Catalunya, català Miguel Carrasquer Vidal, June 2013 The etymology of Catalunya and català remains an unsolved mystery. There are obviously a large number of proposals, but none of them are totally convincing. A fair summary of the current state of our knowledge was given by Joan Corominas in his Onomasticon Cataloniae, under the lemma català, Catalunya (vol. III BI-C, 1995, pp. 335-339): ¿D’o à e àelà o sàcatalà i Catalunya? Ben mirat no ens consta, però valdrà la pena de p e isa àlesàideesàdie tàelà ueàa uiàseàsapàd’a uestaà üesti .àLa realitat més important és que aquests mots apareixen en un moment històric més aviat tardà. La primera menció de Cathalonia és de 1114: Catalania, Catalanicus i Catalaniensis s’use àe àelàLiber Maiolichinus, poe aàes itàe à ai àllatíà oltàpo sàa sà sàta d.àÉsài po ta tàd’o se a à ueàtotesà a uestesà e io sàdeàp i e aàho aàs àe àes itsàd’italia sàiàesà ela io e àa àl’e pedi i à aliada que emprengueren el Comte de Barcelona i la República de Pisa contra els pirates o osàdeàlesàBalea s.àDesàdeà itja sàdelà“.àXIIàl’úsàd’a uestesàpa aulesàesàge e alitzaàe àelsà païsosà istia s,à e t eà ueà oàa i à aiàaàeste d e’sàalsà o osà ueàa o e a e à f a s à els habitants de Catalunya). Vetàa uíàl’ú i àfetà ueàsa e àa àtotaàsegu etat.1 Let’sà e ie àso eàofàtheàtheo iesàsuggestedàsoàfa ,àskippi gàtheà ildl ài plausi leào es. 1. Otger Cataló The first attestation of the word Cathalonia, as we saw, is from 1114. We know that three centuries later, knowledge about the origins of the word had already been lost, as it is then that the first of the hypotheses to explain the name appears in the form of the legend of Otger Cataló, narrated by Pere Tomic in his Histories e conquestes dels Reys de Arago e Comtes de Barcelona (1438): Deueu sa e à ueàe àl’a àdeàlaài a a ioàdeàJesuàCh istà ue’sà o tauaàDCCXXXIIIàe àlaà prouincia de Guiana hi ha un castell qui es apellat Cathalo per totes les gents e era apellat lo dit princep Otger Cathalo com en aquella prouincia hagues un princep qui regia la dita terra per Pipi rey de França e lo princep e regidor fos apellat per son propi nom Otger Golant mes per tal com lo dit princep feya la sua habitacio en lo castell Cathalo per totes les gents eran apellats los Cathalons2. The myth of Otger Cataló appears to combine elements of the Chanson de Rolland with the legend of Otger Danès (Ogier the Dane, Holger Danske). In any case, as an explanation for the words català and Catalunya, it falls short in several respects: 1 Where do the names català and Catalunya co eàf o ?àállàthi gsà o side ed,à eàdo ’tàk o àfo àsu e,à utàità would be worth it to make our thoughts more precise by saying what we know today about this question. The most important reality is that these words appear in a rather late historic moment. The first mention of Cathalonia is in 1114: Catalania, Catalanicus and Catalaniensis are used in the Liber Maiolichinus, a poem written in low Latin a few years later. It is important to observe that all these early mentions are in Italian writings, and are related to the allied expedition which the Duke of Barcelona and the Republic of Pisa undertook against the Moorish pirates of the Balearic Islands. From mid-12th century, the use of these words spreads generally through the Christian lands, while it never reached the Moors (who called the inhabitants of Catalo iaà F a ks .à“eeàhe eàtheào l àfa tsàthatà eàk o àfo àsu e. 2 You should know that in the year 733 of the incarnation of Christ, in the province of Guyenne there was a castle called Cathaló by all the people, and the said prince was called Otger Cathaló, as there was in that province a prince who ruled for Pepin King of France, and the prince and ruler was called by his own name Otger Golant, but as the said prince lived in Cathaló castle, they were called by all people the Cathalons.   