Navigation – Plan du site

AccueilNuméros14I. Descriptions et analyses de la...Written in stones: The Amazigh co...

I. Descriptions et analyses de langues berbères particulières

Written in stones:
The Amazigh colonization of the Canary Islands

José Farrujia de la Rosa
p. 115-138


Gravée dans la pierre : la colonisation amazighe des Iles Canaries

Selon les données archéologiques, l’ancienne colonisation des Iles Canaries fut initiée au début du 1er millénaire avant notre ère par des populations Imazighen. Cette colonisation s’accompagna de l’introduction, dans l’archipel des Canaries, d’inscriptions lybico-berbères, parmi d’autres éléments culturels du monde Amazigh d’Afrique du Nord. Dans cet article, nous analysons la colonisation ancienne des Iles Canaries à la lumière des incriptions libyco-berbères, des écritures latines des Iles Canaries et de la culture matérielle indigène.

Haut de page

Texte intégral

1In memory of Werner Pichler

1. Introduction: The Imazighen of the Canary Islands

2The Imazighen, the indigenous populations of North Africa, have maintained a constant presence since ancient times in the nowadays so called Tamazgha (Hachid, 2000; Chafik, 2005). This African region has experienced various forms of colonisation which, through contacts established with the indigenous people, have given the area a special character. North Africa has been the focus of interactions with “late-comers”, from the founding of Carthage in around 814 BC to the arrival of the French and Spanish colonisers in the twentieth century. The Amazigh-speaking peoples of ancient times, having already encountered the Phoenicians in Carthage, then came into contact with the original “globaliser” (Rome), later resulting in Byzantium. This was followed, more profoundly, by Islam, with the Muslim presence, starting in around 647 AD, proving the most significant (El Aissati, 2005).

3By the 17th and 18th centuries, Arabic had come to predominate in Tunisia and Algeria, although in Morocco the majority of the population continued to live within Amazigh-speaking tribal frameworks. It was only in the nineteenth century that Europe returned to the Maghreb in full triumph, inaugurating another wave of integration within the world economic system through “imperialism” (Maddy-Weitzman, 2006).

4In the case of the Canary Islands (Fig. 1), the Imazighen from North Africa settled in the Canarian Archipelago since the beginnings of the 1st millennium BC and developed a culture on the islands that can be linked to native North African societies and magical-religious practices associated with the religions of the ancient Amazigh (Farrujia, 2014). But we are still far from being able to form a final opinion regarding the situation for the Archipelago as a whole, since the extent of research varies widely from island to island. Also, the present-day situation is unpromising since, although research in recent decades has consolidated the Canarian-African relationship, it is clear that there is still no consensus in terms of origins (how did the islands become populated and colonised? How did the first settlers arrive?). In addition, isolated radiocarbon dates obtained recently are not representative of the entire Archipelago. In this sense, for example, those from the Buenavista site in Lanzarote, which produce a date of C-14 for the 10th century BC, suggest an earlier occupation of Lanzarote (Atoche, 2011), an island which is closest to the African coast. In the case of Tenerife, for example, the most ancient ones produce a date of C-14 for the 5th century BC, and in the case of La Palma, for the 3rd century BC (Farrujia, 2014).

Fig. 1. The Canary Islands

1.1 What archaeology tells us about the insular Imazighen

5The indigenous Canarian culture can only be explained by a continental-Africana ethnogenesis which is inseparable from the culture of certain ethnic Amazigh groups that lived approximately 3.000 years ago. The culture developed in the Canarian Archipelago by Imazighen societies was clearly influenced by insular isolation and adaptation to the island environment under conditions which meant that they were virtually cut off from contact with the African continent and other ethnic Amazigh groups.

6On the basis of current research, it is possible to refer to the existence of relations between some islands during the indigenous period (Tenerife-La Gomera, or Lanzarote-Fuerteventura-Gran Canaria), since certain aspects of the material culture would appear to indicate this. However, some cultural features in certain islands are not found in others. Although they share the same base, the indigenous island cultures developed in isolation, with very little contact with the exterior. Given this, the poor quality of the ceramics, except in the case of Gran Canaria, leads to the conclusion that later inter-island and even cross cultural ex-changes were rare, indicating cultural isolation until the time when the islands were conquered by the Europeans in the 14th century.

  • 1  See the work of González Antón & Arco Aguilar (2007) or Atoche (2011), amongst others.

7This has made the indigenous archaeology of the Canary Islands an extraordinary, marginal and almost unclassifiable historical example of Amazigh or (North) African culture. In other words, the indigenous Canarian universe was unarguably Amazigh, although from the point of view of “positive culture” it is a unique case and an extraordinary product of involution (due to isolation) and adaptation to an island environment. The archaeological evidence (ceramics, rock inscriptions, etc.) and anthropological/genetical type (DNA) evidence are indisputable (Mederos & Escribano, 2002; Farrujia, 2014), as we will argue in forthcoming pages. However, there are many gaps in our understanding of the circumstances in which the first settlers arrived in the Canary Islands. We still do not know how or why the North African Amazigh landed in the Canary Islands in the first millennium BC, although the early colonisation of the islands has recently been related to Phoenician-Punic influence in the Atlantic: the islands could have been colonized by Phoenician traders who brought over North Africans.1

8The indigenous Canarians lived mainly in natural caves (and to a lesser extent in man-made caves cut into rocks), usually near the coast, 300-500 m above sea level. These caves were sometimes isolated but more commonly formed settlements, with burial caves nearby. Gran Canaria is the only island where settlements with stone houses forming important urban concentrations can be found, although isolated houses have also been documented. In terms of subsistence, animal husbandry was the main means of support for the indigenous societies in the different islands, with the exception of Gran Canaria where agriculture was more developed, including both dry and irrigated farming. The herds basically consisted of goats and sheep and, to a lesser extent, pigs, all of them imported from West North Africa and adequately adapted to the climate and enviroment. Gathering plants and fishing also provided significant food resources (Mederos & Escribano, 2002).

