Navigation – Plan du site
Études et recherches

Six days towards the polar star: orientation among Tubu Teda

Six jours vers l’étoile polaire : l’orientation chez les Tubu Teda
Tilman Musch
p. 282-310


En sciences sociales ainsi qu’en sciences comportementales, il y a actuellement un débat pour savoir si, pour s’orienter dans l’espace, l’homme utilise des cartes mentales ou se base plutôt sur sa maîtrise pratique. Un troisième procédé d’orientation, la navigation à l’estime, est, surtout en sciences sociales, souvent négligé par les chercheurs, alors qu’il pourrait être de la plus grande importance dans les vastes environnements qui sont peu ou pas peuplés, comme par exemple les déserts. Le présent texte a pour objectif d’étudier l’orientation chez les Teda. Il est basé sur des observations ainsi que sur des preuves orales et dessinées qui ont été collectées lors de travaux de terrain au Niger. Ces preuves montrent que la navigation à l’estime pourrait être le principe d’orientation de base chez les Teda naviguant dans le désert. Pendant une telle navigation, des traits du paysage acquièrent une fonction polyvalente, parce qu’ils ne permettent pas seulement la vérification mais servent également d’indices temporaux et spatiaux, supportant ainsi une ligne et un cercle de position.

Haut de page

Texte intégral

1As an officer of the French colonial camel corps in Niger and later prefect of Chad, Jean Chapelle was involved in the recruitment of “indigenous” guides and troops. In his monograph on the Tubu (1982 [1958]), living between Niger, Libya and Chad, he wrote regarding their qualities of orientation:

  • 1 Translation of French texts into English done by the author.

Each Tubu knows like the back of his hand the corner where he lived during childhood and where he usually lives a nomadic existence and moves around. It is not on such ground that one can judge his qualities as a guide. Recognizing an itinerary travelled only once many years ago; arriving at an unknown destination only by means of given information; reaching a known destination by a new itinerary; traveling at night guided only by stars; maintaining perfectly his direction on flat ground with no landmarks for several days at a time; this and many other checks allow one to put a guide to the test (Chapelle 1982: 173-174)1.

Chapelle, in his monograph, praises the excellent orientation abilities of the Tubu’s, their readiness to move at any time, their experience in travelling over large distances and their faculties of coping with vast spaces (cf. Baroin 1986). Other authors also underline such qualities, in particular concerning the Tubu Teda, a sub-group of Tubu inhabiting, in particular, areas of the Sahara. Ferrandi, for example, a French lieutenant, stated in the early period of French colonization in Eastern Niger, that the Teda’s faculties of guiding and describing routes contributed to a large extent to the progress of the French army in the central Sahara (1910: 39). Chaliand emphasised, when explaining the Teda’s success during the Chadian civil war (1965-79), their “particular use of semi-desert spaces” and their “extreme mobility” (1981: 33), and Cline attributed to the Tubu a “remarkable memory […] of topography” and a “mentality perfectly suited to a desert life” (1950: 47).

2Nevertheless, the term “mentality”, as used by Cline, is close to the idea of “supernatural” abilities or “senses” for orientation, which Westerners are likely to attribute to nomadic peoples. Such a sense does not exist. As Chapelle underlines, the Teda’s orientation “is not a question of possessing a special sense; but they always need to locate themselves and recognize automatically the finest references” (174; cf. also Spittler 1990).

3In fact, nomads who are often on the move and who inhabit large and scarcely populated spaces have faculties of orientation which seem to be much sharper than those of sedentary peoples. But how does such orientation work? A large number of experiments has been carried out by behavioural science in order to resolve this question. However, the experimental spaces were often artificial and were never of a scale as large as in the Sahara and its bordering regions and could not provide a setting comparable to the regions inhabited and crossed by Teda. The present research is an experiment to investigate orientation in large-scale spaces among Tubu Teda of Niger by adopting an ethnographic perspective. The evidence discussed here considers whether dead reckoning is the basic procedure of the Teda’s orientation.