2. The origin of the name of the castle Cathaló itself is left unexplained. Inhabitants of a castle Cataló, or followers of a person called Cataló, would most likely have been called catalonins or catalonesos, and not catalons (but since the actual name is in fact catalans,àitàdoes ’tà eall à atter). The Catalauni One possible explanation for the castle name Cathaló is to connect it with the name of the Catalauni, a Celtic tribe from the neighbourhood of modern Châlons-sur-Marne, famous amongst others for the battle of the Catalaunian fields (451), where Romans and Visigoths defeated the Huns. The theory that Catalonia comes from Catalaunia was first proposed by Jeroni Pau (1491) and Francesc Calça (1588), and was more recently defended by Bonfante (1944). There are two problems with this:  There is no historical evidence that the Catalaunian Celts ever had any connection with Catalonia.  The name Catalauni (French Catalaunōs > *čaðalo s > Châlons) would have given Catalan cadaló and Cadalonya, and not català, Catalunya. 3. Gotholandia etc. A name that was assuredly given to the territories of the Spanish March that were later to become Catalonia is Gothia (since these were old Visigothic territories, taken from the Moors by the Franks). This led some investigators (Torres Amat, Covarrubias) to suggest that Catalunya, català might be derived from something like Gotholandia. There are obvious problems, both with respect to the vocalism and the consonantism, with this theory. In order to remedy the linguistic problems, it has been suggested to derive instead from Goth-Alania or Chatt-Alania, by bringing in the Alans, which founded a kingdom in Hispania, but never passed through Catalonia, or the Chatti (i.e. Hatti), a Germanic tribe from Hessen, which is not known to have ever been linked with either Catalonia or the Alans. In any case, these would most likely have given *godalà, *Godolanya and *atalà,*Atalanya instead of català, Catalunya. 4. Castellani An old theory, defended by Ocampo, Zurita, Andreu Bosc, Balari and Rubio García, connects català with castellanu through a reduced form *castl(l)án-. The regular development can be seen in cat. Castella and castellà, which obviously refer to another part of Hispania. In Catalan toponymy there certainly are cases derived from a reduced form castl(l)an-, but the possible developments are either catllà /kaʎʎá/ or carlà. A development *castlan- > català is quite impossible. Rubio García argues that the development originally occurred in the South of France, to account for the loss of s before the consonant cluster. It is true that the modern Occitan dialects have lost s in this position at the northern fringe of the dialectal area, but that surely happened in modern times. Around 1100, when the word Cathalonia first appears, the s had not even been lost in the la gue d’oïl. 5. Montcada The toponymist Paul Aebischer proposed to derive català from montecatananus > montecatalanus, after the place name Montcada, near Barcelona, with elision of the prefix mont-. As the name Montcada itself proves, that could only have given *(mont)cadalà (with plausible dissimilation d..d > d..l, instead of the unlikely t..t > t..l). From an historical point of view, it remains completely unclear what the significance of Montcada castle may have been in the naming of Catalonia. 6. The Lacetani In 1917, the German phonetician Ernst Schopf, in his dissertation Die konsonantischen Fernwirkungen made a side remark about the possibility of a metathesis lacetani > catelani. The proposal was endorsed by Grammont and Ronjat. The Iberian tribe of the Lacetani is known to have inhabited a good part of modern Eastern Catalonia in pre-roman times, so the theory is certainly attractive from an historical point of view. Linguistically however, it is unconvincing. If the metathesis is old enough to have happened at a time when c before e was still k, then the metathesized form is old enough to have undergone the weakening of t to d between vowels, and the result would have been *cadalà, *Cadalanya. Joan Corominas was obviously