.M. Pasangun 

9.30 P.M. arr. Kum 12 

Time : 17 hrs. (3 hrs. in saddle, the rest in gari} 
Distance : 18 farsakhs, estimated at about 70 miles 

TUESDAY, MAY 19 
1.00 A.M. Iv. Kum 

Travel in a diligence, sleeping most of the way 
9.00 P.M. arr. TEHERAN 25 85 

Time : 20 hrs. 

Distance : 25 farsakhs = 85 miles 



Time of the journey : 7 days 

Total cost : 125 tomans (about $120) 



2m 



1 



CHAPTER XXVI 
TEHERAN, THE MODERN CAPITAL 

Where in sunshine reaching out 
Eastern cities, miles about, 
Are with mosque and minaret 
Among sandy gardens set, 
And the rich goods from near and far 
Hang for sale in the bazaar.' 

ROBERT Louis STEVENSON, 

A Child's Garden of Verses. 

4 ISFAHAN is fair, Shiraz is beautiful, but Teheran Tahrdn 
khaili khub ast Teheran is very beautiful,' this is the en- 
comium which I heard bestowed, over and over again, upon the 
modern capital of Iran as I journeyed northward from Yezd. 
During a week's stay in the metropolis I found the city suffi- 
ciently attractive to make me sympathize, in part at least, with 
the enthusiasm of the Persians, even if Teheran cannot boast 
of many of the natural beauties of such a city as Shiraz. 

East and West combine imperfectly in its mixed civilization, 
with a far greater preponderance of the Orient, as is natural. 
Landau carriages in the public square, a post-office with 
bilingual notices in Persian and French, well-equipped tele- 
graph headquarters, an imposing Imperial Bank, a so-called 
Boulevard des Ambassadeurs, along which the ministers of 
the foreign legations ride in official dress, not to speak of shops 
with European goods, two 'hotels,' a claim to the use of gas, 
and a pretense of having a jingle-bell tramway, all these 
tell something of the influence of the Occident. But all the 
rest mosques, minarets, and madrasahs, camels and caravan- 
sarais, bazaars crowded with scuffling mon and veiled women, 

418 



TEHERAN AS A CAPITAL 419 

with the survival of many a custom that seems to antedate the 
time of Cyrus are characteristics that belong to the Orient 
and make Teheran as Oriental as any capital in the East, 
although I missed those signs of national greatness which be- 
longed to the days of Persia's by-gone glory and are visible at 
Persepolis even in the ruins. 

Looked at from the historic standpoint, Teheran may be 
considered to be the inheritor of the ancient honors of Pasar- 
gadae and Persepolis, and the successor to the imperial rank 
held a few centuries ago by Shiraz and Isfahan. With the rise of 
Teheran to power, Media has been able once more to reclaim 
the supremacy she lost to Persis in the time of Cyrus, and the 
present capital occupies a site that is almost identical with the 
ancient city of Rages (Avestan Raghd, Old Persian Raga), now 
Rei, its ruined suburb, which shared with Ecbatana in antiquity 
the honors of supremacy over Iran. And yet, comparatively 
speaking, Teheran is a modern town, a city that came into 
existence less than seven hundred years ago, about the time 
when Ragha began to sink into oblivion ; its rank, in fact, as 
capital has been held only since 1788, when the present Kajar 
dynasty came into power. Of so little importance was Teheran 
at the beginning of the thirteenth century that Yakut (c. 1220) 
spoke of the place merely as 4 a stronghold, one farsakh dis- 
tant from Rei,' adding that the inhabitants lived in dwellings 
dug beneath the ground and were rebellious to all authority 
and in a constant state of warfare. 1 If then unimportant, 
the place must have grown extensively during the next four 
hundred years, as we may judge from the accounts by the dif- 
ferent European travellers who visited it during that period; 
and so important had it become by the end of the eighteenth cen- 
tury, that Agha Mohammed Shah made it his capital and 
bestowed upon it the laurels won by the overthrow of the Zend 
dynasty of Isfahan. 

The present size and appearance of the city are largely due 

1 Yakut, p. 399. 



420 TEHERAN, THE MODERN CAPITAL 

\ to the late Shah Nasr ad-Din, who devoted himself with genu- 
ine enthusiasm, after his first visit to Europe, to developing 
and beautifying his chosen seat of government. The old walls 
were torn down for the most part, the moat filled up with 
the debris, and an entirely new rampart was constructed, 
fully a mile outside the older enclosure, thus making the com- 
pass of the city vastly larger. This wall, over ten miles in 
circuit, is entered by a dozen gates, the more important of 
which are capped by gayly decorated towers with glazed mina- 
rets, whose shining tiles are seen at a long distance. 

As the city lies rather low in a sandy plain, its location is not 
commanding, but a line of hills to the north, behind which 
towers the Alborz range crowned by the magnificent crest of 
Damavand, 19,400 feet high, forms an imposing background in 
that direction, in contrast to the miles of level tracts that stretch 
southward over the territory traversed in coming from Yezd. 
The mountains temper the wind from the north in winter, 
and the foot-hills, with their 'gardens and orchards, afford 
a pleasant place of resort during the heat of summer. 

Descriptions of Teheran are so numerous that I may be al- 
lowed to confine myself merely to the principal features with- 
out going into details. 1 This is not due to alack of interest in 
the capital or to a scarcity of material at hand, but it is owing 
to lack of space for a more extended account of the place. 

A plan of the city, if we had a sketch-map before us, would 
show that Teheran is laid out in octagonal form, roughly speak- 
ing, and the heart of the town lies considerably north of the 
middle point of this walled enclosure. 2 The principal square 
in this section is called Meiddn-i Top Khdnah, 4 Arsenal 
Square,' a handsome quadrilateral, about 300 yards long and 
150 broad, with its larger side running east and west and its 
surface paved in a rough sort of way with cobble-stones. The 
central portion of this extensive parallelogram is occupied by a 

1 The description given by Curzon, ous accounts and should be consulted. 
Persia, 1. 300-353, supplants allprevi- 2 Cf. Curzon, Persia, 1. 305. 




THE SOUTHWEST GATE OF THE MEIDAN AT TEHERAN 



Ji 


i X 


J.. 


1 


^^ ^^M^^aMM| 














THE AMERICA.N LEGATION AT TEHERAN 



AVENUES AND PLAZAS 421 

large basin of water fenced in by an iron railing and lined 
on all sides by old mounted cannons. The eastern end of 
the plaza is practically devoted to the building and grounds 
of the Imperial Bank of Persia, a rather imposing white 
edifice, in Perso-European style, with an arched gateway at 
the left leading into an attractive garden, where the English 
members of the bank find an opportunity on holidays to indulge 
in tennis. Out from the north side of the great plaza runs the 
principal driveway of the city, Khiabdn-i '-Aid ad-Daulah, 
'Boulevard des Ambassadeurs,' along which are built the 
residences of the foreign legations and some of the finer houses 
of the city. On the west side of the square is the Arsenal 
{Top Khdnah, lit. 4 Cannon-House '), with quarters for the 
soldiers and some small buildings of the same character as those 
on the northern and southern sides of the parallelogram. The 
arched gateways that are raised over the half-dozen avenues 
which lead from the square are conspicuous because of the 
gaudy effect they produce by their glazed tiles and fancy deco- 
rations. The most noticeable of these portals is the one that 
guards the entrance to the Khiabdnri Almdsiah, or 4 Avenue of 
Diamonds,' leading from the southwest corner of the plaza to 
the palace, and the royal flag floats from the top of this gateway 
when the Shah is in Teheran. 

By this exit we leave the main square to visit that portion of 
the city which lies to its south and which is the most interest- 
ing section for foreign visitors, because it is the oldest and 
most characteristic, containing minor public squares, the origi- 
nal fortifications, the palace grounds, and the bazaars. The 
first point to engage our attention is a small square, outside 
the palace enclosure and somewhat to the south; it is the Meiddn-i 
Ark, ' Citadel Square,' or Meiddn-i Shdh, 4 Place de 1' Empereur.' 
Here, by the side of a large tank of water, is mounted a huge 
cannon, known as the Top-i Murvdrld, c Cannon of Pearls.' 
According to some accounts, its muzzle was originally deco- 
rated with a string of pearls, but other explanations are like- 



422 TEHERAN, THE MODERN CAPITAL 

wise given. The history of the gun is also variously related. 
But whatever the case may be, this large piece of artillery 
has now a semi-sacred character, because it affords a place 
of refuge (bast) for criminals who seek shelter beneath its 
shadow, and superstition has further endowed it with miracu- 
lous powers, even with the gift of granting children to barren 
women who touch its brass mouth. Facing one side of this 
square is a stately arched portal, the Nakdrah Khdnah, 4 Band- 
Tower,' or 'Hall of the Royal Music,' from whose chambers sun- 
rise and sundown are signalled by the same noisy accompani- 
ment that I have described in speaking of Urumiah and Isfahan. 1 
The second place of interest to the south of the Top Meidan 
is the area of the old fortified enclosure, nearly a quarter of a 
mile square, known as the Ark, or 4 Citadel,' within whose mud 
walls is located the Palace of the Shah. It is needless to 
say that the royal residence and the various buildings and 
grounds which belong to it, the courtyards, pavilions, fountains, 
and gardens are a source of pride to the imperial heart. The 
Museum of the palace contains, among other objects of interest, 
the sword of Tamerlane and the mail-coat of Shah Abbas, as 
well as a priceless collection of crown jewels, together with 
much that is tawdry, according to the judgment of those best 
entitled to speak on the subject. The most often described 
among the treasures, because the handsomest from the artistic 
point of view, are the magnificent jewelled globe a geo- 
graphical study in emeralds, diamonds, and turquoises and the 
famous Peacock Throne (Takht-i Td'us), which is said to have 
been brought from Delhi in India by Nadir Shah, the Great 
Mogul, about the middle of the eighteenth century ; but, ac- 
cording to Lord Curzon's claim, it is not the original Peacock 
Throne, but was made for Fath AH Shah, more than half a 
century later than Nadir Shah. 2 

1 See pp. 104, 267, above. details, as I neglected the ordinary 

2 See Curzon, Persia, 1. 317-322, 'sight-seeing' duty of a visit to the 
to whom I must refer the reader for palace and museum. 