9Ceramics are the artefacts most commonly found in archaeological excavations and also the most widely studied. This is due to the research possibilities they offer, using a cultural-historical approach (which still prevails amongst Canarian archaeologists), for establishing timelines, food consumption patterns, stylistic trends, etc. In the Canary Islands they are typically varied, with each island presenting both formal and decorative differen-ces. The only common feature is that they were coil-built instead of using a wheel. Ceramic items were often incised or burnished, in particular in Gran Canaria, where painted decorations in shades of red, black and white were also common. The La Palma ceramics provide the best stratigraphic sequence, in four phases, although the Gran Canaria ceramics are undoubtedly the most complex, due to the variety of shapes, handles and decorative features. Gran Canaria ceramics also exhibit the clearest affinities with North African Amazigh ceramics (Fig. 2). The Tenerife vessel forms enable parallels to be drawn with those documented in Mauritania and in the central and southern Sahara regions (Farrujia, 2014).

Fig. 2. Decorated ceramic from Gran Canaria.
Photo: El Museo Canario

10The raw materials used in the lithic industry were obsidian and basalt, and the most common tools were burins, borers, racloirs and scrapers. Typical polished stone items were ground stones, used to grind cereals. In terms of the bone industry, which was based mainly on ovicaprine bones, grainers and awls for use in leatherwork were common, as well as fish hooks. Antlers were also used to make tools for ploughing or were set in wood to be used as projectiles (spears). There was also an important wood industry, primarily represented by shepherd’s crooks, combs, shields, containers and doors for man-made caves and houses.

11The majority of the plant fibres used by the indigenous Canarians for clothing and basket-making came from the round-head bulrush (Holoschoenus vulgaris), which wasused to make mats, baskets, bags and shrouds, as well as garments, which were also manufactured from goatskin (Arco, 1993).

12According to the first chronicles and historical sources written by the Europeans just after the conquest and colonization of the Canary Islands (see Tejera & González, 1987; Farrujia, 2014), in terms of social and political organisation, there was a system of matrilineal descent in most of the islands, in which inheritance was passed on via the female line. Social status and wealth were hereditary and determined the individual’s position in the social pyramid, which consisted of the king (known as the Guanarteme in Gran Canaria and Mencey in Tenerife), the relatives of the king, the lower “nobility”, villeins, plebeians and, finally, executioners, butchers, embalmers and prisoners.

13With regard to faith, the indigenous Canarians, like the North African Imazighen groups, worshipped two celestial divinities, the sun and the moon, and sacred natural places such as particular mountains, rocks and caves. Their religion revolved around the need for rainwater, on which the pasture land and crops, and therefore the food for the indigenous people and their livestock, depended. Religious offices were usually held by men, although in Gran Canaria and Fuerteventura these duties were performed by women. Some indigenous sites have been associated with this cult, such as the cave paintings in Gran Canaria whose interiors display painted geometric motifs, the cup and groove sites, consisting of small spherical depressions carved into the rock and linked by man-made channels which are related to the spilling of libations, and the sacrificial altars, built in stone and varying in shape although mainly circular, used to burn animals sacrificed as offerings to the gods (Tejera, 2001; Mederos & Escribano, 2002).

14The world of death was also related to cultural practices. In the Canary Islands, burial sites are one of the most common finds, although they have been plundered continuously since the 18th century. The indigenous Canarians beloved that life continued in another form after death and therefore supplied the corpse with provisions (ceramics, food, awls, beads, rush bags, etc.). They laid the bodies to rest by placing them on beds of stone, vegetation, animal skins etc. to avoid physical contact with the earth. The most common method of laying out corpses was to place them supine inside natural caves or shelters. In Gran Canaria they were also placed in excavated caves or tombs (Arco, Jiménez & Navarro, 1992).

  • 2  For a more extensive and detailed perspective on the archaeology of the Canary Islands, see the wo (...)

15Indigenous funeral rituals also included mummification, which was reserved for members of the “nobility” (as a prestige practice) and has been documented primarily in Tenerife and Gran Canaria.2 In this sense, and according to Abreu Galindo (1602), in the isle of Gran Canaria, nobles and gentry were mummified in the sun and then deposited in burial caves. Also, as Arco (1976) has argued, the mummified remains studied in the Canary Islands present a size above average, an important level of leptosomia, and they lived longer. These, in fact, are the main biological characteristics of the upper social strata. It is also relevant the following aspect: mummification is represented by a small percentage compared to non mummification, and this is indicative of the existence of social differences expressed at the time of burial.