aware of this: Aquesta etimologia la va sugerir per primera vegada, i en termes lacònics, el fonetista ale a à“ hopfàl’a à 9 9à Die konsonantischen Fernwirkungen,àp gi aà 96 ,àiàl’a epta e à sense discussió ni raons Grammont, Ronjat i algú més. Però ningú fins ara no havia provat de fer-laàa epta leàdesàdelàpu tàdeà istaàhist i ,àiàelsàseusàpa tida isàpotse às’e ui o a e àe à presentar-la com si correspongués a una evolució popular i hereditària: aleshores el canvi de Catelanus en català oàesta íaàd’a o dàa àelàt a ta e tàfo ti àdeàlaà-t- entre vocals3. However, Corominas, in his Onomasticon Cataloniae and several other publications, proposes another variant of this theory as the most likely explanation for the words català and Catalunya: Vetàa uíà o àitalia sà ultesà e o da e à ueàa uellaàpa tàdeàl’Hisp iaàe aàha itadaàpelsà Lacetani, i començarien a usar de nou aquest nom clàssic, tal com es va fer amb Aquitania, Belgica,Lucania i altres noms clàssics, ressuscitats amb èxit més o menys constant i pe a e t.àElà o àpogu à i ula àp i e a e tàe àelsàes itsàd’algu s:àdesp s,àt o a t-lo útilàiài dispe sa le,àd’alt esàp o a ie àd’i ita -los, i dirien Catelani o Catalani, per un lapsus (de llengua o de ploma) gens inusitat en tals resurreccions mig-sàvies. [] En el cas de catelani a uestà lapsus àt o a iaàa epta i àge e al,àpe àtalà o àlaàfo aàalte adaàe aà sàf ilàpe à aquells que ja coneixien els castellani i els catalauni4. Finally, to explain the discrepancy between the vowels of català and Catalunya, Corominas adds: E àfi,à ’adhe i iaàaàl’e pli a i ,àjaàdo adaàpe àalt es,àdelà a iàdeàlaàte i a i àdeàCatalania en Catalonia com a degut a la influencia del altres nombrosos noms de país que terminen amb -onia: Vasconia, Gasconia, Slavonia, Pannonia, Franconia, Brittonia, Livonia, etc. [] El model immediat de Catalonia [] devia ésser especialment Vasconia (> Gascuenha), ja que és 3 This etymology was suggested for the first time, in laconic terms, by the German phonetician Schopf in the year 1919 (Die konsonantischen Fernwirkungen, page 196), and was accepted without discussion or reasons by Grammont, Ronjat and some others. But no-one until now had tried to make it acceptable from a historic point of view, and its supporters were perhaps mistaken in presenting it as if corresponded to a popular and inherited development: in that case the change from Catelanus to català would not be in agreement with the phonetic treatment of -t- between vowels. 4 See here how learned Italians remembered that this part of Hispania was once inhabited by the Lacetani, and they would have started to use again this classical name, as was done with Aquitania, Belgica, Lucania and other classical names, resuscitated with more or less permanent and lasting success. The name might have circulated at first in the writings of some: later, finding it useful and indispensable, others would try to imitate them, and would say Catelani or Catalani, by a lapsus (of the tongue or the pen) not unusual in this kind of halflea edà esu e tio s.à[]àI àtheà aseàofàtheàCatela i,àthisà lapsus à ouldàfi dàge e alàa epta e,àasàtheàalte edà form would have looked easier for those already familiar with the Castellani and the Catalauni. l’ú i à ue,à o àCatalunya, mostra breu: hi ha un paral·lelisme rigurós entre els occitans Cataluonha i Gascuonha, cast. ant. Catalueña i Gascueña, cat. Catalunya i Gascunya, etc5. Indeed, formulated as a metathesis of the consonants in the written word, the development lacetanu- > catelan- > català is perfectly possible. A similar, bookish, explanation would work even better (without the need for metathesis) for Catalauni,àalthoughàthatàdoes ’tàtakeàa a àtheàla kàofà historical connections between Châlons and Catalunya. 