Canonical calligraphy on the Ottoman frontier: Preliminary considerations on thuluth inscriptions from the Balkans

During the month of June 2023, I had the opportunity to conduct field work in Greece and Türkiye thanks to a grant from the École française d’Athènes. Starting in Athens, I travelled to Istanbul via Northern Greece and Turkish Thrace. On the way, the aim was to stop at sites where dedicatory inscriptions dating from the early Ottoman period were possibly still visible and accessible, such as in Thessaloniki, Komotini and Edirne.

Fig. 1. Distribution by country of TEI-referenced inscriptions from the Balkans and Anatolia dated between the 14th and 16th centuries (GIS & CAD : V. Thérouin)

Early Ottoman inscriptions in the Balkans have been known for a long time as documentary sources. Indeed, dedicatory inscriptions affixed to monuments contain valuable information about individuals, including the founder (e.g. name, gender and title reflecting a position in the Ottoman social hierarchy), the function of the building, and a date. The inscriptions were therefore deciphered, edited and studied since the early 20th century. The Thésaurus d’Épigraphie Islamique (or TEI, supported by the Max van Berchem Foundation and the University of Liège) is a remarkable tool for a first approach to this material.

However, the contrast between the number of TEI entries relating to Anatolia and the number relating to the Balkans is striking (Fig. 1). In the TEI, the overwhelming majority of Ottoman inscriptions dating from the 15th and 16th centuries are located in Anatolia. This observation can be seen as the reflection of two dynamics. Firstly, it testifies to the disappearance of Ottoman inscriptions in the Balkans, following the gradual withdrawal of Ottoman rule in the region from the early 19th century. Shifts in political authority, from an Ottoman to a nascent national one, were accompanied by the destruction of Ottoman heritage, which was perceived to be foreign to the region. Secondly, the disappearance of these objects has completely marginalised the Balkans in studies of Ottoman material culture, which tend to limit themselves to the borders of contemporary Türkiye.

Fig. 2. Shaikh Hamdullah ibn Mustafa Dede, Album of Calligraphies Including Poetry and Prophetic Traditions (Hadith), ca. 1500, Made in Turkey, probably Istanbul, Purchase, New York, The Metropolitan Museum of Art, Accession Number: 1982.120.3

Seen primarily as historical texts rather than material objects, Ottoman inscriptions have rarely been approached for the stylistic and morphological features of the scripts. The paucity of art historical approaches to epigraphy might be due to disciplinary compartmentalisation, but also to primacy given to the calligraphic canon. Indeed, these inscriptions were almost exclusively characterized as “thuluth script”. Thuluth is one of the “Six Styles” (ar. al-aqlām al-sittah) initially codified within the Abbasid courts between the 9th and 13th centuries. Later, at the turn of the 15th and 16th centuries, the Ottoman calligraphy school continued to develop this style under the aegis of Şeyh Hamdullah[1] (1436-1520), making thuluth one of the primary display scripts used both in manuscripts and on monuments (Fig. 2). Treatises of Ottoman calligraphers have also played a major role in shaping the understanding of calligraphy by modern scholarship. By relying on them, scientific research inscribed itself in an older discourse applied to the history of calligraphy, perceiving it through a limited number of standardised and unitary stylistic canons.

Still, variations from the calligraphic canon in terms of composition and character shapes are discernible from one inscription to another. In the light of the following examples, documented in situ and subsequently studied with the help of master calligrapher Nuria Garcia Masip, preliminary formal considerations can be suggested. In the present blogpost, a representative panel of fifteenth-century inscriptions, all produced by anonymous calligraphers, will be shown with the aim to highlight the formal diversity of thuluth inscriptions displayed in the Early Ottoman Balkans. In parallel, we will try to answer the following underlying question: what do we mean by thuluth, after all?

          Starting from Athens, I first visited Thessaloniki, in Northern Greece. Despite the 1917 fire and its subsequent reconstruction, Thessaloniki still preserves a significant Ottoman heritage, now promoted by the local Ephorate of Antiquities.