A GRADUATING CLASS OF THE AMERICAN SCHOOL FOR BOYS AT TEHERAN 




A PERSIAN FAMILY 



THE ROYAL RESIDENCE AND CHIEF BUILDINGS 423 

The chief bazaars of the city are located also to the south 
of the Top Meidan. They are similar to the vaulted, covered 
structures that I have already described in other cities, with 
the customary shops, booths, mazy passages, and courtyards for 
caravans, but all on a larger scale than elsewhere in Persia. 
The opportunities which I found in them for shopping were ex- 
cellent, but I sought in vain for one object which I wished to 
purchase ; it was a flag with the Persian emblem of the Lion 
and the Sun. The flag has little significance in Persian patri- 
otism ; for that reason it is not commonly on sale, so I had 
to have the banner made to order. I may add, however, that 
it was handsomely painted, and it now hangs in my study 
as a memento of my journey to Iran.. 

Regarding, the other native edifices in the older part of the 
city there is little to say. The mosques of Teheran are of 
minor importance, considering that they are in the capital, and 
none of them can compare in sanctity with the shrine of Shah 
Abdul Azim near its ruined suburb, Rei. There are a num- 
ber of madrasahs, or religious colleges, and several educa- 
tional institutions on royal foundations, including the Shah's 
College, which is supported by funds from the Crown and 
employs European instructors as well as native teachers ; but, 
although this institution furnishes free instruction, clothing, 
and food, many young Persians go to the schools established 
by the foreign missions, and over a hundred are in attendance 
at the American School for Boys in Teheran. 1 

As we return from the older part of the town to the south- 
east entrance of the chief Meidan, we pass the large building 
occupied by the Indo-European Telegraph Company, where 
there is an opportunity to send a communication home by 
cable, and then we again enter the main square. Crossing 
to the northwest of the great quadrilateral, we leave it by 
the gateway that arches a street leading to another grand 

1 See Sixty -seventh Annual Report the Presbyterian Church in the V. S. A. 
of the Board of Foreign Missions of p. 238, New York, 1904. 



424 TEHERAN, THE MODERN CAPITAL 

enclosed area, the largest, although not the most important, 
plaza in Teheran. It is the city's great parade-ground, more 
than a quarter of a mile long and nearly as wide, and is called 
Meiddn-i Mashk, 4 Drill Square.' This vast area is one of the 
largest enclosed grounds for manoeuvring that there is in the 
world, 1 and here the Shah's troops go through their military 
exercises, having been trained in Occidental tactics by Euro- 
pean drill-masters. But this immense Champ de Mars is com- 
paratively little used, and it served as an admirable playground 
for boys, when I saw it, or as a place where stray animals that 
had died from disease or some other cause were allowed to lie 
until devoured by dogs. 

The northernmost section of Teheran is largely European, as 
I have already intimated. A short distance from the great 
parade-ground there are to be found the grounds and buildings 
of the American Presbyterian Mission, with its dozen or more 
laborers in evangelistic, medical, and educational work. From 
the Mission it is an easy walk to the English Hotel. This small 
hotel is the place where foreigners usually stay when on a brief 
visit. It is conveniently located near the principal thorough- 
fare, the Avenue of the Ambassadors, along which we drive 
past the various diplomatic residences. The Legation of the 
United States is pleasantly situated, and its grounds are attrac- 
tive and have been occupied for about twenty years, as the 
Legation was established in 1883. As I passed through its 
gateway and entered the enclosure, I remember the thrill 
which I felt when I raised my hat to salute the Stars and 
Stripes, and I was welcomed by our American Minister, Mr. 
Richmond Pearson. To our Secretary of Legation, Mr. John 
Tyler, who has resided in Teheran for more than thirty years, 
I was indebted, besides other favors, for the privilege of a 
visit to the Persian Minister of Foreign Affairs, whose two 
sons I had met at Berlin and St. Petersburg, where each 
occupies a position as Minister from the Persian Court. 

1 Pekin alone can claim a larger one, if I understood my informant aright. 




THE PARSI SCHOOL AT TEHERAN 




ZOROASTRIAN SCHOOLBOYS AT TEHERAN 



V 



THE AMERICAN LEGATION 425 

I found in him the grace of manner and polished behavior 
that belongs to the cultivated Persian, for a gentleman is a 
gentleman all the world over. There was no oppressive for- 
mality, the conversation proceeded easily, aided by Mr. Tyler's 
skill, and turned eventually to the subject of travelling in 
Persia. My host expressed an interest in my journey, espe- 
cially in my experiences on the Behistan Rock. He requested 
me to pronounce some of the Ancient Persian words in the 
inscription of Darius in order to compare them with the Modern 
Persian form, and we conversed also about the platform of 
Persepolis, a magnificent representation of which in tapestry 
completely filled one wall of the large room in which we were 
sitting. Sweetmeats, tea, and cigarettes were served, and a 
gracious word at parting completed the visit. 

Because of my interests and because of the many associations 
of Zoroaster's name with Rei, the suburb of Teheran, I was 
anxious to meet the Zoroastrians, and was happy that among 
the first visits I received was one from the secretary of the 
Zoroastrian Amelioration Society, Ardeshir Edulji Reporter, 
agent in Teheran for the Parsis of Bombay. This gentleman, 
whose brother I had previously known, gave me excellent oppor- 
tunities for becoming acquainted with the circumstances of his 
co-religionists at the capital, and he added material to my stock 
of information concerning the Zoroastrians throughout Iran. 
His statistics show that the number of Zoroastrians is increas- 
ing slightly and not declining. The figures are as follows : 

Yezd and vicinity between 8000 and 8500 

Kerman . approximately 2400 

Teheran 324 

Kashan 45 

Shiraz .42 

Kum 8 

Isfahan 6 

Sultanabad 4 

Total number of Zoroastrians in Persia about . . 11,000 



426 TEHERAN, THE MODERN CAPITAL 






The Zoroastrians of Teheran, taken as a whole, are in better 
circumstances than those in any other city of Persia, because 
of the more liberal conditions that prevail in general at the 
capital. The most prominent member of the community is a 
rich banker, Arbab Jamshid Bahman, whose wealth is estimated 
in the hundreds of thousands of tomans, and to whom recogni- 
tion at Court is also accorded a fact which helps the position 
of the Zoroastrians considerably, since an appeal to the Shah is 
possible through his mediation. His integrity is of the highest 
order, and his esteem, even in the eyes of the Mohammedans, 
who would naturally despise him as an 'infidel,' is so great, 
that the Persians place implicit faith in his honesty a tribute 
paid to no other native banker as yet, so far as my knowledge 
goes. This is a tribute not alone to his uprightness (arshtdt 
in the A vesta), but also to the teachings of Zoroaster, who made 
truth and honesty a watchword of the religion and whose creed 
was 4 good thoughts, good words, good deeds.' 

Arbab Jamshid called upon me shortly after my arrival 
and invited me to visit his home and his beautiful garden 
a privilege of which I availed myself twice. The garden- 
court adjoining his house in the city is laid out in the 
characteristic Persian manner with fruit trees, flowering shrubs, 
trellised arbors, pathways, and a fountain, and in addition to 
this a door opens into a small chamber on one side that is used 
as an Izashnah-Gah, or chapel, in which the rites of the Zoroas- 
trian faith are occasionally conducted by a priest. We spent 
some time comfortably seated beneath the fruit trees, chatting 
on general topics, eating sweetmeats and dates that had been 
sent by the Zoroastrians of Yezd, and drinking tea, but not 
smoking. The business quarters of Arbab's banking establish- 
ment form a part of his own residence, arid a large staff of 
clerks and assistants are occupied in conducting his affairs. 
Some idea of the number of this corps of employees may be ob- 
tained from the picture of the group which I reproduce. The 
Zoroastrian banker himself sits in the second row, to the right 



% 



THE ZOROASTRIANS OF TEHERAN 427 

of the middle as we face the picture, with a scarf thrown over 
his shoulders, and with his two little sons squatting in front 
of him, in Oriental fashion, near the fountain. 