7. The Kattelanoí Although all the oldest sources (Plutarchus, Plinius, Livius, Frontinus, Dio Cassius) mention the Iberian tribe as Lacetani, in Ptolemy the approximate place of the Lacetani is taken by the tribe of the Καστελανοἰ Καστελλανοἰ in a few manuscripts, no doubt under the influence of lat. castellani). The earlier Corominas had interpreted this form as a possible alternative origin of the metathesis, not in 10th-century Italy, but in 2nd-century Alexandria, under the assumption thatàΚαστελανοἰ actually stood fo àΚατελανοἰ (with the initial curl of the tau mistaken for a sigma). However, in later years, citing a personal communication from Ni olauàd’Ol e àthatàtheào igi alàfo ài àPtolemy might have been Καττελανοἰ, Corominas became more and more convinced, as was his overall trend, of a preroman origin for català and Catalunya: D’alt aà a daàta àesàpotàsospita à o àse laàpe sa àE àNi olau à ueàe àΚαστελανοἰ el malentès causat pel caragol paleogràfic []àestigu sàpe àΚαττελανοἰ amb dues taus i es tractés d’u aà a ia tàp e-romana amb duplicació intensiva. Llavors podría haver-hi hagut transmissió popula à atala aàju tàa àlaàt a s issi àe uditaàpe à iaàd’It lia6. However, it seems highly unlikely that the name of the tribe in pre-roman times was *kattelan-:  The vast majority of the attested tribal names in the general neighbourhood show a suffix -et(an)- consistent with a form Lacetani: Iacetani, Laietani, Edetani, Ceretani, Turdetani, and further Ilergetae, Indigetes, etc.  There is no evidence that Iberian allowed geminated stops. The native Iberian alphabetsyllabary does not distinguish between p and b, t and d or k and g7, let alone geminated variants thereof, and texts in the Greek script, such as the lead plaque from La Serreta (plomo de Alcoy), show no evidence of geminated stops. 8. Talunya The Catalan arabist Joan Vernet identified a possible first attestation of the word Catalunya in the Tarsī al-axbā , by the Muslim historian and geographer Al-ˁUd īà à– ca. 1085). There, a place called Talunya or Taluniya (‫ تلونية‬or ‫ )?طلونية‬is mentioned, in the neighbourhood of modern Montsó (Monzón). With the addition of a prefix ca-, short for ‫ قلعة‬alˁah astle (or why not cat. ca, short for casa house ? , the name Catalunya would have been born. Corominas objects that it is highly unlikely that an obscure location at the very edge of the Catalan lands would have been chosen to name the whole country. On the other hand, Montsó at the time 5 Finally, I would second the explanation, already given by others, of the change in ending of Catalania to Catalonia as due to the influence of the many other country names ending in -onia: Vasconia, Gasconia, Slavonia, Pannonia, Franconia, Brittonia, Livonia, etc. [] The immediate model for Catalonia [] must have been especially Vasconia (> Gascuenha), as it is the only one that, like Catalunya,àhasàaàsho tà :àthe eàisàaàst i tà parallelism between Occitan Cataluonha and Gascuonha, Old Cast. Catalueña and Gascueña, Cat. Catalunya and Gascunya, etc. 6 O àtheàothe àha d,ào eà ouldàsuspe tà asàNi olauàsee sàtoàthi k àthatài àΚαστελανοἰ the misunderstanding caused by the palaeographic u là[]àsta dsàfo àΚαττελανοἰ with two taus and that we are dealing with a preRoman variant with intensive gemination. In that case there might also have been Catalan popular transmission, alongside the erudite transmission via Italy. 7 Although the distinction was made in the dual variant of the North-Eastern script, used precisely in presentday Catalonia. would have been near the border between the Catalan-speaking, Castilian/Aragonese-speaking, and Muslim areas, just within the Catalan area, which would make it a suitable appellative if the name arose precisely in such a border area. 9. Bilād al-qalˁāt Another theory which I have encountered on the Internet, but the source of which I have been unable to trace, was apparently formulated by the French-Catalan linguist Enric (Henri) Guiter (when? where?). The suggestion is made on similar historical grounds as the theory about castellani above, except that here the origin is traced to Arabic (bilād al-) alˁāt la dàofà astles .