Fig. 3. View of the Heptapyrgion/Yedikule gate, Thessaloniki
Fig. 4. Dedicatory inscription of Murad II, Heptapyrgion/Yedikule gate, Thessaloniki

A first inscription to be mentioned here is the one placed on the portal to the citadel of Thessaloniki known as Heptapyrgion or Yedikule. It is important to note that the inscription was placed in a space already characterised by the accretion of various iconographic narratives relating to the city’s Byzantine, Latin and Venetian pasts (Fig. 3). Bearing an Arabic text, the inscription is dated from 834 AH (1430-1431), the year in which the city was conquered for the second time by the Ottomans, by the sultan Murad II (1421-1444), named in the epigraph.[2] The inscription is carved in bas-relief on a rectangular marble plaque and distributed across five bands, the first two of which are divided by a tuğra, the Ottoman sultans’ personal monogram (Fig. 4). The characters are interwoven but can be distinguished. Maintenance of proportions as well as consistency along the letterforms can be observed, although the ascenders are kept relatively short.

Fig. 5. View of the men’s section portal, hammam of Murad II or “Bey Hamam”, Thessaloniki
Fig. 6. Dedicatory inscription, hammam of Murad II or “Bey Hamam”, Thessaloniki

In the intramural part of Thessaloniki, many other structures founded in the 15th century still present a dedicatory inscription. Among them is the “Bey Hamam” (Fig. 5), founded by Murad II in 848 AH (1444-1445), as indicated on the Arabic inscription placed on the men’s section portal (Fig. 6).[3] The inscription has been carved on a plate in the form of a tabula ansata (although only extending to the right). No band has been drawn but the inscription consists of three lines of interlaced bas-relief characters, punctuated by a few floral motifs (designated as rumi). Respect for proportions and regularity in letterforms characterise the Bey Hamam’s dedicatory inscription as well, despite a noteworthy rigidity and verticality of the shafts (e.g. alif, lam and ṭā’/ẓā’).

Fig. 7. View of İshak Paşa mosque or “Alaca Imaret”, Thessaloniki
Fig. 8. Dedicatory inscription, İshak Paşa mosque or “Alaca Imaret”, Thessaloniki

North-east of this hammam is a structure known as “Alaca imaret” or İshak Paşa mosque (Fig. 7). According to the text of the Arabic inscription placed on the gateway to the central space of the building (Fig. 8), its founder was İshak, son of İbrahim, the grand vizier of Mehmed II (1451-1481) and then Bayezid II (1481-1512). The function of the building (imaret and/then mosque) is specified in the inscription (ar. dār al-khayr)[4] as well as in the vakıfname, both dated to the same year, 889 AH (1484-1485). The inscription is borne by a rectangular plaque and is divided into eight identically sized cartouches with poly-lobed ends. The intertwined characters in bas-relief share consistency in terms of proportions and forms. Like the Bey Hamam inscription, the shafts of several ascending characters tend to be rigid and vertical, however the perfect equilibrium of each composition into eight sections shows a high level of calligraphic mastery and skill.

Fig. 9. View of Gazi Evrenos Bey imaret, Komotini

          On the way from Thessaloniki to Istanbul, I stopped in Komotini (tr. Gümülcine), located further east, in Greek Thrace. This site is home to one of the oldest Ottoman structures still standing in the Balkans, the imaret of Gazi Evrenos Bey (Fig. 9), which was built between 1362-63 (the Ottoman conquest of Komotini) and 1383 (the conquest of Serres, an Ottoman authority pole further west)[5]. The structure now houses the Ecclesiastical Museum of Komotini. According to H. Lowry, two inscriptions are attached to the imaret. The first one, in situ, is placed above one of the entrances leading to the imaret’s interior spaces. The other one is kept in the Komotini Archaeological Museum. Unfortunately, due to restoration work in the imaret as well to the restructuration of the Archaeological Museum’s lapidary collection, both inscriptions were not accessible.