The Teheran Parsis, as I learned, take an interest in educa- 
tion, and I am able to present also a photograph of their Boys' 
School, with the teacher, Kayomars Vafadar, in the foreground 
and some of the members of Arbab Jamshid's staff of assistants 
standing by the side of the scholars. The attendance at the 
school, so far as my information goes, is fair, and is proportion- 
ately larger than that at the Jewish schools in Teheran, con- 
sidering the relative size of the respective communities, there 
being 5100 Jews, in all, as against 324 Zoroastrians. Out 
of the five thousand Jews, statistics in 1904 show a total of only 
372 attending the two schools for boys and girls established 
at Teheran in 1898 by the Alliance Israelite Universelle a 
matter for the consideration of both the amelioration societies 
engaged in the betterment of their co-religionists in Iran. 1 It 
would be interesting and instructive, if it were possible, to draw 
a comparison between these two isolated religious communities 
and show the results accomplished in each case by the special so- 
cieties which have their interests in charge. Such a comparison 
should include also the work of the schools of the various Chris- 
tian missions in Teheran, and should be summarized in connec- 
tion with the Mohammedan institutions, before a final conclusion 
is drawn. 

But that is a matter about which I am not now qualified 
to speak, nor have I time to devote more attention to the 
subject at present, but must turn rather to the theme of the 
succeeding chapter, closely connected with Zoroastrianism and 
Teheran Rei, the ruined suburb of the capital. 

1 See Bulletin de V Alliance Israelite 242 in the boys' school, 130 in the girls' 

universelle, deuxieme serie, no. 29, p. school. Compare likewise my article 

126 (cf. also pp. 168-169), Paris, 1904, ' Teheran ' in the Jewish Encyclopedia, 

where the school-attendance is given as 12. 73-74, New York, 1906. 



CHAPTER XXVII 
RUINS OF REI, THE ANCIENT RAGHA 

4 How doth the city sit solitary that was full of people ! 
How is she become as a widow, she that was great among the nations.' 

JEREMIAH, Lamentations, 1. 1. 

SOME Iranian Jeremiah might well find cause to lament over 
that vast heap of ruins, six miles southeast of Teheran, which 
once formed the city of Ragha, or Rages, the metropolis of 
ancient Media and one of the oldest centres of civilization in 
Iran. Sanctified as the cradle of Zoroast nanism, hallowed for 
a day by the presence of the angel Raphael, exalted by 
princes and cast down by conquerors, this city that was great 
among the nations is now but a mass of crumbling walls, 
mounds, hollows, and ruined watercourses, with but few signs 
of life amid the dust of ages. If marked on the map at all, it 
is designated as the Ruins of Rei, or Rhey (pronounced like 
English ray), for Rei is the modern form of Ragha, Rhagse, or 
Rages. Treasure-hunters dig for coins and pottery amid its 
deserted tumuli, and brick-hunters demolish its walls for 
building-materials to be used in Teheran. In a few places, 
it is true, the irrigator has restored an ancient aqueduct or 
cistern and thus converted a sand-heap into a cultivated 
plot of ground, but otherwise desolation reigns everywhere. 

To the modern Persians Rei is known chiefly as the place 
adjoining the shrine of Shah Abdul Azim, which is visited 
annually by thousands of pilgrims and within whose precincts 
the late Shah, Nasr ad-Din, was assassinated in 1896. Within 
the confines of Rei itself there is also a renowned spring, 
Ohashmah-i All, 'Fountain of AH,' named after the cousin 
and son-in-law of Mohammed and therefore more or less sacred. 

428 




3 a 
g .1 
^ 43 

I! 

os *3 
2 

I 



V 



ANTIQUITY OF THE RUINS AT REI 429 

Upon a hill in a northeasterly direction from Rei is the 
Zoroastrian Tower of Silence, where the Gabars expose their 
dead. Among the ruins at Rei there are some which have a 
special interest, and I shall therefore describe them in detail ; 
but one of the chief distinctions of Rei nowadays is that it 
forms a terminus of Persia's only railroad, an insignificant line 
running from Teheran to the mosque of Shah Abdul Azim. 
Over the six miles of track, with little attention to regularity 
or punctuality, a train of a few cars is drawn by a c mawsheen,' 
as the natives call the engine, having Persianized the French 
word machine. In visiting Rei, however, experience proves 
that it is better to rely on a good horse as a means of transpor- 
tation rather than trust to this uncertain conveyance, since the 
distance can be covered in an hour's canter. The road, lined 
with caravan after caravan, all on their way to or from the 
capital, is not uninteresting, and I felt I had been but a short 
time in the saddle when I found myself dismounting with my 
friends to enjoy a morning cup of tea beneath the ruined cita- 
del and the huge moundlike walls that mark the northern out- 
line of this ancient city. 

The date of the founding of Ragha, or Rei, is lost in oblivion, 
although tradition assigns it to Hoshang, the first king of Iran, 
in the fourth millennium before Christ. 1 According to the 
Bible, Rages, or Ragau, as the name is given, must have been a 
flourishing city in the eighth and seventh centuries before the 
Christian era, since the books of Tobit and Judith mention it 
as an important contemporary of Nineveh and Ecbatana, and 
the strange story of the angel Raphael's visit to Rages is 
familiar to all who know the Apocrypha. 2 The Avesta twice 

1 See the legends cited by Barbier Lb'hr, in Kautzsch, Die Apokryphen 
de Meynard, Diet. geog. p. 273, n. 1. des Alten Testaments, Freiburg, 1898. 

2 See Tobit 1. 14 ; 4. 1, 20 ; 5. 5 ; 6. The correct reading is also given in the 
12 ; 9. 2 (and for ' Rages ' in Tobit 6. 9 old Latin version, Itala, in the Hebrew, 
read ' Ecbatana,' which is rightly given and in the Aramaic; cf. Neubauer, 
in two texts, the Codex Vaticanus B Book of Tobit, introduction, pp. 36, 
and the Codex Sinaiticus K) ; see also 55, 78, Oxford, 1878. 



430 RUINS OF RE I, THE ANCIENT RAG HA 

alludes to Ragha in connection with Zoroaster's name, the 
f Pahlavi texts do likewise, and tradition makes Rei the home of 
his mother. 1 The Old Persian inscriptions speak of the dis- 
trict and city of Raga, and the Greek and Roman classics 
refer to it in connection with Alexander and his successors; 
the later Persian and Arabic writings have much to say of its 
importance, one of its claims to renown being the fact that 
Harun al-Rashid was born at Rei in 763. With a fair degree 
of completeness we can trace its annals down to the fifteenth 
century, by which time it appears to have fallen into its final 
decay, as Clavijo, the Spanish ambassador to the court of 
Timur Lang, in 1404, describes it as ' a great city all in ruins.' 2 
A detailed history of the city I shall reserve for publication 
elsewhere, 3 but I shall reproduce here from the Arabic an 
account of Rei, its gates, and its market-places, given by the 
Mohammedan geographer Istakhri in the tenth century. He 
says : 

1 Rei is the largest of all the cities in its vicinity; it is located on 
the route which runs from Arak to the east, and there is no city 
more populous, more prosperous, or larger than it throughout all 
Islam, except Nishapur. The latter has a greater area and has 
lattice-work buildings, 4 a palace, and other advantages, but Rei is 
superior to it. Its length (like Nishapur) is a farsakh and a half, 
and its buildings are of clay, although plaster and brick are also 
employed in it. The city has several famous gates ; among them 
is the Gate of Tak, which leads out toward Jabal and Arak ; the 
Gate of Balisan, which leads to Kazvin ; the Gate of Kuhkin, toward 
Tabaristan ; the Gate of Hasham, toward Kumis and Khorasan ; 
and the Gate of Sin, toward Kum. The chief market-places, or 
bazaars, are Ruzat, Balisan, Dahak-i Nu, Nasarabad, Sarbanan, and 
the Bab-i Jabal, Bab-i Hasham, and Bab-i Sin. The most important 
of these is Ruzat, which does the principal business and has the 

1 For the passages, see my Zoroas- edited by Jivanji Jamshedji Modi. I 
ter, pp. 202-205. do not agree with Rawlinson's view 

2 See Clavijo, Narrative of an Em- that the. ruins at Veramin represent 
bossy, Hakluyt Society, 26. 99 ; cf. ancient Ragha. 

also Curzon, Persia, 1. 349. 4 Or does it mean a network of 

8 See Spiegel Memorial Volume, buildings ' ? See p. 342, n. 1. 



v 



ARABIC DESCRIPTIONS OF RE I 431 

largest caravansarais ; it is a broad street with, lattice-work caravan- 
sarais and buildings. 1 The citadel is in one quarter of the city and 
in it there is a mosque. Most of the city is in ruins, and the palace 
is in the rampart. 2 The water comes from wells, and there are 
water-conduits (Jcanats). One of the two streams for drinking-water 
in the city is named Surakani, and it flows toward the Euzat 
quarter ; the other is named Jilani (or Gilani), flowing toward the 
Sarbanan quarter. These streams furnish the drinking-supply, 
but as there are many canals (kanats), the people distribute over 
their farms the water that is not needed for drinking purposes. 
The coins that are current in the city are the dirham and the dinar. 
The inhabitants resemble the people of Arak in appearance and 
are of refined manners. They are chiefly occupied with trade and 
commerce.' 3 

The description of Rei given by Ibn Haukal, whose work 
(about 975) is based on Istakhri, but whose description, espe- 
cially of gates, is sufficiently independent to deserve quotation 
in full, reads : 

