Place-names, tree names and linguistic archaeology in Swāt. A conversation with Matteo De Chiara

Source: M. De Chiara

I met Matteo De Chiara in 2016, at the Mondes Indien et Iranien CNRS research unit just outside of Paris, home to several researchers studying the languages, histories and geographies of West, South and Central Asia, as well as to the excellent Cartorient mapping project. A renown specialist of contemporary and ancient Iranian languages, Matteo teaches Pashto at the Institut National des Langues et Civilisation Orientales (INALCO) and has notably published a volume titled Iranian Languages and Literatures of Central Asia (from the eighteenth century to the present)  (with Evelin Grassi, 2015). His more recent work has seen him delve into the linguistic geography of Pakistan’s Swāt valley and develop a fascinating study of its toponyms. Matteo kindly acepted to tell me more about this research project, his methodology and interest for language geography.


Can you introduce yourself in a few words ? What is your background ?

My name is Matteo De Chiara, an Italian from Rome. My mother tongue is Italian. I am a linguist by training: I fulfilled my Master studies in linguistics and Sanskrit at the University of Rome “La Sapienza”. After that, I pursued doctoral studies at the University of Naples “L’Orientale” in Iranian studies and, in particular, in Pashto historical phonetics and lexicology under the supervision of A.V. Rossi. Now, I am Maître de Conférences HDR of Pashto and Iranian Philology at the Inalco of Paris.


What is your current research topic about? How do you use linguistic maps?

At present, I’m continuing my linguistic research on Pashto. Recently my volume on the toponymy of the Swāt valley – a district in the province of Khyber-Pukhtunkhwa, north of Pakistan (Matteo De Chiara, Toponymy of the Swāt Valley. Linguistic archaeology, Lahore 2020) – has been published by the editor Sang-e Meel and thanks to the funding of the Italian Archaeological Mission in Swāt, the Ismeo of Rome and the Inalco in Paris. I have been working on this research project for at least the last seven years, under the auspices of the Italian Archaeological Mission in Swāt, founded in 1956 by Giuseppe Tucci, and now directed by L.M. Olivieri, and thanks to funding from the Inalco of Paris. The main goal of this work is to offer new materials on the linguistic and ethnic situation of Swāt from a diachronic point of view.

At the beginning of this project, all toponyms were originally sourced from the Government of Pakistan’s maps from 1950 and 1984. The data was later verified with the 1977 maps from the Soviet Union and from the US Army, which were consulted and used to integrate the cartographic base, and lastly, with Google maps and Google Earth, before being confirmed, corrected and improved with the help of Pashto speakers and residents in loco.

Swāt toponyms have a long history and have been treated in many sources since the 19th century and even before, not only in all English gazetteers, but also in the numerous archaeological surveys. Nowadays, the main language spoken in the Swāt valley is Pashto, in particular the Yusufzai dialect, with Urdu as national language and some minor Indo-Aryan languages, like Gujari. In the northern part of the valley two Dardic languages are also spoken, Torwali and Bashkarik, representing the last remnants of a Dardic substrate.

In recent times, at the time of the “Yusufzai State of Swāt”, the ruler of Swāt “Pashtunised” many village names, and hence the importance of the historical sources. However, notwithstanding this reform, non-Pashto names still exist today. In many cases, very ancient toponyms remain for modern designations, especially with regards to geonyms, i.e. the generic names of place, where we find names of non-Pashto origin – the majority Indo-Aryan.

Starting then from a review of all toponyms and thanks to the program QGIS, I could draw the maps of all toponyms, Tehsil (sub-divisions of the District) by Tehsil and typology by typology, according to their linguistic origin. As an example, I include here the map containing the etymology of rivers.


Source : M. De Chiara


What initially lead you to develop an interest in toponyms and geonyms, as well as for linguistic maps?