  • 3  Autochthonous (E-M81) and prominent (E-M78 and J-M267) Amazigh Y-chromosome lineages were detected (...)

16Further evidence of the North African origins of the indigenous Canary Island populations can be seen in the rock engravings, featuring a script classified as Libyan-Berber that shows clear affinities with scripts recorded in Libya and Algeria. Moreover, from a genetic point of view, the closest counterparts to 55% of the descendants of the indigenous populations are found in the Maghreb.3

1.2 The gradual disappearance of the indigenous Canarian heritage

17The entire indigenous culture, which had existed in the Canary Islands since the middle of the first millennium BC, began to disappear irreversibly following the conquest and colonisation of the Archipelago, that began in the Late Middle Ages, at the end of the 14th century.

  • 4  Indigenous female lineages have survived in present-day populations since the conquest, experienci (...)

18With the passing of time, the legacy of the individual communities that inhabited the different islands in the Archipelago up to the time of the European colonisation –and provided evidence of their way of life and adaptation and survival strategies in this island environment– began to disappear irreversibly as a direct result of the gradual disappearance of the indigenous Canarian societies. In the case of the Canary Islands, unlike the situation in the American continent (Pérez, 2006) or in Australia (Veracini, 2006), there is no historical continuity between the indigenous, pre-colonial and post-colonial societies, since the conquest and subsequent colonisation of the Canary Islands by the Crown of Castile led to the gradual physical destruction of almost the entire indigenous society.4 This is why in the Canary Islands there is no indigenous archaeology –i.e. archaeological research and heritage management produced by and for indigenous people. In fact, the Canarian economy became characterised essentially by a model based on agricultural development, which, in turn, formed part of the Atlantic and international economy. This led to the development of a dependent and peripheral economy on the islands from the beginning of the 16th century. The resulting colonial society was composed of a ruling group (the aristocracy, clergy and merchants i.e. those who held political positions and controlled the economy) and the majority whom they ruled (labourers, those marginalised for religious reasons, and slaves, including the indigenous Canarians).

19In the case of the Canary Islands therefore, the concept of “indigenous” implies clear temporal connotations: it refers to the populations present on the islands since the time of the first settlements in the Archipelago (in the middle of the first millennium BC) until its rediscovery in the 16th century by Europeans.

20The rediscovery of the Canary Islands by the Europeans therefore led to the gradual disappearance of the indigenous settlements, the elimination of material items from the indigenous culture, the appropriation of indigenous areas, which were occupied by the new settlers and by the emerging colonial society, and the imposition of ways of life and systems of social organisation and production governed by a Western-European world view, values and regulatory mechanisms that were alien to the indigenous Canarian world. Over time, the arrival in the Archipelago of an imperialist archaeology, developed during the 19th century and a good part of the 20th century, would help perpetuate a Western, European interpretation of the indigenous Canarian past, as already argued in other works (Farrujia, 2009 and 2014). An unequal power relationship was therefore the starting point for any consideration of the Canarian heritage. The archaeology of the Canary Islands was perceived through the “filter” of a knowledge that systematically questioned the Amazigh origin of the Canarian indigenous people, arguing instead for a supposedly European root. Parallel to this, archaeological heritage management underwent certain changes that nevertheless failed to improve the organisation, understanding, protection and dissemination of the indigenous heritage, until well entered the 20th century. And therefore, archaeological evidence was analysed on the basis of this discourse and this knowledge was essentially disseminated via the museums, as it has been recently discussed (Farrujia, 2013b).

2. The origin of the Libyco-Berber script
and the Canarian context

21Regarding the Libyco-Berber script, and even if Algerian colleagues (e.g. Hachid 2000) claim its independent invention, one fact should be beyond discussion: such an elaborated alphabetical script without any previous stage of pictographic or syllabic script would presuppose one of the most ingenious acts of invention during the whole history of mankind. We can find no second example worldwide. The assumption that this brilliant creator took the same signs for the same phonemes in a series of cases as the creators of the Old Phoenician alphabet is beyond any credible probability. So it seems reasonable to restrict to the assumption that the Libyco-Berber script was no own invention in the strict sense of the word but a very creative adaptation.

22There can be no doubt about the autochthon invention of script in Sumer and Central America, probably this happened in China and Egypt too. The search for the probable precursor of the Libyco-Berber script is not at all complicated. Central America and China drop out by spatial and temporal reasons, the highly pictographic Hieroglyphic script of Egypt never can be a model for the extremely geometric Libyco-Berber script. So it is no surprise that more than 90% of all researchers agree in the preference of Sumer as the probable place of origin. Till the beginning of the 2nd millennium BC the idea of script reached the coast of the Mediterranean. In the time about 1700 BC the idea of a consonantal script was born in the region between Syria and the peninsula of Sinai. Out of this region the script developed into two different directions:

to the S-Semitic scripts of the Arabian Peninsula

to the NW-Semitic scripts.

23The clear favourites for precursors among these two groups are:

the so-called “Thamudic” scripts

the Old Phoenician script.

24In fact the decision is not really complicated: How should the Berber people of NW-Africa ever have been able to get in contact with the nomads and oasis settlers of the Arabian desert or vice versa? Of course a contact with the seafaring Phoenician people which colonized a great part of the Mediterranean coast is much more plausible.