4Two competing theories, in particular, try to explain how people orientate in space. In anthropology, they are labelled “mental map theory” and “practical mastery theory” (cf. Istomin and Dwyers 2009). Two terms are associated with these theories: route and survey knowledge. Survey knowledge-based navigation happens when people imagine their way and the surrounding landscape as a map, whereas route knowledge refers to recognition of sequences of a (known) route. Both theories have been sufficiently discussed (cf. Istomin and Dwyers 2009), so I will only give a brief summary of them here.

5The practical mastery theory, which probably has its origins in Bourdieu (1977), explains navigating as a movement along a chain of linked landscape images, which Ingold calls “vistas” (2011: 219). According to Ingold, “there is no such [mental] map” (2011: 238), but a navigator travels from one perceived surface (“vista”) to another. This is why Ingold considers the process of way-finding as more comparable to storytelling than to using a map and claims that knowledge about environment is formed when moving and not before (ibid. 230, 239).

6In the same perspective, but from a behavioural viewpoint, Wand and Spelke state, that internal representations on which humans base wayfinding seem to be far from the geocentric and all-embracing properties of maps. Humans navigate by forming and maintaining, “a representation of their momentary relationship to significant environmental locations” (2002: 380). Humans also compare a current environment to stored view representations of places and reorient themselves “by analyzing the shape, but not other properties, of the surrounding layout” (ibid. 2002: 380).

7Gell, on the other hand, states that Bourdieu’s assumption of an almost exclusively egocentric space is difficult to accept, as “we are obliged at all times to locate our bodies in relation to external coordinates” (1985: 278). And Widlok writes that the practical mastery theory “overgeneralizes bodily experiences” but cannot account for orientation skills, which are not centered on direct experiences (1997: 325).

8According to Gell’s definition of a mental map, the latter is “any system of spatial knowledge and/or belief which takes the form of non-token-indexical statements about the spatial locations of places and objects” (1985: 278). People, in fact, seem to be able to build spatial representations without having before explored that represented environment. Such representations build in particular on linguistic descriptions of space, and mental maps can even conserve metric properties of what they represent (cf. Bryant 1997: 242; Denis and Zimmer 1992: 296; Mallot and Gillner 2000: 45). In fact, it is not possible to imagine how Chapelles’s “ideal” guide may orientate in an unknown space without, for example, mental maps based upon oral descriptions of others. As Bryant puts it, “the development of spatial language was very important in human evolution because it allowed individuals to convey environments to others and guide their behavior” (1997: 239). Furthermore, if people use abstract terms for cardinal points representing non-token indexical statements on space, then mental mapping cannot be completely absent in their culture.

9Rather than opting for one of the two competing theories, one should, according to several scholars, suppose that human way-finding is based on both, practical mastery and mental mapping (cf. Istomin and Dwyer 2009: 36; Mallot and Gillner 2000: 45). It seems probable that human wayfinding in well-known spaces is more based on practical mastery, whereas unknown spaces or large-scale environments demand more advanced techniques of orientation, as for example mental mapping and dead reckoning (cf. Widlok 1997).

10In behavioural science, considerable attention is paid to dead reckoning as a technique of way-finding, whereas theoretical discussions of way-finding in current social sciences show little interest in it. Istomin and Dwyer, for example, in their exhaustive review on spatial orientation (2009), mention this technique only in a footnote. This is surprising, as, especially in ethnographies of marine navigation, dead reckoning seems to play an important role. Puluwat navigation, for instance, was described as “entirely a dead reckoning system” (Gladwin 1970: 144).