' The most considerable city of those we have mentioned is Rei. 
After Baghdad there is not in the eastern regions any city more 
flourishing. Its gates are much celebrated; one of them is called 
the Darwazah Natan, facing the mountainous country, or the 
Kohistan of Arak; another leads to Kazvin; another, called the 
Darvazah Gerhak, 4 is in the direction of Kum. And there are 
many remarkable streets and quarters in this city, such as Rudah, 
Kalisan, Dahak Nu, Nasarabad, Sarbanan ; Bab al-Jabal, or the 
Mountain Gate ; Dar-i Hasham, or Hasham's Gate ; the Dar-i Ahanin, 
or the Iron Gate ; and the gate called Dar-i Athab : but the quarter of 
Rudah is the most populous and flourishing of all. In this place 
(Rei) are many bazaars, caravansarais, and market-places. In the 
suburbs there is a mosque. The citadel is in good repair, and there 
is a wall round the suburbs, which is, however, falling to decay and 

1 See preceding note. (thanks again to Dr. Yohannan and 

2 Judging from the description of Professor Gottheil for assistance with 
Yakut (tr. Barbier de Meynard, Diet, the Arabic). For a briefer description 
geog. p. 277), the palace, like the of Rei by Istakhri, see De Goeje, 1. 202. 
mosque, must have been situated within 4 The precise reading of these names 
the rampart of the citadel ; but the in the Arabic letters is not certain, and 
text is not absolutely clear to me. this probably accounts for some of 

8 Istakhri, ed. De Goeje, 1. 207 the seeming variations from Istakhri. 



432 RUINS OF REI, THE ANCIENT RAG HA 

almost desolate. Here they have both river-water and water 
V brought by canals or trenches; one of these is called the Kariz 
Shahi, or Royal Aqueduct ; it passes by Sarbanan ; another, called 
Gilani, also passes through Sarbanan. For the most part the inhabit- 
ants drink the water of these aqueducts. There are many canals 
besides. Here they cultivate the land and practise husbandry, and 
traffic for gold and direms. The people of this place are hospitable 
and polite. Here they manufacture fine linen, cotton, and camelots, 
which are sent to all parts of the world.' 1 

The northern wall, which I have mentioned, is the best-pre- 
served portion of the entire circumvallation, and the mound 
rising before it is the site of the ancient citadel. The natives 
call the mound Kal^ah-i Rei, c Castle of Rei,' and it is evidently 
this elevation to which Yakut refers when he speaks of the 
height as Eei-bandi, 'Stronghold of Rei.' 2 According to 
Yakut, who quotes liberally from his predecessors in his long 
article on Rei, the citadel was situated outside the 4 city ' itself 
and formed a part of the so-called c exterior town,' or suburb, 
named Mohammadiab. 3 Its walls and moat, he adds, were 
completed by al-Mahdi in A.D. 775 (A.H. 158), and the great 
mosque was also built by this same ruler. 4 Yakut's description 
of the ' outer city,' the 4 citadel,' and the ' city ' itself to 
repeat again his own designations would be worth quoting 
in full if there were space, because his remarks make us more 
sure of our ground when endeavoring to trace among the ruins 
the main features of by -gone Rei. By far the best description 
of these remains, however, was given by an Englishman, Ker 
Porter, nearly a century ago. His account was also accom- 

1 Ibn Haukal, Oriental Geography, cites, is a misreading of the Arabic 
tr. Ouseley, pp. 176-177. In this quo- letters for Eei-bandi. 

tation, I have made insignificant 8 See the French translation of 

changes in punctuation and spelling, Yakut's long article on * Rei,' and 

so as to conform with the rest of the * Mohammadiah ' in Barbier de Mey- 

book. nard, Diet. geog. de la Perse, pp. 

2 Yakut, p. 277. The name Zobei- 273-280, 516-518. 

diah given to the citadel in the text of * Yakut, pp. 277, 517. 
Jafar ar-Razi, which Yakut (p. 517) 



> 



KER PORTER'S DESCRIPTION OF THE RUINS 433 

parried by a well-sketched plan, which will remain authoritative 
until careful archaeological surveys and researches point out the 
changes that have occurred since his day owing to the lapse of 
time, excavations by the natives, and the building of the rail- 
road to Rei. 1 Ker Porter's description is so excellent and his 
book so little accessible nowadays, that I regard it worth while 
to reproduce the main paragraph here, making such changes 
in punctuation as would be required to-day, and including his 
excellent map of Rei, which may be helpful in locating the 
gates mentioned by the Arab geographers, and adding explana- 
tory footnotes of my own. 2 

'The ruins lie about five miles southeast of Teheran, extending 
from the foot of the curving mountains and running in that direc- 
tion across the plain in an oblique line southwest. The surface of 
the ground, all over this tract, is marked by hollows, mounds, 
mouldering towers, tombs, and wells. The fabric of all being chiefly 
of that burnt and sun-dried material which seems to bid defiance to 
the last oblivious touch of time. A very strong citadel appears to 
have occupied a high and rocky promontory that juts out consider- 
ably beyond the other huge buttresses of nature which here start 
from the different clefts in the mountains. 3 Along the perpendicular 
sides of this height we easily discovered the foundations of its 
embattled works. And directly from its base a line of massy forti- 
fication appeared, reaching southward and apparently defending the 
eastern face of the city, till it terminated in an immense square bul- 
wark, flanked with towers and making a fortress in itself. 4 Thence 
the wall curved round in an irregular oblique sweep towards the 
northwest till it met another enormous square tower flanked in like 
manner with six round ones. 5 This tower terminated not only that 
line of wall, but another, which had also started from the base of the 

1 It is to be hoped that some future plan; it is the citadel alluded to by 
student or traveller, or an investigator Yakut, as I have already described, 
like M. de Morgan, may undertake a 4 This is the eastern outline marked 
series of careful researches at Rei, be- ACCD in the plan. 

cause the field is a promising one. 5 On the plan this is the broken line 

2 Ker Porter, Travels in Persia, running upward toward the northwest, 
1. 358-360. past the village of Shah Abdul Azim 

3 This is marked A on Ker Porter's and the tomb of the Imam Abdullah. 

2r 



r 



434 RUINS OF REI, THE ANCIENT RAGHA 

promontory and formed the northern front of the city ; l the whole 
fortified space between the three walls taking the shape of a tri- 
angle, its vertex [A] touching the citadel-promontory, its base [DD] 
stretching southwest from the one large, square tower to the other. 
These walls are still many feet in height, of prodigious thickness, 
and have been additionally strengthened by proportionably sized 
towers connecting the wall and placed at point-blank arrow-distance 
from each other. The two enormous fortress-like towers before 
mentioned [DD], which terminate the southeastern and north- 
western points of the triangle, are united with the walls ; but in 
going along the outer side of the longest line which stretches from 
the one square tower to the other, we find a third tower about the 
middle of the wall, but standing out at some distance from it. 2 It is 
nearly of the same dimensions with the others and supported in the 
same way with round flanking towers. Probably a ditch and a bridge 
lay between this great bulwark and the principal gate of the town, this 
entrance lying in almost a direct line with the citadel. The remains 
of other fortifications are near it, as if still more to protect this 
ingress which opens to the southwestern side of the plain. I have 
no doubt that these three square towers [DDD] commanded the 
three great entrances to the city, the northern holding the com- 
munication with Azerbijan and Mazanderan, the southern that 
toward Khorasan, and the southwestern pointing to Hamadan, the 
ancient Ecbatana. 3 Therefore, by this last gate, it is probable that 
Tobit's celestial messenger entered on his embassy to Gab[a]el. 4 
At the foot of the great promontory [A] which crowns the apex 
of the fortifications, and projecting within their area, is another 
range of equally strong walls, embracing a considerable space and 
forming a lower citadel [B], within which, in all likelihood, 
were the royal palace and other buildings of state. 5 Another 

1 That is, running from A westward 4 Through the same approach Darius 
to D. The outlines of the northern Codomannus probably entered when 
wall are best preserved. fleeing before Alexander. 

2 It is also marked D and is near the 6 Ker Porter's inference is borne out 
tomb of the Imam Abdullah in the by Yakut, pp. 277, 617, who cites older 
southwest part of the plan. authorities also to the effect that the 

8 With these remarks on the en- citadel, where the Khalif resided dur- 

trances it is interesting to compare the ing his stay at Rei, dominates the 

notes on the city gates made by the grand mosque of Mahdi and the palace 

Mohammedan writers quoted above, of the governor. Both of these build- 

whose statements antedate Ker Porter ings, together with the city prison, 

by nearly a thousand years. were situated in the ' exterior city,' or 




\ 



KER PORTER'S PLAN OF REI 




TOWER ON A HILL NEAR REI 



r 2 



THE MASSIVE RAMPARTS 435 

wall, 1 exterior to the city-rampart, connects the height on which 
the first great citadel stands, with another rocky projection of the 
mountain, where every tenable spot has been strongly guarded by 
outworks, 2 linking themselves across the gorge of a deep ravine to 
the side of a third citadel, or fortress, finely built of stone and on 
the summit of an immense rock which commands the open country 
to the south/ 3 

Some idea of the size and sweep of the massive ramparts 
may be gathered from my photographs. In several places the 
moldering walls are fully fifty feet high, and the sun-dried 
bricks which composed them can still be seen plainly in outline, 
and they reminded me of the bricks used in the old Zoroastrian 
fire-temple near Isfahan, although somewhat larger, since 
some of the specimens measured 17 by 7 inches (44 by 18 cm.) 
along the face. 4 I was afterward told in Teheran that they 
are known as l Gabar bricks,' a designation that would seem 
to confirm my impression of their antiquity. 