During my doctoral studies, I spent one year in Paris to study the Pashto language (Paris is the only place in Europe where a stable academic master exists). After that, I was invited to Swāt by the director of the Archaeological mission to Swāt, L.M. Olivieri, to improve my Pashto proficiency. I stayed for four months in Swāt in a Pashtun environment. This was at the end of 2006, and I left a few days before the start of the war of 2007-2008 in Swāt. Some years after that, L.M. Olivieri, who developed an innovative methodological model of study for this region, based on a multidisciplinary approach, proposed that I study the toponyms of the Swāt valley.

As toponyms are the names of places, this project obliged me to develop an interest in geography and linguistic geography. I must confess that I was never interested to these topics before. However, thanks to this research, I discovered how fundamental an understanding of the physical features of the environment together with the dynamics of the history are, in order to appreciate linguistic contact: this is particularly true in a milieu of strong multilingualism.

I spent some time, guided by Emmanuel Giraudet of my research unit at the CNRS (UMR 8041 – CERMI), to learn how to use the program QGIS, which enabled me to firstly draw geographic maps, and then the resulting linguistic maps.

As a consequence of this project, I now see how important linguistic maps are for the identification of isoglosses. In my specific case, as I dealt mainly with a single dialect, i.e. Yusufzai Pashto, I individuated some main lexical isoglosses existing in the toponymy of the Swāt valley.

Another interesting issue, which I am confronting at the moment, are the geonyms, i.e. the generic names of places, as ‘mountain’, ‘valley’, etc., from which one can derive useful information on linguistic geography and on toponymical dynamics as well as methodological suggestions on the semantics and etymology of toponyms.


Source: Designer429/Wikimedia Commons


What sources do you use? To which extent do you rely on fieldwork, and materials by other scholars?

As stated above, I firstly relied on the maps of the Pakistani Government, of the Soviet Union and of the US Army, implemented by Google maps and Google Earth. The maps contained in the archaeological surveys of the first part of the 20th century (in particular to those contained in the books by A. Stein and G. Tucci), as well as the more ancient Gazetteers of the 19th century (cf. for instance those by H.G. Raverty and H.W. Bellew), also provided useful and complete data.

However, all of this without checking with Pashto speakers and residents in loco would have been useless. In order to verify all data and to take an advantage of their linguistic expertise, I brought three missions to Swāt, in particular to Saidu Sharif, not far from Mingora, where the headquarters of the Italian Archaeological Mission are located. In Swāt (I travelled up to Bahrain, in the north of the valley) I could verify my material with language informants: I always received excellent hospitality and warm friendship. I am really grateful for all the encouragement and information I received, and for the many agreeable moments I had there during my visits.

The next step of my research was to study in detail the toponyms in two directions: from one side, a comparison with the typology of toponyms, from the other, the linguistic analysis proper. For the first task I compared Gazetteers and toponymical works near or further from my context: for instance, I carefully consulted the useful work of L.W. Adamec, Historical and political gazetteer of Afghanistan, in six volumes; the works of W. Eilers and M. Mokri, among others, on the toponymy in other Iranian languages; and also the Atlante toponomastico della provincia di Cremona, edited by V. Ferrari, dealing with the toponymy of a province in northern Italy. I must cite here also the work of W.Kh. Hakimzay, in Pashto, entitled Daɣa zmunǧ kәlay day [This is our village], which provides useful etymologies, not always reliable, but pioneering, even if quite neglected in Swāt.

For the second task, I consulted dictionaries and repertories of more or less local languages, belonging to many different linguistic families: Iranian, Dardic and Kafiri, Indo-Aryan, Turkish, Mongolian, Dravidian, Muṇḍa (Austroasiatic).

L.O. Olivieri provided me with useful data, such as local chronologies, toponyms of Alexander’s time, etc.: all of this doubtless facilitated my research.

Source: M. De Chiara


According to you, what can your approach bring to the study of linguistic geography?

