As we have already seen, the first paid Ottoman infantry force, probably only established after 1340, and possibly after 1350, comprised the yaya or piyada. This was in truth no more than a levy of tax-exempted Turkish irregulars raised for occasional siege operations, paid an asper a day for the duration of a campaign, and disbanded on its completion. Although properly organised in units of 10, 100 and 1,000 men whilst on active service, they remained no more than a booty-seeking rabble, occasionally even being referred to as ‘vagrants’, and proved something of a failure in military terms. They were therefore replaced soon after by a new, standing army called the Yeni-ceri (‘New troops’), a name which later western writers were to corrupt into ‘Janissary’. This was most probably founded in 1362, by the application of the law of ghanimat (booty) being extended to include prisoners, one-fifth of whom were subsequently to be surrendered to the sultan.2 The Janissaries were therefore a slave army initially made up of Christian prisoners converted to Islam; Doukas reports that ‘among them could be found neither Turk nor Arab, but all of them without exception were Christians — Byzantines, Serbs, Albanians, Bulgarians, Wallachians, and Hungarians.’ The practice of drafting such prisoners into the ranks of the Janissaries continued throughout this period (for example 320 Serbs when Novo Brdo fell in 1455, 800 Trapezuntine youths on the fall of Trebizond in 1461, and 1,500 Genoese after the fall of Kaffa in 1475), but by the 1390s at the very latest a secondary means of maintaining their numbers was also being utilised. This was the devshirme, a levy of children and youths forcibly imposed on the sultan’s Christian subjects. We only first read of it in a sermon of 1395 by Bishop Isidore Glabas of Thessalonika (which was at that time in Turkish hands), but according to Idris al-Bitlis, writing at the very beginning of the 16th century, it may have been introduced simultaneously with the foundation of the Janissaries in Murad I’s reign. After 1395, however, we only next encounter any references to the practice in 1430 and 1438, which has led to the suggestion that it may have been ‘suspended in the years of confusion following the Battle of Ankara’, and reintroduced under Murad II as part of the reorganisation of the Janissaries — mentioned but, alas, not described by Pseudo-Sphrantzes — which took place after the Janissary mutiny of 1446. Certainly the devshirme (which the Byzantines called paidomazoma) was well-established by the 1430-60 period, as witnessed by the appearance of immunity from it as a concession in several terms of capitulation (e. g., loannina in 1430 and Galata in 1453). At first it was seemingly only levied on an occasional basis when the number of boys taken captive in war was insufficient to maintain the Janissaries’ numbers, but by the mid-15th century it appears to have been levied on a regular 5-yearly cycle. European sources record the age limits of those taken by devshirme as 8-20, but later Ottoman sources put the lower limit at 15. Certainly Jacopo de Promontorio reports c. 1475 that 2,500 15-year-old boys were taken by the Ottomans ‘each year’, of whom 1,500 were drafted into the Janissaries (the remainder probably going into the Qapu Khalqi, for which see below). Konstantin Mihailovic says in his ‘Memoirs’ that there were normally ‘about 2,000 of these boys’ under training. He explains that they were sent ‘across the sea’, i. e. to Anatolia, where individual sipahis were responsible for teaching them at their own expense. Other sources give fuller details, claiming that such trainees were placed at the disposal of sipahis in both Anatolia and Rumelia for their first 5-7 years of service in order to learn the Turkish language, customs and religion, and were then posted to Gallipoli (or to Constantinople after 1453) and subsequently selected for palace or military duty according to their abilities, most ending up as Janissaries. Mihailovic himself only appears to have received about one year’s training, but he was not actually a Janissary (he appears rather to have been some son of quarter-master). His account of standard Janissary training says that the authorities ‘ship back [from Anatolia] those who are suitable and these study and train to skirmish in battle. Already the sultan provides for them and pays them a wage. From there he chooses for his own court those who are trained and then raises their wages.’
The Janissary corps was properly called an ocak (‘hearth’). By the latter part of this period it comprised two distinct bodies, these being the Cema’at (translated as ‘Company’ by many modern authorities, but as ‘Reunion’ or ‘Assembly’ in the Encyclopaedia of Islam), and the Segban (‘Hound-keepers’ or ‘Huntsmen’, popularly called the Segmens), which was added following a Janissary revolt soon after the capture of Constantinople. A third division, the Boluk, was later added by Selim I (1512-20). Of these three, the Cema’at remained the largest element even in the 16th century, though the Boluk was by then pre-eminent. The Cema’at’s original function appears to have been to provide the fortress garrisons that were established in conquered provinces during the 14th century. Mouradja d’Ohsson, for instance, wrote at the end of the 18th century that one Cema’at orta (see below) had been based in Vidin ever since its occupation in 1396. It was therefore undoubtedly the Cema’at rather than the Segban that provided the Janissary garrisons that we often read of being put into captured towns and fortresses in the second half of the 15th century — for example, 400 each into Corinth and Sparta in 1460 according to Kritovoulos. The overall command of the Janissary corps was in the hands of the Yeni-ceri Aghasi, or Aga of the Janissaries, with the commanders of the Segban and, later, the Boluk (the Segban-bashi and Kul kahyasi respectively) as his adjutants. After the incorporation of the Segban into the Janissaries the Aga was normally chosen from among its ranks. Other senior officers in order of rank included the Zagharci-bashi (‘Chief of the bloodhound-keepers’), Muhzir-aga (‘Chief of the ushers’). Senior and Junior Khasseki, and the Bash-cavus (‘Chief of the sergeants’).