25Actually, Jürgen Untermann (1997) has proved years ago that the Phoenicians have brought the script to the southern part of the Iberian Peninsula during the 9th century BC. What would be more plausible as to presume a similar act of taking over the script south of the Street of Gibraltar?

26To tell it in short: All indications for this event fit together perfectly:

the system of the script, especially the three half vowels

the time

and the place of the taking over.

-The vowel system

27While the inhabitants of the Iberian Peninsula took over the three half vowels from the Old Phoenician alphabet and added two more vowels under Greek influence, the inhabitants south of Gibraltar kept the system of three half vowels.

-The time

28The development of some Old Phoenician characters gives us a temporal corridor for the taking over to Gibraltar between the 10th and 9th century BC and to Northern Africa about the 8th or 7th century BC.

-The place

29If all what we know about the development of the Libyco-Berber script is not totally wrong, we can identify the oldest inscriptions in the mountains of the High Atlas.

30In 2006 trial trenches were made on the Oukaimeden plateau in the High Atlas, one of them exactly on front of the famous “frise aux elephants” with one of the supposed oldest Libyco-Berber inscriptions. The C-14date obtained from the charcoal of a hearth resulted in an age of 2680 +/- 35 BP. The authors of this study summarize: “Il est tentant (mais osé) d’établir un lien entre la date de ce foyer et l’inscription libyco-berbère de la frise aux éléphants. (…) Mais ceci reste purement hypo-thetique” (El Graoui et al. 2008: 107f).

31This is totally correct. In fact, this C-14date is no proof of the age of the inscriptions. Nevertheless, it is a remarkable correspondence with the age deduced from the historical evolution of script in the Mediterranean.

3. The Lybico-Berber script and the Canary Islands

-The archaic script

32The above mentioned original type of alphabet, which can be called the archaic one (Fig. 3), spread to the East until the Kabylia and to the West till the Canary Islands. The most plausible time of this taking over to the islands deduced from the epigraphic development is the 6th century BC.

33We can find these archaic inscriptions especially on El Hierro and Gran Canaria and single ones on La Gomera and La Palma. Their main features are:

they are pecked (and not carved or scratched)

they prefer round variants (instead of angular ones)

they occur in a clearly definable context of linear and geometric depictions: circles, serpentines, labyrinths, nets etc.

Fig. 3. The main differences between archaic and classic inscriptions

34It is evident that this context shows striking similarities to the so-called “megalithic” repertoire of signs. During the last century D.J. Wölfel and his students tried to search for these roots of an “Atlantic culture”, but we have no evidence of such a connection so far and probably we never will find one (Pichler 2003; Farrujia, 2013a). However, it is no secret that the North African rock art provides hundreds and thousands of examples which show an evident similarity to Canarian ones.

35Thus it is no surprise that we can find in these archaic inscriptions personal names which are identical with ones from African inscriptions:

36NGRN = NGRH (RIL 441)

37STN = STH (RIL 980)

38MZL = MSHL (RIL 586)

39MSKL = MSKL (RIL 713)

40KSN = KSN (RIL 719) etc.

41Who were the creators of these archaic inscriptions? At least since Wölfel it was usual to ascribe the Libyco-Berber inscriptions to seafarer who made a short stop or unintentional visitors like ship wreckers sailors: “My provisional impression is that the greatest part of the inscriptions is situated near the coast. Therefore, it seems reasonable to think of seafarer who visited the islands to bunker fresh water and to have a rest” (1940: 306). Wölfel felt confirmed in this assumption by his own translation of one line from La Caleta/El Hierro (Fig. 4):

42l)£ta = LRYT = lereita = was here.

Fig. 4. Details of the inscription of La Caleta (El Hierro)

43From the epigrapher’s point of view it is necessary to state that this transliteration is definitely wrong. The first sign lstands for /w/ in vertical lines, the sign for /l/ would be P (l turned 90°).

44Nowak, also imbedded by an evolutionist positioning, took over Wölfel’s ideas and resumed: “An assignment of these alphabetiform inscriptions to the original inhabitants seems to be just as bold as improbable” (1986: 70). Only a few very rough inscriptions in a greater distance from the coast were considered to be naive copies of inscriptions made by the inhabitants. All these conclusions were based upon a sample of some dozens of inscriptions from El Hierro only, and as already stated, imbedded by evolutionism. In fact, nearly 90% of the panels on this island are situated nearer than one kilometre, nearly 20% nearer than ten meters from the coast.

45Nowadays we have the counter example of Fuerteventura: on this island all sites except two can be found in distances of 6–12 kilometres from the coast. If we consider that the typical width of the island lies between 18 and 26 kilometres this means a nearly maximum distance of the sites from the coast. In addition we register typical altitudes of 200 – 570 meters, half of the sites lie on the top of mountains. In view of these facts the thesis of occasional or unintentional visitors as creators of the inscriptions becomes obsolete.

46Everyone who ever has dealt with the topic of the Libyco-Berber script knows that we have a second group of inscriptions on the eastern islands of the archipelago. For a better understanding let us first have a look on the further development in the northernmost part of Africa.

-The classic script

47Since the 3rd century BC two Numidian kingdoms developed: the one of the Masaesyli and the one of the Massyli. The Libyco-Berber script of this time, which can be called the classic one, was adopted as an “official script” of these kingdoms, especially used for monuments and gravestones. The considerable influence of the Roman and Punic cultures upon these inscriptions is documented by a series of bilingues.

