According to Nesser [1], jihadis in general have the idea¹ “that they are to reoccupy European territories² held by Muslims at the height of the Muslim expansion”³ [1] (Figure 1 showing 7th and 8th century expansion; for later-Ottoman–expansion, Figure 2). By the “height of Muslim expansion” the author probably means the first mentioned Arabic expansion period, before the territorial losses in southern Europe became very substantial. As an example, he mentions Andalus ,⁴ the Muslim-occupied part of the Iberian peninsula , which was conquered in the 7th century but from the 9^th century on gradually reconquered by Christendom,⁵ and finally lost in 1492 (Figure 3).⁶ [2] This paper tries to examine to which extent and in which historical periods the now obsolete toponyms were in general use in the Arabic Muslim world, and whether IS uses the obsolete toponyms in the same meanings as their religious ancestors did.

Figure 1: Islamic expansion in the 7^th and 8^th centuries. (Source: christianworldviewpress.com). The ‘Islamic Caliphate’ or ‘Patriarchal Caliphate’ (633-661) is the one founded directly after Mohammed’s death; it was followed by the Umayyad Caliphate (661-750). Caliphs were supposed to be the successors of the prophet Muhammad.

Figure 2: Islamic expansion after the 8th century. (Source: http://nazaroo. blogspot.nl/2014/11/isis-is-now-islamic-state.html).

Figure 3: The seven phases of the Reconquista, starting from the North and ending in the South, and rendered by the colors yellow (before 914), orange (914-1080), etc. Source: https://nl.wikipedia.org/wiki/Reconquista_(Spanje).

Attention will be paid to obsolete toponyms which refer to geographic features that nowadays carry other names than those used by medieval Arabic-Muslim authors. ‘Arabic-Muslim’ authors are those whose mother language is Arabic, as well as those who wrote in Arabic, but had another mother language (e.g. Persian).⁷ [3] Data with regard to such sources were mainly collected from Konig [4] and other publications on relationships between the Muslim empire and western countries.

If obsolete names do not correspond with modern names, they have been ignored in this paper. The following names were left aside:

1. Khorasan: In IS language, it refers to a large territory stretching from eastern Iran into Afghanistan, not corresponding to any actual official name. In modern Arabic it refers to an Iranian province.

2. Alkinana: This name (rarely used) refers in IS l;anguage to northeast Africa (including Egypt) and does not correspond to any actual official name.

Obsolete toponyms have been collected, among others, from a map, published in 2014, and showing supposed future Muslim conquests in the world (Figure 4). Although the provenance of this map (henceforth indicated as Conquests Map) is unclear, and has been the object of criticism because of multiple mistakes and incongruities,⁸ it seems to reflect IS ideas.⁹ [5] Moreover, IS never dissociated itself from it, which seems a good reason to involve it in our material. In this study, the rationality of the (presumed) IS claims as they are expressed by the map will be disregarded. This applies, for example, to the suggestion, evoked by the map, that Ethiopia as a whole would have been Muslim in the past.

Figure 4: Conquests map. Source: HOSKEN 2016: x. Orobpa=‘Europe’, Qoqzaz=Caucasus, Sham=Levant, Habasha=Ethiopia, Yaman=Yemen.

As mentioned above, four obsolete toponyms were found (Table 1) ¹⁰.

Table 1: Obsolete toponyms used by Muslim state, with actual copgnates or synonyms.

Andalus

In Table 2, the name Andalus , as used by Muslim State, is compared with name use in Islamic-Arabic sources. Figure 5 shows the Muslimruled area in the Iberian peninsula .

Table 2: Territories referred to by means of the toponym Andalus by Islamic State, compared with territories referred to by historical Islamic-Arabic sources, from the 9^th until the 17^th century. Sources are in chronological order.

Figure 5: Al-Andalus (green territory) circa 1000 AD. Source: Wikipedia s. v. Al-Andalus. ‘Caliphate of Cordoba’ refers to the Umayyad Caliphate of Cordoba (929-1031).

As the Table 2 demonstrates, IS is using Andalus in three meanings. Two of these correspond with historical meanings.