11Dead reckoning is based on the combination of two constraints: distance and direction; combining both generates a position in space. By choosing a direction in space, a navigator establishes a line of position on which a further position during his travel is located. When he specifies a distance from a current position, then he establishes a circle of position. His further position is located at the intersection of the line and the circle of position (cf. Gallistel 1990: 58; Hutchins 1996: 52-56; Loomis et al. 1999: 127). This is an elementary explanation of how to describe positions in space, and the ability to perform dead reckoning is not only an attribute of humans: In 1965, von Frisch described the “waggle dance” by which bees communicate direction and distance of forage to conspecifics of their hive.

12Nevertheless, dead reckoning is also subject to cumulative errors if it contains faulty data (Lewis 1976b: 27). This is a risk, especially in large-scale environments. The greater the distance travelled on a faulty line of position, the greater the resulting bearing difference becomes. This is why dead reckoning should be complemented by other procedures, as for example view-dependent place recognition and reorientation by the shape of the surroundings (Wang and Spelke 2002: 376). For the same reason, dead reckoning is, in reality, often combined with piloting: “There is a constant interplay between processes like piloting, which requires a map, and dead reckoning, which is in principle sufficient unto itself” (Gallistel 1990: 89; cf. Gladwin 1970: 147).

  • 2 A hypothesis is proposed in which the integration process, an operation that may be difficult for (...)

13A procedure close to dead reckoning but more sophisticated is path integration, which means the ability to compute shortcuts. Foraging desert ants (Cataglyphis sp.), for example, and also other animals, are able to return directly to their housing without repeating inversely the exact route they covered in order to reach forage sources (Collett et al. 1999; Müller and Wehner 1988). Probably, animals are able to generate and store vectors to landmarks, and may have mechanisms “for generating the angular and distance components” (McNaughton et al. 1991: 191). In the case of humans, neuroscience suggests that the ability to perform path integration is due to “linear associative mapping” implemented by means of neurons and completed by the incorporation of visual landmarks (ibid. 190)2. The awareness of distance and direction and the ability to also compute deviations or turns of a path from a linear direction allows a traveller to locate a destination in space and to estimate his current position (cf. Loomis et al. 1999: 129).

14Despite the above-mentioned lack of interest of contemporary anthropology in dead reckoning, this procedure seems to be a basic means of orientation, especially in large-scale environments and also in unknown ones, which are however difficult to study. Dead reckoning seems then to be complemented by two other procedures: updating during the travel and piloting when closer to a destination. As I will show in the following section, dead reckoning should be considered as the basic principle of orientation among Teda.


15The aim of the present research is to investigate orientation by dead reckoning in large-scale environments, in particular in the desert. The research has a strong ethnographic background but also ties in with topics initially debated in behavioural science. It seems that an ethnographic approach, in particular, is able to carry out research in large-scale environments, where behavioural science often encounters problems due to the difficulties of framing such environments into experimental settings. Istomin and Dwyer (2009: 32), for example, criticize the fact that the evidence of Wand and Spelke (2002) on way-finding is provided from experiments in small-scale environments of room-size, and that such results would be problematic to apply to large-scale environments. Desert environments, especially, which are the focus of interest of this paper, may be difficult or impossible to reproduce under experimental conditions. For example, the so-called “desert world”, which Foo et al. (2005: 198) use in one of their experiments, consists “of a blue sky above a circular ground plan (100 meters in diameter) that was textured with a multiscale random gray pattern”. Such an artificial and limited environment cannot at all represent a real desert.

16Of course, from the viewpoint of behavioural science and its current understanding of methodology, the present research may be criticized too. Criticism may concern the heterogeneity of informants and spaces, and the fact that there are no travelling experiences which were gathered or shown under similar experimental conditions. This is, however, a feature which is inevitable in the nature of the things being studied: large scale environments are per se heterogeneous, and deserts, as potentially dangerous environments, cannot be used as experimental spaces where an optional number of experiments may be carried out. The aim of each desert traveller is to cross and to leave the desert as fast as possible (cf. Spittler 1990). A researcher who would make a desert journey several times “just” to observe how different drivers and guides interact with their environment would be misunderstood by locals and even seem suspicious to them, putting at risk confidence between the researcher and his interlocutors which is extremely important in an ethnographic approach.