48Exactly in this era there happened a second wave of immigration to the Canary Islands. Berber people which were accustomed to the Roman culture and script brought a second type of inscriptions which differ from the archaic ones in three points (Fig. 2):

they are carved or scratched

they prefer angular variants

they occur often in a context of Latin cursive inscriptions.

49All of these three points cannot be stated as apodictic conditions. That means: not every scratched inscription belongs necessarily to the classic type but there is a striking preference. The same goes for the preference of angular variants which is caused by the technique: it is much easier to scratch/carve straight lines than round ones.

50The second type of alphabetic inscriptions on the Canary Islands (The Latino-Canarian type) plays an important role for the dating of the associated Libyco-Berber ones. The first examples of this type were discovered in the late 1980s on Fuerteventura and Lanzarote. A research project carried out by Werner Pichler, and sponsored by the Austrian FWF, widened the basis for examination and interpretation from a handful of lines to about 240 lines.

51As it was already demonstrated in the 1990s, this special type of cursive script, which is typical for the border-territories of the Roman Empire, can be dated to the time about Jesus Christ’s birth (Pichler 1994, 1995). This is exactly the time of the colonies of Augustus in Morocco (Tingis, Lixus, Zilis, etc.: 27 BC – 14 BC) and of the Mauretanian king Juba II. (25 BC – AD 23) with his crimson manufactories at Mogador, and probably also in the Canary Islands.

52Among the Latino-Canary lines we can find personal names which are well-known from North African inscriptions (Fig. 5):



55The preferred destination of this wave of immigration were the eastern islands Lanzarote and Fuerteventura, where the classic type of script is the only one, occasional examples seem to exist on Gran Canaria, Tenerife and El Hierro.

Fig. 5. Latino-Canarian inscriptions from Morro
Pinacho y Barranco del Cavadero (Fuerteventura)

56The most frequently asked question is if we can read these ancient inscriptions of the Canary Islands. For a considerable part of them the answer is: Yes. The transliteration of the Canary inscriptions is solved to nearly 100% by new discoveries of the last decades. A basis group of the sign inventory never changed during the 2500 years of development:

57à = M, t = T, O = N, P = L, r = R, ¢ = Y

58Some additional signs did not change their form and phonetic value till the invention of recent Tifinagh:

59m = D, l = W, ú = P/F, I = S

60A series of further characters easily can be recognized as variants of a basic form:

Image 20000009000003C6000003F987DFD0A6.wmf = ú, = ∏ ,  = ¢

61In these last cases a statistical analysis of the frequencies can help to establish assignments to phonetic values with a high degree of probability. In any case the result must be a complete alphabet and a frequency of each character which is usual in the used language/s.

62In fact, in 1994 there were documented some examples of bi-scripts (Libyco-Berber and Latino-Canary) on Fuerteventura (Pichler 1994). Since this time the transcription of Libyco-Berber inscriptions in general is far beyond speculation.

63In the case of these few bi-scripts we can be very sure that the writer had control of both alphabets. He surely comes from a territory inside the borders of the Roman Empire, growing up with the Berbers own script and learning a second one by getting in contact with the Romans. Such a Romanized Berber writes his name in Latin characters and adds his filiation in Libyco-Berber characters: he feels as a participant of the Roman culture but at the same time he is proud of his Berber origin.

64In general we can notice a very close connection between the Latino-Canary and the Libyco-Berber inscriptions of the Canary Islands:

We have some bi-scripts: the same word in both scripts on the same panel: e.g. WMKRN – AVMACURAN

We have examples of the same personal name in Latino-Canarian and Libyco-Berber script on different panels/sites: e.g. WSM ’ - VASIMA

We have examples of Latino-Canarian lines with one Libyco-Berber character for a phoneme which cannot be represented by a Latin character (all sibilants except of /s/): eg. IUFAS

We can notice a distinct influence of the Libyco-Berber way of writing on some Latino-Canarian inscriptions, e.g. the direction of writing

65The second step of reading the Canarian inscriptions is the transcription of the lines, mainly consisting in the addition of the unmarked vowels. This is no problem in the case of several personal names which are attested in Latin inscriptions on the Canaries:

66SM ’ = SIMA

67or – in most cases – in North Africa:


69In some cases we even can translate a Libyco-Berber line:

70WMKRN = AVMACVRAN = son of MAKURAN (PN, male).

The transitional and the Tifinagh scripts

71During a period of some hundred years the evolution from the archaic to the classic alphabet consists only in minimal changes: nearly all alterations of signs can be classified as variants of the basic forms. But sometimes later –we do not know when it happened up till now– the appearance of the Libyco-Berber script changed totally. All phonemes except six were represented by totally new signs from that time on. We call this new type of alphabet Tifinagh. This change didn’t happen in a sudden act of innovation but in a continuous process of transition. We can find examples of this transitional alphabet all over Northern Africa: from Mauritania and Morocco over Ahaggar, Adrar and Air till to the Tassili and Messak. The most obvious feature of this alphabet is the appearance of dotted signs in addition to linear ones:

? together with l

q together with þ

72A further characteristic is the presence of the sequence q r þ (according to Aghali-Zakara 1999:3: “ə(re(“ = “I call”, followed by a personal name).