Everywhere about the walls and within the desolate area 
that once formed the city there are scattered fragments of 
pottery and shards of earthenware. Several of the pieces 
closely resembled the bits of a rough yellow jar I had picked 
up on the hillside of the Atash Gah near Isfahan and 
probably dated from the same period. 5 I found also a few 
specimens of finer ceramic art which showed the exquisite 
lustre or glint known to porcelain-connoisseurs as the reflet 
metallique, and I added them to my collection of mementos. 
The natives are aware of the value of these curios, and they 
treasure up for sale the more artistic specimens unearthed in 
digging new wells or restoring the old canals that still run 
from the active stream which flows through Rei to restore life 

suburb, of Mohammadiah (Yakut, pp. the tower F in the plan. A photograph 
277, 517). of the tower F may be seen in Curzon, 

1 This wall is shown by the curved Persia, 1. 351. 

line running from the citadel A toward 3 The position of the fortress is in- 

F in the plan. dicated by G in the plan. 

2 The outworks are indicated by 4 See pp. 253, 255, above, 
circles running from the hill toward 6 Cf. p. 253, above. 



436 RUINS OF RE I, THE ANCIENT RAG HA 

to the gardens within and without its otherwise desolate 
j walls. 

Within the compass of the ancient city, and nearly midway 
between the citadel and the central point, there rises a high 
brick tower which the natives say is 4 old ' (kadlm), but whose 
claim to antiquity cannot exceed a thousand years. 1 Some 
writers have supposed it to be the burial-place of the grandson 
of Tamerlane, Khalil Sultan, and his beloved consort, Shad al- 
Malik, ' Joy of the King,' in the fifteenth century ; others 
ascribe it to a date four centuries earlier and see in it the 
mausoleum of Tughral Beg, who died A.D. 1063. I am 
not yet prepared to pronounce judgment as to which view 
may be nearer the truth. 2 I can only say that the tower is not 
mentioned in Istakhri, Ibn Haukal, or Yakut, and that it re- 
minded me of two similar towers I had seen in Hamadan. 3 

This structure, cylindrical in form, but corrugated in outline 
because of its many angles, rises to a height of about seventy 
feet, is fully forty feet in diameter, hollow within, and open at 
the top. The arrangement of the bricks of the outer casing is 
such that they form a succession of huge perpendicular prisms, 
some twenty in number, and the cement is curiously indented 
in a honeycomb fashion, attention to which was called years 
ago by a Presbyterian missionary from the United States, the 
Rev. Dr. J. Perkins, in an unpublished letter to the Amer- 
ican Oriental Society. A paragraph in his ' Diary of a 
Visit to Teheran, the Capital of Persia,' dated Dec. 2, 1845, 
on p. 58 of the manuscript, briefly describes the tower and 
concludes as follows : 4 Another peculiarity which I noticed 
is, that the mortar in the interstices between the ends of the 
bricks is curiously indented and wrought into honeycomb sec- 
tions, which imparts to the whole an exquisite delicacy, con- 

1 The position of this tower (which is of the tower, which was taken before 

roughly indicated by E in Ker Porter's the structure was restored, 
plan) may be seen from the photo- 2 For the views, see Benjamin, 

graph that I have reproduced. I Persia, 1. 59 ; Curzon, Persia, 1. 350. 
have added also an older photograph 8 See p. 162, above. 



> 




TOWER AMONG THE RuiNS OF REI, AFTER RESTORATION 




TOWER AMONG THE RUINS OF REI, BEFORE RESTORATION 



AN ANCIENT BRICK TOWER 437 

nected with durability.' 1 The entrance to the tower i& 
through a high, arched doorway surrounded by a square portal, 
as will be seen in a photograph that was taken before the two 
were restored, some fifteen or twenty years ago. The picture 
shows why the Kufic inscription, which Ker Porter says en- 
circled the top, has disappeared, no attempt being made to 
restore it when the new cornice was added. Care, however, was 
taken to develop a garden around the tower and to plant nu- 
merous kalam trees, whose slender and straight lines harmonize 
admirably with the perpendicular outlines of the structure. 

An ascent to the summit of the edifice is worth making, as 
the view of Rei from its top is one of the best that can be ob- 
tained. Far to the north is se*en the lofty crest of Damavand. 
In the near foreground the broken outline of the northern wall 
of ancient Ragha may easily be traced from east to west. 
As one looks toward the hills on the northeast, the Gabar 
Dakhmah, or Tower of Silence, appears like a white speck 
against a rocky background, while on another hill to the east 
is a small tower ('old Gabar,' the guide suggested), and 
again somewhat beyond it to the east-southeast rise the walls 
of an elevated fortress. Far to the southeast and south, one 
notices on the plain a series of mounds of earth and elevations 
that are said to be a farsakh's distance apart and are explained 
(for lack of a better interpretation) as ' signal towers ' for pil- 
grims making the journey to and from Kerbela. To the south 
and southwest, in the nearer foreground, the hamlet of Shah 
Abdul Azim nestles amid green trees, and the western line of 
Rei is skirted by the railway that approaches the shrine. 

While resting and taking luncheon near the gateway that 
forms the entrance to the tower garden, we had a chance to 
converse with the natives and to ask them some questions to 
see whether they knew anything of Zoroaster or were even 
acquainted with his name as Zardusht in Modern Persian. In 

1 This honeycomb effect in plaster hammedan architecture of Merv, Bo- 
is familiar in a larger way in the Mo- khara, and Samarkand. 



438 RUINS OF RE I, THE ANCIENT RAG HA 

answer to the inquiry as to who was the prophet of the Gabars, 
two of the Mohammedans with whom we spoke replied, 
wholly independently of each other, 'that 4 the prophet of the 
Fire-Worshippers was Ibrahim,' that is, Abraham. This re- 
sponse is of interest because of the frequent association of 
Zoroaster with Abraham by the Moslems. 1 Somewhat later 
there came along a bright-faced young Mullah wearing a white 
turban ; on being interrogated, he promptly replied : ' Zardusht 
was the prophet of the Gabars and lived about six thousand 
years ago ; to find out more about him, one must consult the 
Gabar books, for the Gabars have books.' 2 

Although there would have been time for a special visit to 
the Shrine of Shah Abdul Azim, I decided rather to devote it 
to continuing my survey of the ancient ruins. Returning 
to the upper section of Rei, I first examined a sculpture, carved 
in a space something less than ten feet by twenty, on a rock to 
the east of the ancient citadel. It has something of a history, 
as the rock is really a palimpsest, an old carving having been 
chiselled away in the first part of the nineteenth century to 
give place to a new one. In its present form the monument 
represents Fath Ali Shah in the act of spearing a lion, and has 
little, if any, historic value. The old sculpture which the mon- 
arch caused to be obliterated in order to afford a space for 
his own picture was really a historic one and dated back to 
Sasanian times. Fortunately it had been sketched by Morier 
(1809), W. Price (1811), and Sir William Ouseley (1811-1812), 
and described also by Ker Porter (1818), so that it has not 
been lost entirely, and I reproduce here the drawing by 
Ouseley, because of its archaeological interest. 8 The sculpture 

1 See p. 404, above, and my Zoro- 8 Sir W. Ouseley, Travels, 3. plate 
aster, p. 157 ; cf. also Hyde, Historia 65; compare also Morier, Second Jo ur- 
Religionis Veterum Persarum, pp. ney, 2. 190, London, 1818 ; W. Price, 
28-31, Oxford, 1700. Journal of the British Embassy to Per- 

2 The Moslems always lay stress on sia, p. 37, 2d ed., London, 1832 ; Ker 
any religion that has books, even if Porter, Travels, 1. 363; and seethe re- 
they despise the believers as infidels. marks by Curzon, Persia, 1. frTl-frV?. 




SASANIAN SCULPTURE NOW DESTROYED 
(From the drawing of Sir William Ouseley) 




MODERN BAS-RELIEF REPLACING THE SCULPTURE SHOWN ABOVE 



AN OBLITERATED SASANIAN SCULPTURE 439 

represented a mounted warrior charging at full speed with a 
couched spear against an antagonist, the head of whose hor> 
was roughly indicated in outline on the stone. The figure 
with the spear was undoubtedly a Sasanian king, as was shown 
by the balloon-like head-gear and other typical accoutrements 
of Sasanian times, but we no longer know which particular 
monarch may have been intended. Ker Porter thought it 
probable that Ardashir, the founder of the dynasty, was the 
hero, and that the combat was against the last of the Parthians ; 
Sir William Ouseley, on the contrary, was inclined rather to 
attribute it to Ardashir's son Shahpur, but both scholars were of 
the opinion that the entire sculpture had never been finished. 
We can but regret that it is no longer possible to solve such a 
problem, and we must deplore the destructive zeal of Fath Ali 
Shah, which has robbed us of a monument of antiquity. 