A multidisciplinary approach, including geography, linguistics, history, archaeology, but also botany and others, can doubtless bring new fuel to the study of linguistic geography. For instance, in my research I identified many names of pastures, passes and villages containing tree names, as shown in this partial list:

  • °Banr (baṇ) ‘forest’
  • Amlukbanr (amlukbaṇ) ‘forest of the sloes’
  • Asharbanr (ašārbaṇ) ‘forest of the silver firs’
  • Ghuzbanr (γuzbaṇ) ‘forest of the walnuts’
  • Gujarbanr (guǰarbaṇ) ‘forest of the Gujars’
  • Loebanr (loybaṇ) ‘big forest’
  • Mabanr (mābaṇ) ‘garden of the moon’
  • Surbanr (surbaṇ) ‘red forest’
  • Tut banrin (tut baṇin) ‘densely wooded by mulberries’
  • Achro Banda (ačro bānda) ‘pasture of the silver firs’
  • Ranjro Sar (ranǰro sar) ‘mountain of the deodars’
  • Nakhtar (naxtar) ‘Himalayan pine’
  • Talgo Nakhtar ‘id.’
  • Shaurai Khwar (šawray xwaṛ) ‘river of the small sissoo’?
  • Banj Banda (banǰ bānda) ‘pasture of the quercus ilex’
  • Tarkana (tarkāna) ‘maple’
  • Baret (baret) ‘bird cherry’

In order to do that, I also consulted the volumes by M. Alam on Trees and Shrubs in Afghanistan (2011) and Sultan-i-Rome on Land and Forest Governance in Swat. Transition from Tribal System to State to Pakistan (2016), as well as Qadeer M.A. Khan’s thesis (n.p.) on Working Plan for the Forests of Swat and Swat-Kohistan, Swat State (Malakand Agency, 1965-1980). The approach of F. Barth in his article on the “Ecologic Relationships of Ethnic Groups in Swat, North Pakistan” (1956) is also important.



This has consequences on our understanding of the economic processes as well as on the drawing of linguistic maps. In this way, I think that my approach can really contribute to linguistic geography and, in particular, to the lexical domain.

Clearly, a more in-depth investigation, valley by valley, could also enhance our knowledge. In this concern, I think that my future research in this domain will follow two paths: a micro-toponymy collection, devoted to local names and traditions; and a macro-toponymy collection, outlining more general comparisons and historical influences.

All these aspects deserve further investigation and my hope is that this work will lead to new interest and energy, and will be considered not a conclusive result, but an open and collaborative starting point.


Source: M. De Chiara


Can you name a couple of books about linguistic geography that you would recommend and explain why in a few words?

I will leave aside the classics of linguistic geography, which already influenced the research in this domain. I would like to mention an inspiring book for me, and a continuous source of materials: the two volumes of G. Fussman’s Atlas linguistique des parlers Dardes et Kafirs, published in 1972 in Paris.

The first volume contains 167 linguistic maps, arranged word by word, as ‘tree’, ‘cow’, ‘summertime’ or ‘to see’. The area covered by the maps goes from the north of Kabul in the west, to Gilgit and Srinagar in the east, and from the springs of the Chitral river in the north, to Peshawar and Parachinar in the south. The second volume contains the commentary to the linguistic maps. Every map indicates the synonyms for the words in the different Dardic and Kafir languages spoken in the area, in addition to  languages with a different origin: Iranian, Indo-Aryan, Turkic, Burushaski. The three useful indexes contain linguistic matters, the cited words and phonetic treatments.

Not only does this work concern exactly the region I am interested in, and represents an inevitable tool for all those dealing with this area, but it is also an important methodological tool. Starting with the linguistic maps, G. Fussman draws verified and reliable conclusions on the movements of the different peoples. For instance, in my work I could verify the accuracy of Fussman’s analysis when he stated that the Damelis and the Gawars would have been expulsed from Swāt “vers le milieu du XVe siècle par les Pathans” (Fussman 1972, I: 23).