Cema’at and Segban were each subdivided into companies, called orta, of which there were 165 after the incorporation of the Boluk, rising to 196 by the reign of Suleiman the Magnificent (1520-66). Of this latter total 101 ortas were Cema’at and 34 Segban, figures which may also apply for the 15th century. Chalkokon-dyles, writing of Mehmed IPs reign, states that ‘the ortas had each an establishment of 50 men’, but their strength probably varied with the passage of time — Mihailovic, for instance, writing in Bayezid IPs reign, mentions Janissary ‘centurions’ (though he also mentions that the troop of Janissaries with which he was provided as commander of Jajce was 50-strong). The smallest unit was a squad of 10 men; the ‘Ordo Portae’, a survey of the Ottoman army written in vulgar Greek between 1473-81, says that each 10 Janissaries received a horse, tent and trunk, adding that if one of the Janissaries died he was replaced immediately ‘so that their numbers are always complete.’ (This source also tells us that the Janissaries were paid 3-5 aspers per day and their Aga 100 aspers, and we know from other 15th century sources that these salaries were paid out every 3 months.) Each orta was commanded by a corbaci (‘soup-maker’), assisted by a staff comprising oda-bashi (‘chief of the barrack room’, i. e., adjutant); wakil-khardj (‘controller of the expenditure’, or company clerk); beyrak-dar (standard-bearer); bash-eski (‘chief of the veterans’, the oldest soldier in the orta); ashci-bashi (‘chief cook’, i. e., quarter-master); and sakka-bashi (‘chief water-carrier’). Some orta commanders also commanded larger elements of the entire corps. Promotion was by seniority.
As to the numerical strength of the Janissaries, contemporary accounts and modern estimates vary wildly. Some of the latter give them no more than 1,000 men under Murad I and Bayezid I, and only 1,200 even under Mehmed II. However, what evidence there is for such statements is not made clear. Certainly it is undeniable that their numbers in the 14th century may have initially been small (some contemporary accounts do not mention them at all), but there is no indication that they were ever that small. Bertrandon de la Brocquifere, whose account seems reliable, does not mention the Janissaries in his chronicle (unless the infantry amongst the 5,000 household troops he records are they), but at a later date (c. l440) he wrote that the Jehaniceres were 10,000-strong, though he admits that he was given the figure by a Greek whom he met. The figure of 3,000 is given for Murad IPs reign by the ‘Ordo Portae’, which adds that his successor Mehmed II raised their numbers first to 5,000 men, and then to 10,000 during the war against the White Sheep Turks (1472-73); however, of these 2,000 were killed in the campaign so that when the war ended they stood at 8,000. To a certain extent this version of events is corroborated by other accounts — Chalkokon-dyles, for instance, puts Janissary strength in the latter part of Mehmed IPs reign at 6-10,000, while a letter written to the pope by the Venetian Laurus Quirino in 1464 mentions 7,000 Janissaries. Our eye-witness Mihailovic, the so-called ‘Serbian Janissary’, writing of c. 1455-63, says that there were only ‘about 4,000’, though he qualifies this by saying that these were at the sultan’s court, thereby excluding from his figure those posted in the provincial fortresses. Doukas mentions the figure of 10,000 Janissaries repeatedly in his chronicle — at Ankara in 1402 (where an anonymous contemporary Ottoman account mentions 5,000 to have been present), at the siege of Constantinople in 1453, and ‘now’ (c. l461) — but it should be borne in mind that he may be referring to the Qapu Khalqi in general rather than the Janissaries alone (see below). Alvise Sagudino, another Venetian, this time writing in 1496, says there were 8,000 Janissaries, while Janus Lascaris, who visited the Levant twice between 1484-92, provides the interesting statement that they numbered 7-10,000, the latter when at full strength. This may be the key to the fluctuating figures found in sources of the second half of the century, which may alternatively be explained by a deliberate policy of increasing Janissary strength prior to major campaigns. All in all it seems certain that the strength of the Janissary corps increased steadily throughout the reigns of Mehmed II and Bayezid II, from 5,000 at midcentury, to 6-7,000 during the 1460s, and 8-10,000 for the rest of the century. There are some even larger numbers quoted in the sources: Leonard of Chios (1453) and Arnold von Harff (1499) claim 15,000 and
32,000 Janissaries respectively, but these figures are totally unconvincing.
In closing, it should be noted that the appearance of the Janissary ocak in Murad Ps reign did not automatically result in the total disappearance of the yaya, who are mentioned as being present at the Battle of Kossovo Pole in 1389, and fighting for Murad II both in the civil war of 1421 and in his attack on Corinth in 1446. However, at least some of these occasional references may nevertheless be allusions to the Janissaries, since one source says that the men of the Cema’at were themselves often called yaya beyleri. Indeed, if the reference to yaya at Kossovo is not intended to denote the Janissaries, then neither they nor the Qapu Khalqi are mentioned in a single contemporary source as being present at that decisive battle (though a later source allegedly mentions that the Ottoman van included ‘the Aga of the Janissaries with his 2,000 archers’).
World History