73We can find only one inscription all over the Canary Islands which can be related with this stage of evolution of the Libyco-Berber script. It was documented at the site Llano de Zonzamas/Lanzarote (Pallares Padilla 1991: 59 - Fig. 12) and contains – quite astonishing - exactly the sequence q r þ - but in opposite order (Fig. 6).

Fig. 6. Inscription from Llano de Zonzamas (Lanzarote)

74The final stage of the historical evolution of the Libyco-Berber script is called Tifinagh. Its evidence is thousands of rock inscriptions all over the whole territory of North Africa – but only very few in Morocco. This last restriction may be the reason for the fact that we can find no Tifinagh inscription on the Canary Archipelago. So the assertion cited again and again in a series of Spanish publications that the Libyco-Berber inscriptions of the Canary Islands originate from Berber/Moor slaves from the time after the Conquest is not legitimate at all. Probably these slaves came from the north-westernmost part of Africa and it looks like if literacy had not survived till to modern times in this region.

The colonization of the islands: a diachronic hypothesis on the basis of Libyco-Berber inscriptions

75Of course we are far away from being able to give any final opinion about the situation of the whole archipelago because the level of research differs enormously from island to island. Nevertheless, the examination of the Libyco-Berber inscriptions indicates a division of the Canary Islands into two parts of Berber influence from different times, but geographically overlapping:

An Archaic Berber culture of the 6th century BC including El Hierro (El Julan, La Caleta etc.), parts of Gran Canaria (Barranco de Balos, Arteara etc.), the one inscription on La Palma (Cueva de Tajodeque) and probably the one of La Gomera (Las Toscas del Guirre) too. According to other archaeological data (Atoche, 2011), Lanzarote, and Fuerteventura probably too, were colonised since the 10th century BC.

  • 5  Some other rupestrian sites in Tenerife have already been related to this period, by means of styl (...)
  • 6  For a deeper analysis of this archaeological proposal, taking into consideration other archaeologi (...)

A Romanized Berber culture since the time of Augustus and Juba II, including Lanzarote, Fuerteventura, probably parts of Gran Canaria (Hoya Toledo, Llanos de Gamona etc.), some inscriptions on El Hierro (Barranco de Tejeleita, Barranco Cuervo etc.), and the one inscription on Tenerife (Cabuquero).5 The connection of the inscription from Tenerife with additional linear-geometric engravings similar to those of Lanzarote and Fuerteventura increases the probability of this assignment. The assignment of the inscriptions of Gran Canaria and El Hierro is highly hypothetic.6

76In addition to these two significant waves of cultural influence we have a very small indication for a contact in the era of the transitional alphabet. But we have not the slightest indication that North African people imported the Tifinagh alphabet in the time after the Conquest.

4. Conclusions

  • 7  In the case of La Gomera it is important to highlight the archaeological site of “Las Toscas del G (...)

77Scientific research into the archaeology of the Canary Islands has been systematically characterized by an interest in issues such as timing and significance, in some cases from evolutionist approaches and, more recently, from a culture-historicism perspective. In this context, and in the case of Libyco-Berber inscriptions, it should be mentioned that the level of research differs from island to island: while we have a satisfying quantity of data from Lanzarote, Fuerteventura, Gran Canaria, El Hierro and La Gomera, we know only one inscription each from Tenerife and from La Palma.7 In this sense, some of the present statements develop in this paper are of course provisional ones and can be modified by new discoveries.

78Libyco-Berber inscriptions have played a really important role from a diachronic point of view, been studied with the aim of sequencing the prehistory of the islands. In recent years this has been accompanied by the emergence of numerous publications, in many cases produced outside research programs and divorced from theoretical discussion, the isolated study of certain rupestrian sites consisting of small-scale historical units (district, ravine, etc.), the failure –in most cases– to incorporate the perspective of spatial archaeology, and the development of research limited to the formal description of rock motifs that does not explore the inherent chrono-cultural or interpretative issues.

79This problem in recent Canarian archaeology is a direct consequence of how the Canarian scientific community deals with the study of rupestrian manifestations and, when trying to unravel such questions as origin and meaning, the scenario remains quite bleak. Currently there are no research programmes that provide for systematic prospecting and excavation. As long as there is no provision for this, the study of archaeology will remain at a standstill.

80In this sense, we are far away from being able to give any final opinion about the situation of the whole archipelago, because the level of research differs enormously from island to island. Nevertheless, we can affirm that the ancient colonization of the Canary Islands (Archaic Berber culture) was initiated by the 6th century BC in El Hierro, La Palma, La Gomera, Tenerife and Gran Canaria. Lanzarote and Fuerteventura, according to other archaeological data, were colonised since the 10th century BC. In a second stage it was introduced a Romanized Berber culture since the time of Augustus and Juba II, including Lanzarote, Fuerteventura, Gran Canaria, El Hierro and Tenerife.

Abreu Galindo J. (1977 [1602]). Historia de la conquista de las siete islas de Canaria. Introducción y notas a cargo de Alejandro Cioranescu. Santa Cruz de Tenerife. Goya Ediciones.

Aghali-Zakara M. (1999). “Les marqueurs d’orientation dans la lecture des inscriptions”. La Lettre du RILB, 5: 2-3.