17The large-scale desert spaces where my interlocutors navigate are not homogenous, unlike the situation of an experimental setting. On the contrary, each of the four cases presented below is different in its landscape patterns. Even if one chose only one restricted geographic area for investigation, one could never be sure that people moving in this space would take the same way since navigating in the desert happens in a “network of tracks” where a traveller chooses his way according to “seasonally variable conditions” and also according to “individual considerations” (Ritter 1992: 147).

18Is homogeneity of investigated space necessary in order to do research on orientation? The heterogeneity of spaces in Eastern Niger, in which the present travelling experiences were gathered, offers the chance to investigate how the same basic procedures of orientation are used in several different environments: the Manga, a pastoral region in the borderlands of the northern Sahel and the beginning of the Sahara; the Kaouar, a chain of oases alongside a north-south escarpment not so far away from Libya and Chad; the Erg of Bilma between Agadem and Bilma; and the Tenere desert between the Kaouar in the east and Agadez in the West.

Fig 1: Map Overview

Fig 1: Map Overview

19In the following sections, I will present evidences of way-finding by my interlocutors, in particular oral route descriptions and mental maps. By means of these descriptions and maps, it becomes understandable how Teda orientate themselves within the four above mentioned regions. Such descriptions and maps are, however, more than ethnographic material: they tie in with the human spatial behavior of conveying space to somebody by language (cf. bryant 1997) and of using survey knowledge. Each of my interlocutors who described space or drew a map for me not only furnished an illustration for my research but also wanted me at least to understand the respective space and perhaps even to make me able to orientate and travel there alone.

  • 3 Fieldwork and the present research were financed by the German Research Council (DFG).

20The data I will use in this paper was gathered during several fieldwork visits among Tubu in Eastern Niger. Most of the evidences on way-finding was obtained during a fieldwork visit of more than three months at the beginning of 20143. During that time, I travelled a lot with locals in the four relevant regions and was able to do research by participant observation and open-ended conversations. Travelling, and staying with them, offered the possibility of discussing with Teda the topic of orientation and also of collecting narratives of navigating and mental maps. Most of the research was done on the route, some also in towns or settlements. As Spittler (1998: 30-31) pointed out, travelling offers a productive moment for ethnographic research, because the travelling community is somehow “disconnected” from every-day social relations (cf. also Lewis 1976a: 249). Moving with Teda and, at the same, doing research on how they navigate, allowed me thus both to work during particularly productive (travel-) moments and to observe by participating in the move.

21I used mostly French as the working language (many Teda, even those who did not attend school, can express themselves in French), but some evidence was gathered in Tedaga. Numerous differences in the orthography and the spelling of place names may constitute a problem. In fact, one can encounter in the relevant regions names of Teda and Daza (another Tubu group), and also of Kanuri, origin. Almost all of these names also have their “francized” variant(s), which can be found on maps. I tried to conserve the more or less “original” orthography when a place represents a landscape feature which is important for the Teda’s orientation. For other names, in particular these ones of bigger places (Seguedine, Dirkou…), I use the (or one of the) French orthographies.

Dead reckoning among teda

22Some principles of dead reckoning and orientation among Tubu Teda Many colleagues who have done fieldwork among nomadic peoples know situations similar to the following: A traveller is searching for a camp, a well or another place and asks a local the way. The latter takes the traveller to an open place, points with his hand in one direction where, at that moment, nothing particular is visible and indicates a distance: this is how the traveller will find the way to his destination. This is a basic principle of way-finding: dead reckoning by combining direction and distance.

23Among Teda, cardinal points are basic parameters of orientation (yala: north; ônum: south; ma: east; dî: west). As for landscape patterns, “the nomenclature of geographical aspects is poor” (Le Cœur 1956: 114) and “it is indispensable to know the necessary time for crossing these wide spaces, and according to the time distances are evaluated” (ibid. 114).

