The skirmish is described in varying levels of detail by four separate chroniclers, who all attest to different dates. The Arab historian
Ibn al-Athir maintains that it took place after the
execution of a caravan at
Otrar in 1218. al-Athir's near-contemporary
Nasawi explicitly corrects him by attesting that the battle took place in the year 612 of the
Islamic calendar, or 1215–6. The later Persian historian
Juzjani gives the date of 615 (1218); while
Juvayni records that the Shah departed on his campaign at the end of 1218, giving a date of 1219 for the battle itself. Adding to the confusion, all the chronicles contain errors of differing magnitude: for example, Nasawi indicates that the battle came after the defeat of
Kuchlug, which is known to have happened no earlier than 1218; by contrast, Juvayni's account states that the Sultan remained in
Bukhara from 30 October to 30 December "because it was springtime" — a contradiction in itself. The later historian
Rashid al-Din also provides an account of the Mongol engagements with the
Merkit tribes in his
Jami' al-tawarikh; however, his chronology of Mongol interactions with the Khwarazmids is suspect because his chronicle misses out years. in 1215 AD. The Irghiz river, the site of the battle, is located on the Khwarazmian border, north of the
Aral Sea.|alt=Map of the Khwarazmian Empire. Some historians propose that the battle took place in 1209. Drawing upon sources such as the
Secret History of the Mongols and extant biographies of the general
Subutai and citing similarities between different campaigns in Juvayni's account, they suggest that the skirmish took place just after Subutai had defeated a
Merkit confederation. Because Kuchlug, then a vassal of the
Qara-Khitai, managed to seize the key city of
Samarkand in 1210, Paul Buell postulates that the indecisive result at the Irghiz River weakened Muhammad's reputation and made the otherwise inexplicable loss of Samarkand possible. According to Christopher Atwood, it is certain that both the Shah and the Mongols were campaigning in the drainage area of the
Syr Darya in 1209/10, which lends this account credence in geographical terms; it is however supported by fewer sources than the other version. Other historians suggest that the battle took place in 1218/19, following the accounts set down by al-Athir and Juvayni: this version emphasises the
deterioration in relations between Shah Muhammad and
Genghis Khan.
Jochi is present in this version. It is recorded that Genghis Khan praised Jochi's leadership on his return from this campaign; historians have assumed that a large battle against a high-quality enemy such as the Shah would be more worthy of high praise than the more routine elimination of the Merkits. Carl Sverdrup suggests that the Shah, who already knew that his provocative diplomatic behaviour would result in a retaliatory Mongol strike, seized the opportunity to weaken both the Merkit raiders and his future Mongol enemies. This is supported by a passage in Juvayni, who records that the Shah wished "to kill two birds with one stone", implying that he had already decided on open hostilities. However, there are logistical difficulties with this account — namely that
Jebe was near
Wakhan in modern
Afghanistan when the disgraced Kuchlug was killed in late 1218, but the Irghiz River flows nearly a thousand miles away in modern
Kazakhstan. == Battle ==