Arco Aguilar M. C. (1976). “El enterramiento canario pre-hispánico”. Anuario de Estudios Atlánticos, 22: 13-112.

Arco Aguilar M. C. (1993). Recursos vegetales en la Prehistoria de Canarias. Santa Cruz de Tenerife: Cabildo Insular de Tenerife.

Arco Aguilar M. C., Jiménez Gómez, M. C. & Navarro Mederos J. F. (1992). La arqueología en Canarias: del mito a la ciencia. Santa Cruz de Tenerife (Spain): Interinsular. Ediciones Canarias.

Atoche Peña P. (2011). “Excavaciones arqueológicas en el sitio de Buenavista Lanzarote): nuevos datos para el estudio de la colonización protohistórica del archipiélago canario”. Gerión, 29 (1): 59-82.

Chafik M. (2005). Treinta y tres siglos de la Historia de los Imazighen (Bereberes). Rabat (Morocco): Institut Royal de la Culture Amazighe. Centre de la Traduction, de la Documentation de l’Edition et de la Communication.

El-Aissati A. (2005). “A socio-historical perspective on the Amazigh (Berber) Cultural Movement in North Africa”. AfrikaFocus, Vol. 18 (1-2): 59-72.

El-Graoui M. et al. (2008). “Recherche d’indices chronologiques sur le passage des graveurs de rochers de l’Oukaimeden (Haut Atlas, Maroc)”. Sahara, 19: 105-108.

Farrujia de la Rosa A. J. (2003). “The Canary Islands under Franco’s dictatorship: archaeology, national unity and African aspirations”. Journal of Iberian Archaeology, 5: 209-222.

Farrujia de la Rosa A. J. (2005). Imperialist archaeology in the Canary Islands. French and German studies on prehistoric colonization at the end of the 19th century. British Archaeological Reports. International Series, 1333. Oxford: Archaeopress.

Farrujia de la Rosa A. J. (2010). En busca del pasado guanche. Historia de la Arqueología en Canarias (1868-1968). Preface by Alain Schnapp. Santa Cruz de Tenerife (Spain): Edicion Ka.

Farrujia de la Rosa A. J. (2013a). “Le colonialisme et le ‘Collège Invisible’ dans l’emergence de l’archéologie des Îles Canaries au xixe siècle”. Bulletin du Musée d’anthropologie préhistorique de Monaco, 53: 53-61, 2013.

Farrujia de la Rosa A. J. (2013b). An archaeology of the margins. Colonialism, amazighity and heritage management in the Canary Islands. Multidisciplinary perspectives in archaeological heritage management. Springer: Nueva York.

Farrujia de la Rosa A. J. (2014). Ab initio. Análisis historiográfico y arqueológico sobre el primitivo poblamiento de Canarias (1342-1969). Nueva edición, revisada y ampliada. Prólogo de Jordi Estévez Escalera. Colección Thesaurus. Santa Cruz de Tenerife: Ediciones Idea.

Farrujia de la Rosa, A. J. (ed.) (2015). Orígenes. Enfoques interdisciplinares sobre el poblamiento indígena de Canarias. Colección Thesaurus Arqueología. Santa Cruz de Tenerife: Ediciones Idea.

Farrujia de la Rosa A. J. & García Marín S. (2005). “The Canary Islands and the Sahara: reviewing an archaeological problem”. Sahara, 16: 55-62.

Farrujia de la Rosa A. J. & García Marín S. (2007) “The rock art site of Risco Blanco (Tenerife, Canary Islands), and the Saharan Horsemen Cycle”. Sahara, 18: 69-84.

Farrujia de la Rosa A. J., Pichler W., Rodrigue A. & García Marín S. (2010). “The Lybico-Berber and Latino-Canarian Scripts and the colonization of the Canary Islands”. African Archaeological Review. Vol. 27 (1), 13-41.

Fregel R., Gomes V., Gusmão L., González A. M., Cabrera V. M., Amorim A. & Larruga J. M. (2009). “Demographic history of Canary Islands male gene-pool: replacement of native lineages by European”, BMC Evolutionary Biology. Retrieved February, 2, 2012 from

González Antón R. & Arco Aguilar M. C. (2007). Los enamorados de la osa menor. Navegación y pesca en la protohistoria de Canarias. Colección Canarias Arqueológica, Vol. 1. Santa Cruz de Tenerife (Spain): Cabildo de Tenerife.

Hachid M. (2000). Les premiers berberes: entre Mediterranee, Tassili et Nil. Aix-en-Provence: Edisud.

Hodder I. (ed.). (2001). Archaeological Theory Today. Cambridge: Polity Press.

Lozny L. R. (ed.). (2011). Comparative Archaeologies: A Socio-logical View of the Science of the Past. Berlin: Springer.

Maddy-Weitzman B. (2006). “Ethno-politics and Globalisation in North Africa: The Berber Culture Movement”. The Journal of North African Studies. Vol. 11: 71-83.

Maddy-Weitzman B. (2012). The Berber identity movement and the challenge to North African States. Austin: University of Texas Press.

Mederos Martín A. & Escribano Cobo G. (2002). Los aborígenes y la prehistoria de Canarias. Santa Cruz de Tenerife (Spain): Centro de la Cultura Popular Canaria.