1. ‘The entire Iberian peninsula ’. In this meaning, the name can be found on the Conquests Map, and this meaning may have been intended in 2015, when IS communicated: “We want, Allah willing, Paris before Rome and before Al-Andalus”.¹¹ [6-8]

2. Muslim-ruled southern Spain, once occupied by Muslim forces (Figure 5) and not coinciding with the present province of Andalucía (Figure 6).

Figure 6: The Spanish province Andalucía. Source: http://www.credoespana. com/spanje/andalucia/ .

In historical Arabic-Muslim sources, however, Andalus had four possible meanings. Table 3 ¹² also shows that some historical Arabic authors, as a matter of fact, used the toponym in more than one meaning, see for example Ibn Khaldūn (nr. 14).

Table 3: Territories referred to by means of the toponym Andalus by Arabic-Muslim scholars from the 9^th until the 17^th century. Sources are in chronological order.

Insofar as the historical word meanings could without any doubt be established in the data collected in this study, Andalus meant mostly ‘the Iberian peninsula as a whole’ (11 times), from the 9^th until the 15^th century. This name meaning has been attested from the very first start: two dinar coins dating from the period between 715 and 717 shows both a Latin inscription with the name Span (short for Spania ‘Hispania’, a terminological heritage from the Roman past,¹³ and the Arabic inscription Andalus .¹⁴ [6] And al-Marrākushī (nr. 7), for example, referred to “the city Silves [Shalab] on the peninsula of al-Andalus” (my italics). In the same period, sources also used a more restricted meaning: ‘the Muslim-ruled part of the peninsula ’ (7 times); the historian ibn Hayyān (Table 3, nr. 6) recorded military confrontations of the Christian Basques “with Andalus ”. Out of 15 authors, four use both meanings in their texts.

Initially, medieval Arabic-Muslim sources did not differentiate between the concepts ‘Spain’ and ‘Portugal’. In the heydays of Islam in the peninsula , parts of the territories now known as Spain as well as parts of present Portugal belonged to Andalus .

The earliest attestation of the name of Portugal in Arabic- Muslim sources dates from the 11^th century (al-Bakrī, 1014-1094): Al-Burtuqālish.¹⁵ [4] It does not refer to Portugal as an independent polity, but as one of the four provinces of Galicia. After three ages, Portugal was mentioned by Ibn-Khatib (1313-1375) and Ibn-Khaldūn (1332-1406) as a polity of its own. In that period, the name probably denoted just a fraction of present Portugal, namely the land between the rivers Douro and Minho, in the north of the present country.¹⁶ The name España for Spain, in the present meaning, emerged in 1474, when the kingdoms of Castile and Aragon merged.¹⁷ Since the 16th century, Spain has been indicated in Arabic by Isbaniya or Ishbanya¹⁸ [7] and discerned from the historical name Andalus . Among some descendants of expelled muslim emigrants, however, Andalus remained in use with reference to their former home country. As late as in 1785, inhabitants of Testour, in Tunesia, made inquiries of the Moroccan ambassador who visited Tunesia, asking after ‘their land, al-Andalus, which I (the ambassador) had observed when I traveled there’. Their ancestors had been expelled in 1609.¹⁹ [7]

By applying the meaning ‘Spain as a whole’, Muslim State is, from a historical point of view, mistaken. Moreover, the irony is that IS, by using the meaning ‘Spain as a whole’, is in a sense more ‘modest’ in its territorial claims than early medieval history might justify: why not claim ‘muslim Portugal’, too? Projecting the modern concept ‘Spain’ on the medieval past is, however, not an IS prerogative. For example, Wikipedia (Dutch version) s.v. Reconquista (Spain) describes Andalus as ‘the Arabic Spain’, while spencer [8] equates Andalus with “Muslim Spain”.²⁰ [8] And even for an Islam expert like Bearman [9] this use matches indeed with the one of those medieval Arab writers for whom Andalus referred to ‘the Muslim part of Spain “whatever its territorial extent”’ (my italics).²¹ [8] The simplication is well understandable, though, considering that Portugal’s name as a regional entity and polity arose in non-Arabic Muslim sources as late as the 11^th century; specific terms for the territories now called Spain and Portugal, were lacking, as mentioned above.