The second work I like to name is the Atlas Linguistique de l’Afgahnistan. Unfortunately, it has never been published, but its materials are kept at the archives of Berne. This remarkable research project intended to provide linguistic maps for the whole Afghanistan. In order to do that, many questionaries were submitted to language informants and several scholars took part in the undertaking, including G. Morgenstierne, one of the leading researchers of Iranian philology in the 20th century.


Source :

Fortunately, an Arbeitspapier was published in Berne in 1974 by the three authors, G. Redard, Ch.M. Kieffer and S. Sana: L’atlas linguistique des parlers iraniens: Atlas de l’Afghanistan. In particular,  the article by Ch.M. Kieffer is remarkable: “L’établissement des cartes phonétiques: premiers résultats”. This work of linguistic geography shows its usefulness and possible developments, as it appears from a draft of a paper presented by Ch.M. Kieffer in the Colloque sur l’établissement d’un Centre d’étude de la langue et de la littérature paṣ̌tō, Caboul, 22-29 novembre 1975, and entitled “Dialectologie du paṣ̌tō. Distribution et typologie des parlers paṣ̌tō d’après les premières cartes de l’Atlas linguistique de l’Afghanistan (ALA)”. My hope is that this work will one day see the light.

Moreover, this project was not isolated: indeed, starting from the ’50s, G. Redard himself had also begun preliminary work for a linguistic atlas of Iran and has published many articles on Iranian dialectology.



See also: De Chiara, M. (2019). Swāt Hydronymy at the Border between Iranian and Indo-Aryan Languages, Iran and the Caucasus, 23(1), 64-74. doi:

The Catalan Atlas, transcriptions, translations and toponyms

I have always loved the 14th century masterpiece now known as the Catalan Atlas. I never saw it with my own eyes though, even when I worked at its home, the Bibliothèque nationale de France, but whether online or in print, it still is an enduring source of awe and inspiration to this day and I regularly find myself coming back to its immense breadth, to the simple clarity of its era-defining cartographical language and to its colourful, mysterious, illustrations.


Thanks to a recent Twitter thread by the excellent press cartographer Jules Grandin, I just discovered the Cresques project, a webpage by aerospace engineer and independent map scholar Juan Ceva. Based in California but originally from Spain’s Majorca island, Ceva has made available fascinating materials and secondary literature, notably by Gabriel Llompart and Jaume Riera.

The illustrious Catalan Atlas was produced in or around 1375, very probably by the Majorcan cartographer Abraham Cresques and possibly his son. Acquired early on by the French monarchy, the portulan chart in 6 sheets is still held in the collections of the Bibliothèque nationale de France‘s manuscripts department and has been the subject of many exhibitions and descriptions. While many mysteries still surround this piece of work and its author, Ceva’s contribution is very enlightening for the non-expert map enthusiast that I am, notably because it provides keys to decipher and understand descriptions and toponyms from the richly decorated maps, for instance in the top left of panel 3: “En Inbèrnia ha moltes illes merevellosas…“.


The full inscription is translated : “In Hibernia [Ireland] there are many wonderful islands whose existence can be credible; among them, there is a small one where men never die, because when they are about to die of old age, they are transported outside the island.”



Other fascinating and very helpful parts of this webpage include transcriptions and translations of the legends as well as of Balearic toponyms. I am by no means a specialist and pre-18th century maps do not relate to my own research work but those toponyms and this map bring back very fond memories of Menorca Island, where I spent time with my grandparents and parents as a child, near Ciutadella, the Catalan Atlas“citaela”.


Congratulations to Juan Ceva for making the language of those maps easily understandable by us all and stimulating original research on this fascinating piece of cartographical history. May it inspire others to do the same for masterpieces such as the ethno-linguistic map accompanying the Dīwān Lughāt al-Turk [ديوان لغات الترك], Mahmud al-Kashgari’s landmark 11th century encylopedia of the Turkic languages and people.