          A final stop was made in Edirne, located in Turkish Thrace. The city was the seat of the Ottoman power from the 1360s until 1453, and afterwards a seasonal residence for the Ottoman elite. For these reasons, Edirne has preserved a significant number of monumental structures, which still display their dedicatory inscriptions. Consequently, this corpus received more detailed attention in scholarship. To illustrate the diversity of the actors involved in commissioning epigraphic productions, the following three inscriptions have been selected.

Fig. 10. View of Gazi Mihal imaret/mosque, Edirne
Fig. 11. Dedicatory inscription, Gazi Mihal imaret/mosque, Edirne

The first inscription is placed on the gateway to the main room of a building today known as the imaret or mosque of Gazi Mihal (Fig. 10). Its founder was the beylerbey Mihal, son of Aziz, as stated in the Arabic text of the inscription dated 825 AH (1421-1422)[6] (Fig. 11). This inscription is placed on a rectangular marble plaque divided into three bands. Unlike the inscriptions presented so far, the bas-relief interlaced characters do not share a regular system of proportion. This irregularity of  shapes can be seen in the variations for a single character (waw, initial ʿain/ghayn, nūn). Furthermore, certain shapes are shared by characters which are usually clearly distinct (nūn and rāʾ/zāʾ). It is therefore conceivable that the inscription was executed by an amateur hand that did not master thuluth script, possibly by a self-taught local calligrapher.

Fig. 12. Dedicatory inscription of Murad II, Üç Şerefeli mosque, Edirne

Our second case study concentrates on the epigraphic programme of the mosque known as “Üç Şerefeli” (i.e. with minarets featuring three balconies), a milestone in the history of Ottoman architecture. The Üç Şerefeli mosque was founded by Murad II, as indicated by the Arabic text of the inscription dated 849 AH (1445-1446)[7] (Fig. 12). This dedicatory inscription is located on the entrance portal to the prayer room, placed on a long rectangular plaque of painted marble. The inscription is not divided into bands or cartouches but is arranged in a complex, three level thuluth composition with one line written in kufic along the upper part. The gilded writing is set against a background of rumi motifs arranged in spirals. The bas-relief interwoven cursive characters share coherent proportions and regular shapes. However, the main feature of this inscription is the elongation of the shafts of alif, lam and ṭā’/ẓā’ which creates a notable ascending effect. Combining two scripts, thuluth and kufic, this inscription is reminiscent of Timurid inscriptions, echoed in the Ottoman epigraphic productions of the 14th and 15th centuries.

Fig. 13. Dedicatory inscription, Sittşah Mürkrime Hatun mosque, Edirne

The third and final inscription comes from the time when Edirne was no longer the Ottoman capital. This is the dedicatory inscription placed on the gateway to the prayer hall of the mosque of Sitti Şah, whose name is mentioned in the Arabic text dated 889 AH (1484-1485)[8] (Fig 13). The founder is Sittşah Mürkrime Hatun, wife of Mehmed II. The inscription is carved on a small rectangular plaque that is divided into three bands. The closely tightened bas-relief characters do not share similar proportions, and the composition is characterized by general irregularity. As noted by F. T. Dijkema and A. Tüfekçioğlu[9], the inscription contains elements of mirror writing (ar. muthanna), as is the case for the word “sulṭān”, as well as the isolated waw and other letters.

In conclusion, despite the formal variety, the fact that all these inscriptions can be characterised as thuluth leads us to question the morphological features of this canonical script. Regular linear thuluth, as developed in manuscripts, is characterised by ascending letterforms slightly inclined to the left, descending letterforms going well below the baseline, and groups of characters sitting slightly oblique on the baseline, underpinning the following characters. In addition to this, certain letterforms are particular to the thuluth style such as the three different forms for rāʾ and zāʾ,  the two forms of mīm or wāw, as well as certain elongations and the large looped tails of jīm and ‘ain[10]. This formal variety has thus made thuluth known as a “display” script, making it particularly flexible for its use in larger compositions. In the epigraphic context, the thuluth script is therefore particularly useful, characterised by its flexibility and ability to adapt to the context, while maintaining formal coherence, as any canonical style.