àTheàadje ti eà *calatà would then, by metathesis, have become català. Curious detail in the Internet account is that p ofesso àGuite ,àappa e tl ,à as part of his study [] asked 1000 students from the Montpellier U i e sit to epeat the o d Calataiud . “u p isi gl , a ou d 70% of the stude ts said Catalaiud i stead of Calataiud . The problem with this, however, is that the Arabic plural of ‫ قلعة‬alˁah astle àisà otà* alˁāt but either one of the broken plurals ‫ قاع‬ilāˁ or ‫ قلوع‬ulūˁ. 10. Conclusion and new proposal In conclusion, the ideal explanation for the words català and Catalunya would offer an historically plausible explanation of the name, accounting for all of the initial c- (k-), medial -t-, and the alternation in the vocalism of català versus Catalunya. The theories reviewed above, in tabular form: theory history name 1. Otger Cataló 2. Catalauni + 3a. Gotholandia + + 3b. Goth-Alania + 3c. Chatt-Alania + 4. castellani + + 5. Montcada + 6a. Lacetani (Schopf) + + 6b. Lacetani (Corominas) + + 7. Kattelanoi + 8. Talunya + 9. qalˁāt + + k+ + (+) + + + + + + + -t+ + + + + (+) -a+ + + + + + + + + -o/u+ + (+) (+) + - Goths, castles and Lacetani (but not *Cattelani) offer plausible historical scenarios. The theories of Otger Cataló and castle Talunya fail to explain the name itself. All but the Gothic (and, in my opinion, the Chattic) theories explain initial k-, but only Otger Cataló, the Chatti, the bookish Catelani, Ptole ’sàKattelanoi and Talunya explain medial -t- adequately. None of the theories explain both -ain català and -o/u- in Catalunya simultaneously. The usual explanation (see Corominas above) that Catalonia is modelled after other country names in -onia (especially Vasconia) is factually accurate, but the motivation remains unclear in view of the no less common country names in -ania (cat. Bretanya, Alemanya,àa d,àlet’sàfa eàit,àEspanya). Vowel alternations (against a fixed consonantal pattern) are a feature of the Semitic languages, which led me to look into the possibilities of an Arabic origin for the alternation between català and Catalunya. I àWeh ’sàdi tio a à eàfi dàaà u instance: e àofà ootsà o espo di gàtoàa àapp o i ateàpatte àK-T-L, for xtlxatala (xatl, atalā ) to dupe, gull, cheat, double-cross, deceive (s.o.) III to deceive, cheat, dupe (s.o.); to behave hypocritically VIII = I xatl and u ātala deception, trickery, double-dealing, duplicity, duping, gulling u ātil deceitful, crafty, wily, foxy a ila ( a al) to talk nonsense IV do. V to strut, walk with a pompous gait; to walk with a proud, swinging gait a al idle talk, prattle a il garrulous, chattering, given to silly talk ; stupid, foolish qtlqatl killing; manslaughter, homicide; murder, assassination [...] qitl pl. aqtāl enemy, foe, adversary, opponent qatīl pl. qatlā killed; killed in battle, fallen ; one killed in battle, casualty […] qattāl murderous, deadly, lethal [...] a ala ( a l) to cut off (s.th.) II = I a īla towel; floor rag ktlkatala (katl) and II to agglomerate, conglomerate, gather into a compact mass, press into a lump (s.th.) V to be heaped up, piled up, agglomerated, clustered, clotted, massed, pressed into a compact mass or lump; to agglomerate, cluster, clot, pile up, gather in a mass; to unite in a bloc or group kutla pl. kutal lump, chunk, clod, clot; bulk, mass; cube; block; bloc; beam, joist, transom, lintel, girder [...] An interesting word is attāl (‫)قتالو‬, pl. attālū (‫) قتالون‬, glossedà àWeh àasà u de ous,àdeadl ,à lethal à adj. .