The sculptural activity of Fath Ali Shah was not confined, 
however, to making palimpsest records; he has left us another 
carving which is chiselled on a huge panel that was specially 
prepared for the purpose on the face of the volcanic rock 
overhanging the waters of the holy and much frequented 
Chashmah-i Ali, or 4 Spring of Ali.' The sculpture represents 
the long-bearded monarch seated upon his throne, surrounded 
by his sons and vizirs, in the same manner that he is portrayed 
in the painting on the walls of the picture-gallery of the Naga- 
ristan in Teheran. 1 

Riding some distance beyond the intrenchments of Rei 
toward a spur of the mountains that surround the ruins on the 
easterly and northeasterly side, I ascended the barren hill on 
whose rocky side stands the old Zoroastrian Tower of Silence. 
Because of the historic association of Ragha with the Zoroas- 
trian religion, it is probable that this dakhmah is located on 
one of the oldest known sites for a Tower of Silence, and its 
situation fulfils all the requirements of the Vendidad canon, 
which enjoins that the dakhmah shall be placed on a hill, far 
1 For a reproduction of the Nagaristan paintings, see Curzon, Persia, 1. 338. 



440 RUINS OF RE I, THE ANCIENT RAG HA 

from human habitations, but accessible to 4 the corpse-eating 
r-dogs and birds.' The location recalled that of the dakhmah 
at A j mere in India, and the structure resembled the Parsi 
Towers of Silence at Bombay in being round in shape, about 
thirty feet in height, and covered with a cement which gives 
it the whitish color noticeable in the Bombay towers. 1 It 
was repaired some years ago, as were the dakhmahs at Yezd, 
through an expenditure of. funds supplied by the Parsis of India 
for the amelioration of the Zoroastrians in Persia. 2 Unlike 
the Bombay towers, it has no door, because, as the Zoroastrians 
in Teheran told me, they are afraid that the Mohammedans 
may desecrate the place. The corpse is lifted over the wall by 
means of ladders and ropes or a chain, as I learned from the 
Rev. L. F. Esselstyn, a missionary, who had helped to lift into 
it the dead body of one of his servants, who was a Zoroastrian. 

I ascended higher up the hill to a point where I could 
see the interior of the tower. The pdvis, or receptacles for 
the bodies, were arranged after a rectangular method and 
not radiated in the wheel-like fashion of the Parsi dakhmahs. 
I could see no bhanddr, or central pit, in which to deposit the 
skeleton after the flesh had been stripped from the bones, but 
I was told by those who had been inside the dakhmah that 
there was a place to which a few steps led down, and this might 
have served as an astoddn^ or repository for the bones. The 
whole arrangement appeared to me rather primitive and less 
systematic and up-to-date than in the Bombay Towers of 
Silence. I may add that there was no evidence anywhere of 
a sagri, or shrine, for a perpetual lamp to burn near the place 
of the dead. 

As we galloped homeward toward Teheran, the sun began 
to cast long shadows from Mount Damavand, the great peak 
of Alborz, the Avestan Kara Berezaiti, whose snowy cap and 
frowning front loomed skyward to the height of nearly twenty 

1 See the accompanying photograph of this dakhmah given above, p. 403. 
of the tower and also the distant view * See p. 397, above. 




SCULPTURED PANEL OF FATH ALI SHAH, ABOVE THE ' SPRING OF ALI 




THE ZOROASTRIAN TOWER OF SILENCE AT REI 



V 



THE DAKHMAH AT REI 441 

thousand feet. No wonder it looked sullen at its ceaseless task 
in crushing beneath its ponderous bulk for ages, as legends tell, 
the giant monster Zohak, or Azhi Dahaka, lest he escape from 
his chains and tyrannize the world. Only in the eleventh millen- 
nium will the mountain be relieved from duty, for then the 
hero Sama Keresaspa will waken from his sleep, slay Zohak, 
and usher in the dawn of a new era. 

But soon we were entering the gates of the capital, and my 
thoughts were turned from ancient legends and ancient ruins 
to themes of the present. 



CHAPTER XXVIII 
THROUGH MAZANDARAN TO THE CASPIAN SEA 

' To the sea-shore he gan his way apply, 
To weete if shipping readie he mote there descry.' 

SPENSER, Faerie Queene, 5. 12. 3. 

I HAD been a week in Teheran and had especially enjoyed 
my visit because I had accomplished most of the immediate 
purposes I had in view. Though loath to depart, I found the 
time demanded my leaving if I were to carry out my plans for 
visiting Central Asia. The official calls had been made, and 
the requisite papers obtained from governmental sources to en- 
able me to travel freely in Transcaspia and Turkistan; the adieus 
to my friends had been said ; and the minor arrangements for 
the journey had been completed. Yet to say farewell to Persia 
without delays is always impossible, and although the carriage, 
horses, and driver were at the door, it took an endless amount 
of time and ado before the baggage was lashed to the wagon 
and the start made for the Caspian Sea which we hoped to 
reach on the second day intkallah, 4 God willing ! ' 

Another hour had elapsed before we came to the outer post- 
station of the city, near the Kazvin gate. A long delay oc- 
curred here to remind me again that we were still in Persia. 
Each piece of luggage had to be taken off the rickety vehicle, 
weighed with a precision that could come only from a desire 
to exact the utmost shdhi for every ounce of baggage and 
extract an additional bakhshish, and then the packages were 
slowly lashed in place again upon the conveyance. 

The fare and toll rate for this uncomfortable journey of some 
240 miles in a rattle-bang old brougham was about 70 tomans 
(dollars), a sum that seemed exorbitant considering the dis- 

442 




THE KAZVIN GATE, TEHERAN 




THE MEIDAN-I MASHK, DRILL SQUARE, AT TEHERAN 



KAZV1N 443 

comforts, but I was promised a prompt and safe arrival (again 
inshdllah), and away we started. 

The road was the best I had seen in Persia, because built 
and managed by the Russians, and it was therefore marked off 
in versts, not farsakhs, as a measure of distance. The weather 
was warm, as it was the end of May ; the way was hot and 
dusty ; and the plain flat and uninteresting. Nothing special 
engaged my attention until we came to the river Karaj. This 
active stream rushes out from a deep gorge in the mountains, 
with high precipitous banks that recalled to my mind the sug- 
gestion I had heard from the Gabars in Yezd, that the Karaj 
might be the Darejya of the Avesta. 1 I photographed the 
scene to fix it firmly in my memory, but I felt no more con- 
vinced of the accuracy of the suggested identification than I 
did when the Zoroastrians proposed it to me. Without halt- 
ing longer we proceeded again and travelled all through the 
night, jolting, lurching, lunging in our cramped vehicle, which 
stopped at irregular intervals for fresh relays of horses, until 
we rumbled into Kazvin shortly before eight the next morning 
after a tour of nearly twenty hours. 

Kazvin was once one of the most important cities of northern 
Persia and was sufficiently well known in England in Milton's 
day to be mentioned in Paradise Lost. 2 The city's long history 
is said to have begun in the fourth century A.D., when it was 
founded by the Sasanian King Shahpur II. Four centuries 
later Harun al-Rashid showed it distinguished marks of favor, 
and so far had Kazvin advanced by the beginning of the 
thirteenth century that Yakut, in his Arab geography, devotes 
several pages to the town. In the sixteenth century it became, 
for the first time, an imperial city, as Shah Tahmasp I then 
made it his chosen residence; but Isfahan and Teheran in turn 
supplanted it as the nation's capital, so that Kazvin is no longer 
a city of the first rank, although the number of its inhabitants 
is variously estimated from 50,000 to 100,000 persons. 

1 See p. 360, above. 2 See Milton, Paradise Lost, 10. 435. 



444 THROUGH MAZANDARAN TO THE CASPIAN SEA 

Despite its level and unimposing situation, Kazvin makes a 
pleasant impression upon the eye because the main avenue 
leading through the city to the Governor's Palace is richly 
lined on either side with shade-trees. The principal inn, 
mahmdn-khdnah, at the other end of this boulevard, is the best 
of its kind in all Persia and has spacious quarters for fairly 
comfortable lodging and ample accommodations for numerous 
relays of post-horses, but it does not possess sufficient attraction 
to induce the stranger to prolong his stay more than is necessary. 
The rows of single-storied bazaars on the main streets gave me 
the impression of being reasonably prosperous, and in one of 
the semi-European shops I replenished my supply of certain 
necessaries for the voyage which one best knows how to appre- 
ciate after long travel in Persia. I was not sorry, however, 
when the time came to resume the journey toward the Cas- 
pian, although I should have liked to visit the mosque which 
was originally built by Harun al-Rashid. 1 I also regret that 
I did not know at the time that the distant hills that I saw 
rising back of Kazvin were once the stronghold of the Sheikh 
al-Jabal, 4 Old Man of the Mountains,' the head of the Assassin 
Band in the twelfth century, for I am sure I should have noted 
the outline of these hills more closely than I did and should 
have sought for local legends regarding the picturesque though 
villanous outlaws whose name, hashshdshin, 4 hashish-eaters,' cor- 
rupted by Europeans into assassin, has become synonymous 
with murderer. 