Navarro Mederos J. F., Renata A. SpringerBunk and Juan Carlos Hernández Marrero (2006). “Inscriptions libyco-berbères à La Gomera (Îles Canaries): Las Toscas del Guirre”. Sahara, 16: 245.

Nowak H. (1986). “Die ‘alphabetiformen Inschriften’ der Kanareninsel Hierro”. Almogaren. Kanarische Studien II. Hallein.

Pallarés Padilla A. (1991). “Las peñas de Zonzamas y su inscripción líbico-bereber”. Lancelot, 416: 59.

Pérez Guartambel C. (2006). Justicia indígena. Facultad de Jurisprudencia de la Universidad de Cuenca. Ecuador: Colegio de Abogados del Azuay.

Pichler W. (1994). “Die Ostinsel-Inschriften Fuerteventuras, Transkription und Lesung”. Almogaren, XXIV-XXV/1993-94, 117-220.

Pichler W. (1995). “Neue Ostinsel-Inschriften (latino-kanarische Inschriften) auf Fuerteventura”. Almogaren, XXVI, 21-46.

Pichler W. (2003). Las inscripciones rupestres de Fuerteventura. Servicio de Publicaciones del Cabildo de Fuerteventura. Fuerteventura.

Pichler W. (2007): “Origin and development of the Libyco-Berber script”. Berber Studies, Vol. 15. Rüdiger Köppe Verlag. Köln.

Sabir A. (2008). Las Canarias prehispánicas y el Norte de África. El ejemplo de Marruecos. Paralelismos lingüísticos y culturales. Rabat: Institut Royal de la Culture Amazighe.

Searight S. (2004). The Prehistoric Rock Art of Morocco. A study of its extension, environment and meaning. Oxford: BAR International Series, 1310.

Springer Bunk R. (2010). “Los orígenes de la escritura líbico- bereber”. Anuario del Instituto de Estudios Canarios, No. LIV. 141-163.

Tejera Gaspar A. (2001). Las religiones preeuropeas de las Islas Canarias. Madrid: Ediciones del Orto.

Tejera Gaspar A. y M. A. Perera Betancort. (2011). “Las supuestas inscripciones púnicas y neopúnicas de las Islas Canarias”. Spal, 20: 175-184.

Tejera Gaspar A. & González Antón, R. (1987). Las Culturas aborígenes canarias. Santa Cruz de Tenerife (Spain): Interinsular Canaria.

Tilmatine M. (2008). Los estudios amaziges. Bibliografía temática. Barcelona: Ediciones Bellaterra. UNED-Melilla.

Uuntermann J. (1997). “Neue Überlegungen und eine neue Quelle zur Entstehung der althispanischen Schriften”. Madrider Mitteilungen, 38: 49-66.

Veracini L. (2006). “A prehistory of Australia’s History Wars: the evolution of aboriginal history during the 1970s and 1980s”. Australian Journal of Politics and History. Vol. 52, issue 3: 439-468.

Wölfel D. J. (1940). “Vorläufige Mitteilungen zu den kanarischen Siegeln und Inschriften”. In Leonardo Torriani. Die kanarischen Inseln und ihre Urbewohner. Leipzig: Anhang III, 304-310.

Haut de page


1  See the work of González Antón & Arco Aguilar (2007) or Atoche (2011), amongst others.

2  For a more extensive and detailed perspective on the archaeology of the Canary Islands, see the works of Tejera & González (1987); Arco et al. (1992); Arco (1993); Tejera (2001); Mederos & Escribano (2002); or Farrujia (2014), among others.

3  Autochthonous (E-M81) and prominent (E-M78 and J-M267) Amazigh Y-chromosome lineages were detected in Canarian indigenous remains, confirming the north-west African origins of their ancestors, thus validating previous mitochondrial DNA results (Fregel et al., 2009).

4  Indigenous female lineages have survived in present-day populations since the conquest, experiencing only a moderate decline, whereas indigenous male lineages have fallen consistently and have been replaced by European lineages (Fregel et al., 2009).

5  Some other rupestrian sites in Tenerife have already been related to this period, by means of stylistic comparisons with several sites located in Western Sahara (Farrujia & García, 2005 and 2007).

6  For a deeper analysis of this archaeological proposal, taking into consideration other archaeological data together with the inscriptions, it can be seen the work of Farrujia et al. (2010).

7  In the case of La Gomera it is important to highlight the archaeological site of “Las Toscas del Guirre”, due to the high amount of inscriptions documented, now under research (Navarro et al. 2006).

Haut de page

Pour citer cet article

Référence papier

José Farrujia de la Rosa, « Written in stones:
The Amazigh colonization of the Canary Islands
Corpus, 14 | 2015, 115-138.

Référence électronique

José Farrujia de la Rosa, « Written in stones:
The Amazigh colonization of the Canary Islands
Corpus [En ligne], 14 | 2015, mis en ligne le 29 août 2017, consulté le 22 juin 2024. URL : ; DOI :

Haut de page


José Farrujia de la Rosa

Sociedad Española de Historia de la Arqueología (SEHA)
History of Archaeology Research Network (HARN)
University College, London

Haut de page

Droits d’auteur

Le texte et les autres éléments (illustrations, fichiers annexes importés), sont « Tous droits réservés », sauf mention contraire.

Haut de page
Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search