àTheà o dàisàofàtheàpatte àfaˁˁ āl, which makes intensive formations, and it also has the substantive meanings kille à o eà hoàkillsàaàlot àa dà attle aàlotàofàkilli g . While intensives from some of the other roots given above are possible in theory, and would make entertaining ( a āl hatte e s , cat. e ai es ) or implausible ( attāl heate s ) sources for the word Catalunya, the realities of life in the borderlands between Islam and Christianity in Hispania around the turn of the first millennium make the root qtl toàkill à àfa àtheà ostàlikel àsou e,à among the ones given above, for the name of the country Catalunya and the ethnonym/adjective català. I am reminded especially of the parallel of the almogàvers, the famous, as well as infamous, CatalanAragonese troops who played an important role in the Christian-Muslim border wars in Hispania, and later in Sicily and in the Byzantine-Turkish Orient. The word almogàvers comes from Arabic al- aghā ī ‎ raiders (from sg. mighwā ). According to Wikipedia: The first historical reference appears in the Arabic chronicle a bā ulūk al-Andalus,à Histo à of theàki gsàofàtheàá dalus ,à itte à et ee à àCEàa dà9 àCEà àá adài à u a adà al-Rāzī,àk o àa o gàtheàá a sà ithàtheà i k a eàál-Ta ījīà theàCh o i le àa dàa o gà Ch istia sàasàtheà oo àRasis.àI àhisà h o i le,àtheàhisto ia àofà u u aàdes i esàtheàareas of Al-Andalus, and the Ebro valley. Al-Ta ījīàide tifiesàfo àfi stàti eài àhisto àtheàe iste eàofà troops called almogavarsài àtheà it àofà“a a us a,àtheàIsla i à)a agoza. These al- aghā ī were Moorish raiders at first, later a mix of Moorish, Aragonese and Catalan people from the borderlands, who made a living out of raiding enemy territory in small bands. By the 12th and 13th centuries, the almogàvers were at the service of the Crown of Aragón. Similar units in the Kingdom of Castille were known as peones. Here is not the place to discuss the bloody war chant of the almogàvers (Desperta ferro!,à Wakeàupà i o ! ,à o àtheàVenjança catalana in Greece after the assassination of the captain of the almogàvers, the former Templar Roger de Flor (1305), culminating in the conquest of Athens for the Crown of Aragón in 1310. Suffice it to say that the Arabic adjective attāl describes some of the actions of the almogàvers very aptly. I would suggest that at the time that these borderland raiders in Aragón, in de area of Zaragoza, were known as al- aghā ī , similar bands in the borderlands at the southern edge of the Spanish March (Gothia), might have been nicknamed al- attālū , and that this name was subsequently adopted, as a geuzennaam (Geusenwort), by the inhabitants of the County of Barcelona and allied territories. Indeed, as Corominas states, the first mentions of the word catalan in the Liber Maiolichinus are in a military context, the Pisan-Catalan expedition against the pirates of the Balearic Islands. It may be significant that the name catalan is mostly associated with the military under Ramon Berenguer III, while the civilians are mostly referred to as gothi. And the fact that the words were never adopted by theà oo sà a àpe hapsà eàdueàtoàtheàfa tàthatàthe àk e àe a tl à hatàthe à ea t… Català would then be derived from the word attāl by adding the Romance adjective-forming suffix -an-, whereas the name of the country, Catalunya, is made by adding the Romance suffix -ia to the plural attālū . The form Cat(h)alonia, with o, can be explained as a hyper-correction for Catalunya on the model of Gascunya ~ Gascueña, Gascuonha. The sound law by which short (open) is diphthongised and then reduced to u in catalan before yod (as well as the parallel development for > i) had already happened in the oldest Catalan document, the Ho ilies d’O ga à (12th century), judging by the attestation of the word nuit ight ( te > nueit > nuit). References Badia i Margarit, Antoni, 1981, Gramàtica històrica catalana, Barcelona. Coromines, Joan, 1989-1997, Onomasticon Cataloniae: els noms de lloc i noms de persona de totes les terres de llengua catalana, 8 vols., Barcelona. Gran Enciclopèdia Catalana (http://www.enciclopedia.cat/enciclop%C3%A8dies/gran-enciclop%C3%A8diacatalana/EC-GEC-0016436.xml?s.q=Catalunya). Puigneró, Jordi, Catalonia: the origin of the name (http://www.unitat.cat/articles/catalonia.html). Rubio García, L., 1978, Catalán - Cataluña, Universidad de Murcia, 1978. Viquipèdia, Catalunya, Otger Cataló, Almogàvers. Wehr, Hans, 1979, A dictionary of modern written Arabic, Wiesbaden.