When examining epigraphy, the stylistic differences across inscriptions are therefore often revealed by the composition of the inscription, and its inner regularity. Rather than focusing solely on specific letter forms, attending to the use of space and compositions can reveal certain peculiarities. As we have seen from this limited corpus, there are distinct degrees of entanglement between letters from one inscription to the other, a gradation in structuring compositions around the vertical shafts of the ascenders, variations in the use of the non-inscribed space, as well as a differences in preserving the letter proportions, with more or less success.

How then do we explain such calligraphic variation in epigraphic production? Causes may lie in the context in which the inscriptions were produced. As such, the verticality of certain inscriptions may be related to stylistic influences on Ottoman material production particular to the 15th century, such as Mamluk or Timurid influences, which raises questions about the circulation of patterns. The respect, or lack thereof, for the exact canonical proportions could be linked to proficiency in calligraphy, which may lead us to identifying the possible producers (as calligrapher, scribe, scholar, etc.). By enlarging the corpus of study and linking descriptions of inscriptions with other material (architecture, status of patrons, resources of pious foundations, etc.), we hope this work will be a first step in contextualising the epigraphic productions of the early Ottoman Balkans.

Vincent Thérouin

[1] Considered to be the founding father of the Ottoman calligraphy school.

[2] Hellenic Ministry of Culture and Directorate of Byzantine and Post-Byzantine Antiquities, eds., Ottoman Architecture in Greece (Athènes, 2008), 216–18. ; TEI entry No. 33218.

[3] Hellenic Ministry of Culture and Directorate of Byzantine and Post-Byzantine Antiquities, Ottoman Architecture in Greece, 232–34. ; TEI entry No. 32972.

[4] Hellenic Ministry of Culture and Directorate of Byzantine and Post-Byzantine Antiquities, Ottoman Architecture in Greece, 229–31. ; TEI entry No. 40507.

[5] Heath W. Lowry, The Shaping of the Ottoman Balkans, 1350-1550: The Conquest, Settlement & Infrastructural Development of Northern Greece (Istanbul, Turquie, 2008), 41–46, 80.

[6] Fokke Theodoor Dijkema, The Ottoman historical monumental inscriptions in Edirne (Leiden, Pays-Bas, 1977), 17–18. ; TEI entry No. 9515.

[7] Dijkema, The Ottoman historical monumental inscriptions in Edirne, 26–27. ; TEI entry No. 849.

[8] Dijkema, The Ottoman historical monumental inscriptions in Edirne, 42–44. ; TEI entry No. 29380.

[9] Abdülhamit Tüfekçioğlu, ‘Symmetrical Compositions in Pre-Ottoman and Ottoman Architectural Inscriptions in Asia Minor’, in Calligraphy and architecture in the Muslim world, by Mohammad Gharipour and Irvin Cemil Schick (Edimbourg, 2013).

[10] Such variations are not the results of the calligrapher’s own interventions: on the contrary, they are precisely codified and standardised, forming a limited repertoire.

OpenEdition vous propose de citer ce billet de la manière suivante :
callfront (1 février 2024). Canonical calligraphy on the Ottoman frontier: Preliminary considerations on thuluth inscriptions from the Balkans. CallFront. Consulté le 25 juillet 2024 à l’adresse

Laisser un commentaire

Votre adresse e-mail ne sera pas publiée. Les champs obligatoires sont indiqués avec *

Ce site utilise Akismet pour réduire les indésirables. En savoir plus sur comment les données de vos commentaires sont utilisées.

Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search