Although the conquest map does in this respect more justice to history, it is erroneous, too: it suggests wrongly that the whole peninsula was under muslim rule, and neglects the Christian territories in the north.

Gaul

After terrorists had assaulted Paris in 2015, Muslim State (IS) communicated that the perpetrators ‘had come from all over the world to Gaul’. I could not find the Arabic text (or a transcription thereof) of the original IS message in which the Arabic equivalent for Gaul might have occurred. So, the terrorists may have employed – apart from Gaul - the present Arabic name El-Ghāl (denoting Gaul as the Roman territory),²² or – less probably - the obsolete variants Ghālīsh or Ghālliya. Obviously, the terrorists referred to France. In present Arabic, France is called Faransa. To my knowledge, it was the first and only time that France was indicated this way by IS.

Tables 4 and 5 show that in medieval days Ghālīsh was the current Arabic equivalent of Latin Gallia , Ghālliya having been used only once, as a variant of Ghālliya in one and the same source. These medieval terms did, however, not refer to France but to other geographical entities. As a matter of fact, the usual medieval-Arabic name for the Frankish empire (Figure 7) and, later on, for France, was Faransa, Faransīs, Ifrans,²³ [4] Ifransa²⁴ [4,7], Ifranjiyya²⁵ [4,10] or Ifrānsiya.²⁶ [4] Since the 16^th century, Ifransa was the common Arabic name for France.²⁷ [7]

Table 4: Territories referred to by Islamic State by means of the toponym Gaul, compared with territories referred to by cognates in Islamic-Arabic sources, from ca. 900 – ca. 1300. Sources are in chronological order.

Table 5: Territories referred to by means of the toponyms Ghālliya and Ghālīsh by medieval Arabic-Islamic sources. Sources are in chronological order.

Figure 7: Frankish empire. Source: https://nl.wikipedia.org/wiki/Frankische_Rijk

The Gaul terminology in Arabic sources originates from the beginning of the 9^th century, when the earliest Arabic translations Ptolemy geographical oeuvre were made and Gallia was transcribed in Arabic script.²⁸ [4] It goes without saying that Ptolemy’s Gallia referred to the Roman province of that name (Figure 8), and so did the Historiae adversus paganos by historian Paulus Orosius (375 – after 418). The latter book was translated ca. 900 into Arabic, which lead to the translations Ghālliya and Ghālish.²⁹ [4]

Figure 8: Gaul as a part of the Roman empire. Source: https://en.wikipedia. org/wiki/Gaul

Why did the jihadis who attacked Paris choose the toponym Gaul with regard to France? Obviously, they did not realize that their religious ancestors did not use this name with regard to France. The following scenario seems probable. (1) A need was felt for an obsolete name for France, expressing nostalgia for the Muslim heydays. (2) Since the late Middle Ages, in many countries the name Gaul adopted a narrowed meaning, ‘France’; Gaul came into fashion among humanists in Western Europe and became especially popular in poetical and more or less anti-French contexts.³⁰ [11] For example, the German poet Friedrich Gottlieb klopstock [12] wrote: “Gallia Sklavin; Gallia frey; sie erniedrigt zur Wilden/Dann sich, schaffend sogar Marat, den Scheusal zum Gott:/ Bleibt, bleibt Wilde.” [Gaul slave, Gaul free; she demeans herself to a savage/Then, while making Marat, the monster, even a god/ remains savage]³¹ [12] Likewise, the English poet Robert Burns [13] chose the Gallia name in 1795, at a critical moment in English-French relationships: “Does haughty Gaul invasion threat! Then let the louns bewaure, Sir!” [If haughty Gaul threatens us with an invasion, etc.].³² [13] In the Netherlands, a poet reported: ‘The thunder of Holland’s and Zeeland’s guns stroke terror into [… ] Castilia, filled Britain with awe, and Gaul with esteem’.³³ [14] Anyway, the jihadi toponym Gaul (or any modern or obsolete Arabic equivalent) should be considered as a would-be obsolete variant: it does not mirror the medieval past. In this respect, ‘Frankish empire’³⁴ [15] or ‘Francia’³⁵ [4] would have been a better (and likewise obsolete) name for France.