The next stages of the journey were devoted to sleeping as 
best I could in the uncomfortable conveyance, but I awoke in 
time to enjoy the grand mountain scenery of the Kharzan Pass. 
This lofty roadway crosses the Alborz range at a height of 
seven thousand feet, or more, above the Caspian, and in winter 
it is one of the bleakest and most desolate places in all Persia, a 
pathway of death where many a beast of burden falls by the 
way. The descent to the village of Peichanar was slow and 

1 See Yakut, p. 443. 




THE MAHMAN-KHANAH AT KAZVIN 




THE RIVER KARAJ 



OF THF 

UNIVERSITY 

OF 



FROM KAZVIN TO ENZALI 445 

winding. The entire region interested me because it lay on 
the borders of Mazandaran, the abode of the Mdzainya Daevas, 
4 Demons of Mazandaran,' in the Avesta, and I confess that 
some of the natives of the rude villages looked like the veritable 
dlvs, 4 devils,' of Firdausi's Shah Namah. 

Early on the second morning we crossed the Safid Rud, liter- 
ally 'White River,' which I believe to be identical with the 
Daitya River of Zoroaster, and I wondered whether its waters 
shimmering in the light of sunrise preserved any memory of 
the prophet's glorification or of the sacrifice offered on its bank 
by his patron Vishtaspa. 1 At nine o'clock we were in Resht, a 
flourishing but unattractive town of possibly 100,000 inhabi- 
tants. 2 I did not stay here more than two hours before 
engaging another conveyance to carry Safar and myself to 
Piri Bazaar, the landing-place where one hires a rowboat as 
a transport to Enzali, the main port. 

Piri Bazaar is a filthy hole, a place where it is almost im- 
possible to avoid a conflict with the shameless watermen who 
seem to have all the vices of dishonesty and fraud that belong 
to boatmen at insignificant ports. Bargains in such cases are 
usually treaties of peace after prolonged warfare in which the 
foreigner comes off worst, at least in his own estimation. The 
bags and packages were finally bestowed somewhere in the bow 
or in the stern of the boat, after an extra bakhshish had been paid 
to insure more careful handling, and we rowed and poled for 
several hours through the lagoons and river until landing at 
Enzali. Here I spent two nights waiting for the steamer Con- 
stantin of the Kavkas-Merkur Line to sail, and I enjoyed the 
company of the British Consul at Isfahan, Mr. J. R. Preece, 
who was on his way to England after a long residence in 
Persia. 

To reach the steamer, which lay several miles from shore, 

1 See my Zoroaster, pp. 40, 42, 45, 49, of the Board of Foreign Missions oj 
52, 196, 197, 211, 221. the Presbyterian Church in the U.S.A. 

2 See Sixty-seventh Annual Report p. 231, New York, 1904. 



446 THROUGH MAZANDARAN TO THE CASPIAN SEA 

it was necessary to embark in a lighter, a huge scow almost as 
unseaworthy as the one in which I had crossed the Persian 
Styx at Julfa on entering Iran, but much more unsafe because 
of the distance to be traversed and the danger from the surf 
and waves of the Caspian, which were running high at the 
time. I could not help thinking what an irony of fate it would 
be to be lost in this wretched hulk of a scow after climbing 
Behistan and going through various hardships on the journey 
now drawing to its close. But all went well. On board the 
Constantin I found good accommodations and excellent com- 
pany, the Austrian Minister and his wife being among the 
passengers on their way from Teheran to Europe. The sea 
journey proceeded pleasantly. 

But Persia all this while was fading gradually from view. 
The steamer called next day at Astara, where I bade adieu to 
my faithful Safar, who had served me so well and whom I 
would gladly, if possible, have taken with me to Central Asia. 
I felt certain it was not easy for him, too, to say good-by. As 
his yawl-boat pushed off from the steamer's side and headed 
toward the shore, I could see Mount Ardabil, where Zoroaster 
once preached the law, raising its lofty brow, no longer mitred 
with snow, but turbaned in green. The color may have been 
emblematic of the change of faith from Zarathushtra to the 
Prophet of Islam, but perhaps, like the green bay tree that has 
sprung from the roof of the Tomb of Cyrus at Pasargadae, it 
symbolized the fact that florescence in Persia is still possible, 
when the season of national greatness once again returns. 

The wheels of the steamer began to revolve ; I waved good- 
by once more to Safar and said a long farewell to the Land of 
the Sun as we sped forward toward Baku, whence I was to 
continue my journey into Central Asia, a description of which 
I shall give in another volume, supplemented by a historic 
account of Susa and of Eastern Persia, which I hope also to 
visit at some future time. 




PEICHANAR, BETWEEN KAZVIN AND RESHT 




ENZALI 



OF THE 

UNIVERSITY 

OF 




INDEXES 



GENERAL INDEX 



Small superior numbers after the page-numbers refer to the footnotes. 

Iranian words are printed in italic type, and when they are not modern Persian 
the language is indicated. 

In the alphabetic arrangement of Arabic names the prefix al- has been disre- 
garded, but not the elements Ibn and Abu. 



Abadah, village, 277. 

Abarkuh, town, 341-4. 

Abbas, Shah, 27, 37. 

made Isfahan capital, 265. 
founded Julfa, a suburb of Isfahan, 

270. 

mosques erected by, 5, 267. 
mail-coat of, 422. 

Abbasid dynasty, 27. 

Abdul Azim, Shah, shrine of, 428. 

Abdullah, governor of Khorasan, 
destroyed Zoroastrian works, 
359. 

ablution of nine nights, 381, 392, 
396. 

Abraham, legends of, 342, 343. 
associated with Zoroaster, 404, 
438. 

Abu Ali ibn Sina, tomb of, 165-7. 

Abu Dulaf Misar, described Tak-i 
Bostan, 224. 

Abu Kalanjar Sultan, fortified Shi- 
raz, 325. 

Abu Said, tomb of, 167. 

Abuzai'dabad, village, 410, 416. 

accipenser huso, 9. 

Achsemenian dynasty, 26. 

Achmetha, Aramaic form of Hama- 
dan, 152. 

Adarian, Zoroastrian chapel, 366. 

Adhargushnasp (Adharakhsh), fire- 
temple of, 133-42. 

Aena, name of Anahita, 153. 

aesha (Av.), plow, 246. 

Afghans, invaded Persia, 27. 

Afrasiab, conflict with Horn and Kei 

Khosru, 137-9, 142. 
legend of defeat by Rustam and 
Kei Kaus, 264. 

2o 449 



Agamatanu, Babylonian form of 

Hamadan, 150. 
Agha Mohammed Shah, 27. 
sacked Hamadan, 155, 159. 
razed the ramparts of Shiraz, 325. 
laid out a garden at Shiraz, 327. 
made Teheran capital, 419. 
Aghri Dagh (Ararat), 17. 
agriculture, implements of, 85-6, 

246-8. 

praise of, in the Avesta, 246, 373. 
Ahmadabad, 122-3. 
Ahmat, ash-hill at, 94. 
Ahriman, 63, 363. 
Ahuna Vairya stanza, as pronounced 

at Yezd, 365. 

recited in the funeral rites, 392. 
Ahura Mazda (Ormazd), 65-6, 362. 
Zoroaster converses with, 62. 
in the inscriptions of Darius, 181-2. 
represented on a sculpture at 

Naksh-i Rajab, 309. 
possibly represented at Tak-i 

Bostan, 220, 225. 
Airyema Ishyo, a prayer, 87. 
Aiwisruthrima Gdh, a time of day, 

394. 

Akdah, village, 402, 403, 415. 
Akhuri, Akori, Armenian village, 17. 
Alborz, Mount, 58. 
Alexander the Great, at Ecbatana, 

157, 158. 

vanquished Darius III, 26, 319. 
burned the archetype copy of the 

Avesta, 306-7. 
legends of, 5, 163-5. 
Aliabad, village, 347. 
Allah ad-Daulah Garshasp, Sultan, 

built mosque at Yezd, 350. 
altars, Achsemenian, 303. 
Alusjird, river, 147. 



450 



GENERAL INDEX 



Alvand, Mount, 146, 147. 
name of, 146 l . 
inscriptions on, 170-3. 
blank niches on, 173. 
Al-Yalvah, the sacred book of the 

Yezidis, 14. 
ambdr, store, reservoir, 42. 

epithet of Hamadan, 149. 
Amesha Spentas, 62, 362. 
Amestris, 12. 
Amr ibn Leith, Safarid ruler, built a 

mosque at Shiraz, 325. 
Amshaspands, 62, 362. 
amulets, worn by children, 378. 
Anahita, temple of, at Ecbatana, 

153. 

temple of, at Kangavar, 240-2. 
identified with Artemis (Diana), 

240. 

possible representation of, 225. 
recognized in inscriptions of Ar- 
taxerxes II, III, 183. 
Anaias, Anaitis, names of Anahita, 

153. 
Andarz-i Buzurg Mihr, Zoroastrian 

work, 359. 

Andarz-i Gavdh, a formula, 385. 
Angra Mainyu (Ahriman), 63, 363. 
animal sacrifice, at Hamadan, 162. 

at Yezd, 371. 