Abelhas e Mapas? : on Valéry, Gilliéron and a “Language and Territory” conference in 2021


A colloquium will be held in the French region of Occitanie next year from 14 to 20 June to “discuss the different ways that languages and territories are linked, and will show the political, social and economic stakes that arise from the relationships between them”.

Paper proposals -in English, French, Spanish or Occitan- can deal with the themes detailed in the whopping list below:

  • “Language and Virtual Territory
  • Language, Territory and Planning
  • Language, Territory and Contacts
  • Language, Territories and Conflict
  • Language, Territory and Law
  • Language, Territory and Economy
  • Language, Territory and Education
  • Language, Territory and History
  • Language, Territory and Identity
  • Language, Territory and Imagination
  • Language, Territory and Literature
  • Language, Territory and Minorities
  • Language, Territory and Mobility
  • Language, Territory and Globalization
  • Language, Territory and Digital Technology
  • Language, Territory and Politics
  • Language, Territory and Philosophy
  • Language, Territory and Nation
  • Language, Territory and First Nations
  • Language, Territory and Religion
  • Language, Territory and Networks
  • Language, Territory and Companies
  • New language(s), new Territory(ies)
  • Languages, Territory and Writing
  • Language “without territory”, territory “without language”.
  • Language, Territory and Heritage”

Deadline for proposals is 21 November 2020. More information here:

I will do my best to attend and will post more details when I receive the full program next Spring. The event, held at Montpellier’s University, which is named after one of my favorite poets, Paul Valéry (1871-1945), will no doubt be of interest for linguists, geographers and accross the social sciences.

Valéry never wrote about linguistic geography, nor did he seem familiar with its development, but it is a pleasant coincidence that “L’abeille” (“The Bee”), one of my favorite poems of his, was published in Charmes (1922) only a couple months after Jules Gilliéron (1854-1926) published his “Complementary notes on the name of the bee”, a sequel to his landmark study Généalogie des mots qui désignent l’abeille [full pdf from Gallica], from the Atlas Linguistique Linguistique de la France (ALF) (1902-1910/1920).

François Jacquesson mentioned this classic bee example on his excellent blog last year, and Mathieu Avanzi published his own take on Jules Gilliéron and Edmont Edmont’s bee on Twitter last month!











A very useful tool here is the Cartodialect website, where it is possible to visualise maps from the ALF. The map for “Abeille” is of course the first of them!

L’attribut alt de cette image est vide, son nom de fichier est Screen-Shot-2020-11-04-at-12.27.54-417x500.png.


Below is Valéry’s Bee in French and in an English translation by Nathaniel Rudavsky-Brody published by The Brooklyn Rail in In Translation recently:


À Francis de Miomandre.

Quelle, et si fine, et si mortelle,
Que soit ta pointe, blonde abeille,
Je n’ai, sur ma tendre corbeille,
Jeté qu’un songe de dentelle.

Pique du sein la gourde belle,
Sur qui l’Amour meurt ou sommeille,
Qu’un peu de moi-même vermeille,
Vienne à la chair ronde et rebelle !

J’ai grand besoin d’un prompt tourment :
Un mal vif et bien terminé
Vaut mieux qu’un supplice dormant !

Soit donc mon sens illuminé
Par cette infime alerte d’or
Sans qui l’Amour meurt ou s’endort !


However keen may be your sting,
However fatal, yellow bee,
Over my basket I have draped
The merest dream of floating lace.

So prick that swelling gourd, my breast
Where Love is sleeping, or has died.
A little of myself will rise
Scarlet to plump, rebellious flesh!

A sudden pang is what I need:
A pain that quickens and is gone
I’d rather than this slumbering grief.

Illuminate my senses with
Your microscopic gold alarm
Without which Love slumbers or dies!

(English translation copyright (c) Nathaniel Rudavsky-Brody, 2014)


Last but not least, Gilliéron’s introductory comments on the names of the bee:



(Jules Gilliéron, Généalogie des mots qui désignent l’abeille : d’après l’Atlas linguistique de la France, Paris: Champion, 1918, p.19)

Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search