Although the humanist Gaul naming habit did not reach the common language,³⁶ [11] the juxtaposition Gallia = France obviously has survived. Firstly, jihadis may have been inspired by the Astérix comics (Figure 9) ³⁷ Secondly, in 1965, René Goscinny and Albert Uderzo published the comic Le tour de Gaule d’Astérix [The Tour de Gaul], alluding to the Tour de France. And in this tradition also fits a quote like this: “Il y a environ deux mille ans, la France c’était la Gaule” [Some two thousand years ago, France was Gaul].³⁸ [16] The humanist identification of Gaul with the French state was probably due to the fact that France, projected on the historical map, was the main part of Roman Gallia .

Figure 9: Map of Gaul in Astérix le gaulois. Source: http://asterix.wikia.com/wiki/ File:Map_of_Gaul.png

Habasha

Table 6 shows the meanings of Habasha as used by Islamic State in comparison with meanings in historic sources.

Table 6: Territories referred to by Islamic State by means of the toponym Habasha, compared with historical Muslim-Arabic cognates, from the 7^th until the 16^th century. Sources are in chronological order.

The meaning used by IS corresponds with one of the three historical meanings. It may seem amazing that Muslim State uses an obsolete, ‘nostalgic’ name for Ethiopia. Considering that Ethiopia has never been considered by anyone as a Muslim country – rather a Christian one with an Muslim minority - it is remarkable that the Conquest map implies that the country has to be reconquered by Muslims. At present, Ethiopia is predominantly Christian. Islam is the second religion (34 percent adherents). Ethiopia adopted Christianity in the 4th century. In 615, the first muslims immigrated.³⁹ The first large scale penetration of Islam dates from the 11^th century. The 16^th century, however, was a critical period: under the rule of Ahmad ibn Ibrahim al-Ghazi (1506- 1543), three quarters of Ethiopia came under Muslim control, some of his fellow combatants being Ottoman Turks.⁴⁰ [17] In 1527, al-Ghazi made plans to conquer all Ethiopia, but the threat soon disappeared. Possibly it is especially this historical period jihadis are alluding to if they use the Arabic name Habasha .

Habasha (habaša⁴¹) is a toponym as well as an inhabitant name. As a toponym, it is the obsolete Arab equivalent of Abyssinia, which, as an official name, was the precursor of Ethiopia. Well into the 20th century, Abyssinia and its cognate variants (like Arab Habasha ) were in use along with Ethiopia and its cognates. In 1925, Abessynia was formally replaced with Ethiopia, at least by the authorities in the country itself.⁴² Outside the country, the name change proceeded somewhat slowly: as late as 1945 the country entered the United Nations as Abyssinia, not Ethiopia. May that be as it is, in the Arab world the Ethiopia name has officially rooted: Atyūbiyā.⁴³ (Table 7). ^44,45

Table 7: Territories referred to by means of the toponyms Habasha, Ḥābeš, Habashat or Habash by Arabic-Islamic sources from the 7^th until the 16^th century. Sources are in chronological order.

Rumiyah

Rūmiyāh is the title of an English language periodical published by IS, being the successor of their periodical Dabiq (Figure 10).The title page says “Rumiyah”, followed by: “O muwahhidin [= monotheists], for by Allah, we will not rest from our jihad except beneath the olive trees of Rumiyah – Abu Hamzah al-Muhajir.” ⁴⁶ [18-20] Obviously, Rumiyah refers to the city of Rome.⁴⁷ [8] This points to some knowledge on Rumiyya as the most frequent name version in medieval Arabic with regard to the city of Rome (Table 6) [21-33]. I could not find this variant in other IS utterances, such as videos. In present-day Arabic, the city is called Rūmā.⁴⁸

Figure 10: Front page of Rumiyah. Source: https://twitter.com/_davidthomson/ status/773052281893814272

The IS name corresponds with Rūmiyyā (Romiya) as an urbanonym in historical Muslim sources (Table 8). Apart from this, the historical sources show three meanings not used by IS (Table 9 ^49,50,51).