Anjuman, Zoroastrian synod, 376. 
at Yezd, meeting with the, 356- 

366. 
Antiochus the Great, found gold at 

Ecbatana, 154, 158. 
Antiochus Epiphanes, came to Ecba- 
tana, 169. 
Antonius, Marcus, besieged a fort 

named Vera, 137. 
Anushiryan (Chosroes I), born at 

Ardistan, 406. 
moved the fire of Adhargushnasp, 

134, 135. 

a work dedicated to, 360. 
Apocrypha, New Testament, legends 

of the Magi in, 102-3. 
Old Testament, see Judith, Tobit. 
aqueducts, 10, 407, 409, 431. 
'arabah, cart, 248. 
Arabian Nights, 31. 

No-Ruz mentioned in the, 99. 
Ararat, Mount, 16-7, 19. 
Aras, Araxes, river, 22. 
archaeological research, need of, at 

Ardistan, 407. 
at Gulpaigan, 250. 
at Hamadan, 163. 
at Kangavar, 238, 242. 
at Marand, 36. 



at Persepolis, 318. 
at Rei, 433. 

around Lake Urumiah, 98 l . 
archangels, Zoroastrian, 62, 362. 
architecture, 28. 

Greek style of, in the ruins at Kan- 
gavar, 237. 

Ardabil, town, 53, 139. 
Ardakan, district, 403. 
Ardashir I (Papakan), represented 
on Sasanian sculptures, 81, 220, 
302, 308, 439. 

ardastdna (OP.), structure, 315, 407. 
Ardavan, see Artaban. 
Ardistan, town, 406, 416. 
Arezurahe Griva, Mount, 124. 
Arguri, Armenian village, 17. 
Armenians, at Hamadan, 148. 
Arphaxad, legendary founder of 

Hamadan, 152. 
Arrian, described the Tomb of 

Cyrus, 285. 
art, Persian, 28, 43. 
Artaban V, represented on sculpture 

at Tak-i Bostan, 220. 
accompanied by a bird of victory, 

76 l . 

Artaxerxes I (Longimanus), 26. 
reputed founder of the fire-temple 

near Isfahan, 259, 260. 
possible tomb of, 296-300. 
Artaxerxes II (Mnemon), tomb of, 

319. 
Artaxerxes III (Ochus), palace of, 

315. 

tomb of, 319. 
Artemis, Anahita identified with, 

240. 

Asadabad, village, 242-3. 
ash-hills, Zoroastrian, 90-8. 
cut down by natives, 91. 
not of volcanic origin, 93 2 . 
good fields for excavation, 98 l . 
at Ahmat, 94. 
at Marand, 36, 54. 
near_Sain Kalah, 121 l . 
Ashirvdd, a formula, 385 1 . 
Ashnar Kuh, mountain range, 251. 
Asmai', al-, mentions Shiz, 135. 
Asmodeus, demon, 169. 
Asnavand, Mount, ancient fire- 
temple on, 139-41. 
believed to be Mount Sahand, 38, 

53. 

possibly Mount Zindan, 141. 
asparagus, derivation of the word, 

30. 

assassin, derivation of the word, 444. 
Astara, town, 446. 



GENERAL INDEX 



451 



astoddn, ossuary, 96-7, 440. 
astrologers, with Kei Khosru, 139. 

consulted at birth, 378. 
Astyages, 26. 

Atash Bahram, at Yezd, 366. 
Atash buzurg kardan, a rite, 396. 
Atash Gahs, ancient, at Naksh-i 

Rustam, 303. 

Atash Kadah, near Isfahan, 342. 
Atash-i mas kartvun, a rite, 396. 
Atash-Parastan, 274, 413. 
audience-hall of Xerxes, 314. 
Auramazda, see Ahura Mazda. 
Aurvant, see Alvand. 
Avesta, the, form and content of, 
67-9. 

translations of, 68 *. 

Geldner's edition of, 363. 

pronunciation of, at Yezd, 363-5. 

manuscripts of, at Yezd, 358. 

no manuscripts of, at Shiraz, 337. 

tradition of archetype copies of, 
68, 306, 307 l . 

chanting of, at Yezd, 367. 

Zoroastrians of Yezd unaware of 
metre in later parts of, 365. 

read by mobeds to cure disease, 
379. 

references to agriculture in, 246-8. 

references to robbers in, 83 l . 

reference to winter in, 81. 

references to conflict with Afra- 
siab in, 138 3 . 

mentioned by Yakut, 343. 

mentioned by Kazvini, 134. 
Avicenna, tomb of, 165-7. 
Avlabar, quarter of Tiflis, 8. 
awning, derivation of the word, 30. 
ayazhdna paiti-darezdna (Av.), sug- 
gested explanation of, 247. 
Azad ad-Daulah, Amir, built a 

causeway near Shiraz, 321. 
dzdj (Arabic, pi. of dzaj), 342 l , 351. 
Azar Kei van, religious teacher, 361. 
Azarbaijan, province, winter in, 71. 

Tabriz the main city of, 39, 40. 



B 



Bab, the (Mirza Ali Mohammed), 

48-9, 328. 

BabaTahir, tomb of, 167. 
Bab-i Roshani, village, 144. 
Babism, 28, 48-50, 375. 
B_actria, 25. 

bad-glr, wind-tower, 349. 
bdgh, garden, 327. 
Bagh-i Shamal, at Tabriz, 50. 



bagldr-bagl, burgomaster, 41. 
Bah-Dlndn, Zoroastrians, 274. 
Bahram, see Varahran. 
bdj-gdh, custom-house, 243. 
bakers, punished for high prices, 273. 
bakhshish, gratuity, 30. 
Baku, Russian city, 4. 
bald khdnah, upper room, 109. 
Baladhuri, al-, mentions Urumiah, 

87. 
Balkh, Zoroaster associated with, 

361. 

Band-i Amir (Bendemeer) , river, 321. 
bandits, 82, 90 S 296, 347. 
Banishun (Wanishun), village, 251. 
Banu-i Fars, shrine of, near Akdah, 

403. 

baodhi (Av.), incense, 370. 
Barashnum ceremony, 381, 392, 396. 
Barashnum Gah at Yezd, 383. 
Barbaro, Josafa, mentions Ardis- 
tan, 406. 

describes Kashan, 403, 411. 

visited Kum, 403. 

mentions Nam, 405. 

visited Yezd, 349. 

refers to ' Abraini' near Yezd, 404. 

itinerary of, from Persepolis to 

Yezd, 352. 

barsom (Av. baresman), 369. 
bas-reliefs, see sculptures. 
bast khdnah, place of refuge, 170. 
baths, at Hamadan, 149. 

at Yezd, 375. 
bazaars, at Hamadan, 148. 

at Isfahan, 267. 

at Kazvin, 444. 

at Shiraz, 326. 

at Tabriz, 44-6. 

at Tiflis, 8, 10. 

at Urumiah, 104. 

Sunday a special day in, 79. 
Bazigrabana, ancient toll-station, 

243. 
Behistan Mountain, the, 186. 

inscription of Darius on, 177, 
181-4, 192-208. 

sculptures on, 180-1, 190. 

blank space on, 187-90. 

ascents of, by Rawlinson, 178. 

ascent of, by the author, 191-6. 

Gotarzes sculpture at, 209. 

newly discovered sculpture at, 

210-2. 

Belinas, 160. 

Bendemeer (Band-i Amir) , river, 321 . 
Berosus, refers to the death of Cy- 
rus, 284. 
bhanddr, pit in a dakhmah, 440. 



^ 



452 



GENERAL INDEX 



bid, willow, 327. 

Birasdd, a prayer, 381. 

birds, 75, 115. 

birth, Zoroastrian ceremonies at, 

378. 
birthday observances, Zoroastrian, 

379. 

Bisitun, see Behistan. 
blocks, stone, of the Tomb of Cy- 
rus, 289, 291. 
of the platform at Meshad-i 

Murghab, 280. 

of the platform near Puzah, 308. 
on the ash-hill at Termani, 93. 
boars, wild, 9. 
boundary stones, 20. 
brass rods, used in Zoroastrian cere- 
monies, 370. 
bread, Persian, 46, 110. 

in Zoroastrian ceremonies, 370. 
bricks, ancient, at Degalah, 92. 
at Hamadan, 162. 
of the fire-temple near Isfahan, 

253, 255. 

at Takht-i Suleiman, 127. 
in the ruins at Rei, 435. 
Bridge of AH Verdi Khan, at Isfa- 
han, 269. 
of Judgment, in Zoroastrian es- 

chatology, 395. 

bridges, building of, a pious act, 75. 
buffaloes, Persian, 5. 
building, square, at Naksh-i Rus- 

tam, 302. 

bulls, winged, at Persepolis, 313. 
Bundahishn, 68. 

mentions Lake Urumiah, 73. 
burial, in Achsemenian times, 305. 
not practised by Zoroastrians 

393. 

butter, melted, in Zoroastrian cere- 
monies, 371. 

Buzo-daghi, mountain, 'Cave of 
Zoroaster' in, 103. 



Cambyses, 26, 43. 

death of, 294 l . 
camels, 21, 33, 109. 

sacrifice of, 162. 

in connection with Zoroaster's 

name, 89, 361. 
caravans, 21. 
caravansarais, at Hamadan, 149. 

at Kermanshah, 232. 
Carduchi, ancestors of present 
Kurds, 89. 



cats, Persian, 149. 
Caucasian Mu