Table 8: Territories referred to by Islamic State by means of the toponym Rumiyah (‘Rome’), compared with historical Islamic-Arabic cognates, from the 9th until the 13th century.

Table 9: Territories referred to by means of the toponym Rūmiyāh and its cognates by Arabic-Islamic sources from the 7^th until the 16^th century. Sources are in chronological order.

1. Comparison of IS toponym use with toponym use in the distant past does not present a univocal picture. Firstly, a discrepancy occurs in the case of Spain: IS tries to relate a present polity to a historical one, Andalus , probably without realizing that in the early Middle Ages Spain was not a polity in itself, coinciding with modern Spain, let alone that medieval Arabs had a special name for the area corresponding with modern Spain. Moreover, Andalus is not exclusively the alleged precursor of Spain. It is Portugal’s, too.

With regard to Gaul, the discrepancy between medieval and IS terminology is bigger – it could hardly be greater. The jihadi name use does not correspond at all with medieval Arabic name use. The jihadi placing of Gaul on the same footing as France does not stem from Arabic linguistic and cultural history, but from the European cultural past – possibly courtesy of the Astérix comics. It is ironic that the jihadi name use reflects – unknowingly – the intellectual legacy of Humanism.

In the case of Habasha , IS terminology is quite consistent with its medieval precursors Hābeš, Habasha and Habashat, all of which were used to denote the country. The Habasha name suggests that, apart from the Arabic Muslim empire, the Ottoman period, too is deemed essential. This would refute Nesser’s [1] opinion that ISIS fosters exclusively the early (Arabic) heydays of Muslim expansion.

The obsolete name Rūmiyyā, as used in IS language, too, corresponds with the name use of medieval precursors. Just like their religious ancestors, they use Rūmiyyā for the city of Rome.

IS toponymic nomenclature is not a completely isolated phenomenon. At least one name, Andalus , is not exclusively employed by IS and other jihadi groups, like Al-Qaida.

Foot notes

¹ I am indebted to prof. Dr. Nico Landman and Mark Demmink for helping me to interpret some Arabic terms.

² IS’ periodical Dabiq employs only the now usual names.

³ Nesser 2015: 32.

⁴ The Arabic definite article al – or el-, which in Arabic toponyms is a more ot less arbitrarily used element, has been omitted in this article.

⁵ The reconquista (‘reconquest of Muslim-occupied territories’) started as early as (probaby) 722 (battle of Covadonga) (see Wikipedia s.v. Reconquista).

⁶ Non-jihadi muslims see the height of Islam, however, in a broader – also cultural – context, and extend the golden age of Islam to the 13th century (Wikipedia s.v. Islamic Golden Age) or even to at least the early part of the Ottoman period (between the 15th century, when Constantinople was conquered by the Turks (see Torres Soriano 2009: 939), and the end of the 17th century (failure of the Vienna siege)).

⁷ I follow here Micheau 2000: 53.

⁸ See, for example, Media Matters for America 3 July 2014 (https://www. mediamatters.org/blog/2014/07/03/why-is-the-media-taking-these-isis-worlddomina/ 199984 ) and Info Wars 27 June 2014 (https://www.infowars.com/warpropaganda- the-isis-five-year-plan/ ).

⁹ See Hosken 2015: 260-261.

¹⁰ See http://en.bab.la/dictionary/english-arabic.

¹¹ Spencer 2015: 253.

¹² In the Arabic text, however, Andalus is equated with Spain.

¹³ Wikipedia s.v. Hispania. Sometimes Hispania denoted exclusively the Christian part of the peninsula , e.g., when Alfonso VI (1040-1109) took the title Imperator totius Hispaniae.

¹⁴ Bossong 2002: 149.

¹⁵ König 2015: 317. Ibn al Khatīb (1313-1375) used the Arab name Burtuqāl.

¹⁶ In non-Islamic sources, the name Portugal has been attested much earlier, in the 9^th century. It derives from Latin Portus Cale, for a city now called Vila Nova de Gaia. In 868 a county was named Portucale ‘land between Douro and Minho’. In 1109, the latter name denoted the territory of this county plus Coimbra.

¹⁷ See Wikipedia s.v. Kingdom of Castile. Some authors, however, continued using ‘Spain’ until the 16th century and even in the 19^th century with regard to the whole peninsula . See egli 1872: 538-9.

¹⁸ Matar 2009: 31.

¹⁹ Matar 2009: 15. The author does not mention any details concerning the background of the ambassador’s visit.

²⁰ Spencer 2015: 211, 213, 253, 342.

²¹ Bearman et al. 2005 et al., s.v. Al-Andalus. Elsewhere in this same article, however, we are correctly told that al-Andalus “in the Islamic world up to the end of the Middle Ages, denoted the Iberian peninsula , that is, modern Spain and Portugal”.

²² The Reverso Dictionnaire (http://dictionnaire.reverso.net/francais-arabe/gaule ) translates French Gaule in لاغلا دالب . I thank Mark Demmink for his transcription. He kindly comments that the word may indicate the country Gaul as well as one of its inhabitants. See also https://www.google.nl/search?q=rabitat+vertaling&oq=r A&aqs=chrome.1.69i59l3j69i60j69i57j69i60.4467j0j8&sourceid=chrome&ie=UTF- 8#q=galli%C3%AB+vertaling+arabisch

²³ König 2015: 228.

²⁴ König 2015: 136. This term was the most usual one since the 16th century (see Matar 2009: 31).

²⁵ König 2015: 213. Also Lewis 1987: 146 (“Ifrandscha”).

²⁶ König 2015: 227.

²⁷ Matar 2009: 31. https://translate.google. om/?text=undefined&hl=en&langpair= auto%7Ctr&tbb=1&ie=undefined#auto/tr/gaul

²⁸ König 2015: 199.

²⁹ König 2015: 43; 197.

³⁰ Brühl 1990:75.

³¹ Klopstock, t, p. 493.

³² Burns 1795: 436-7.

³³ Berkhey 1769-1811: III, 642.

³⁴ Translation in Ragep & Ragep 2009: 427.

³⁵ Translation in König 2015: 213.

³⁶ Brühl 1990:75.

³⁷ Translated a.o. into Arabic.

³⁸ Beauvais 1970: 8.

³⁹ https://en.wikipedia.org/wiki/Abyssinian_people#Antiquity

⁴⁰ J. Spencer Trimingham 1952: 85.

⁴¹ https://en.wikipedia.org/wiki/Ethiopia

⁴² http://www.opride.com/2009/11/02/abyssinia-to-ethiopia-from-obfuscation-toconfusion/ This source refers to a source I could not consult: Ronald K. Starrett (ed.), Documents on Ethiopian Politics, vol. 2 (1976): 1920-1929, The Consolidation of Power of Haile Selassi.

⁴³ http://en.bab.la/dictionary/english-arabic/ethiopia

⁴⁴ One example of a difference between historical and present Habasha can be found in al-Battāni (ca. 858 – ca. 929), who situated the now Egyptian city of Aswan in Ethiopia (Uswān al-Habasha); cf. Ragep & Ragep 1996: 432.

⁴⁵ The author situated the Canarian Islands opposite the Habasha continent (cf. Ragep & Ragep 2009: 432f.)t

⁴⁶ Abu Hamza al-Muhajir was al-Qaeda in Iraq leader († 2010).

⁴⁷ This name variant seems also in use in orthodox – but not nessessarily jihadi – circles. In 2002, Sheikh Yusuf al-Qaradavi was reported to use Romiyya (referring to the city of Rome); see Spencer 2015: 214. In present-day Arabic, the city is called Rūmā; see http://nl.bab.la/woordenboek/engels-arabisch/rome.

⁴⁸ see http://nl.bab.la/woordenboek/engels-arabisch/rome.

⁴⁹ According to Casari [19] some scholars maintain that Ibn Khordadbeh’s description aimed as a matter of fact at Constantinople.

⁵⁰ The spelling Romiyya is used, among others, by Spencer [8]

⁵¹ http://shodhganga.inflibnet.ac.in/bitstream/10603/55329/7/07_chapter%201.pdf, p. 11.