  • 4 Not to be confused with the measurement of time. “I stayed two days” is said, diske čû tudugu (disk (...)

24When asking a Teda about the distance to a place, the latter answers in terms of time, mostly in terms of camel journeys: “Bilma is distant from Agadem six days on a camelback”. On a average camel, a traveller can cover approximatively 40-50 km a day. Normally, he rides 25 km in the morning and 25 km in the evening. Such a one-day-ride is called orro: “I have made two days by a camel” is said orra čû têrgan4. In environments of a lesser scale, one can also express a distance in terms of topography: sentences like “you find the camp after having crossed two depressions [Kanuri: kwaano]” are used especially in the Manga when people describe shorter journeys, for example from a well to a tent.

  • 5 Contemporary Teda seem not to use a large number of stars for orientation, as in the case of “star (...)

25In order to recognize cardinal points, Teda observe the sun and the stars. Two of the latter are especially important: α Ursae minoris which is named “the northern star which doesn’t move” (teski yedi yala), and probably α Crucis, “the southern star which doesn’t move” (teski yedi onum). Significantly, the “guide” in Tedaga is named teskide, a derivation from teski (“star”) (Le Cœur 1956: 49)5.

26If, due to darkness and bad visibility, neither stars nor the sun are visible, Teda are able to recognize cardinal points with the help of other features. The prevailing direction of the wind may allow them to orientate themselves. If there is no wind, they observe dunes or even small piles of sand around blades of grass or other obstacles which enable them to distinguish windward and lee sides and thus to decide on the prevalent wind direction and to work out the cardinal points. Ripples in the sand also allow recognition of the wind directions.

27Observation of landscape features complements the information provided by dead reckoning and allows a traveler to pilot and update his path knowledge. Salient landmarks are mountains, rocks or hills (êĩ, emi, gora). Wells (yȋge) also play an important role especially for the social comprehension of space. Significant attention is paid to changing aspects of the ground. “The earth changes” (besama forostene) is an important sentence when describing a way. Ground can be red (madu), white (čûu), yellow (mîni) or black (yesku). Sometimes, a traveller even has to dig in order to see if the ground changes its colour below the surface and thus to recognize a particular place: “The earth becomes yellow when you dig” (besama ledoã mîni).

28Teda classify sand not only by its colour but also by its granularity. It can be a very fine quartz sand (besama keker), “soft” sand (besama lili), “glittering” sand (besama yiliyili), fine grained (besama neski) or coarsegrained sand (besama gegera). Names also exist for other types of grounds, as, for example, white lime (kalko, kukur), clay (lo), black-reddish basalt (bizi) or a dusty red mixture of sand and clay (wuru).

29A profound comprehension of direction and distance, and the ability to also recognise the finest particularities of landscape, contribute to an excellent comprehension of how to orientate in space. Westerners as many other sedentary peoples have probably lost such faculties. This is why they may consider them wrongly as “supernatural”. As Chapelle stated, many nomads permanently sharpen their senses for orientation, and this fact allows them to display an excellent knowledge of space and locations: “The nomad is living ‘in a state of orientation’, a state where his senses maintain him without reasoning intervening all the time. But orientation is not a sense by itself. This faculty, even if highly developed, has nothing supernatural” (1982: 174).

Wells and trees of the Manga

  • 6 This is a very general delimitation, and not everybody agrees with it. For some of my interlocutors (...)

30The Nigerian Manga reaches Agadem (Tedaga: Ogorom) in its north, N’guigmi in its south, the Chadian border (Kanem) in the east and the Termit massif in the west6. Teda agree that this region formerly offered excellent conditions for camel husbandry due to the abundance of Cornulaca mona-cantha (cf. Also Chapelle 1982: 116). For this reason, many Teda families from the Tibesti or from other regions migrated here after the arrival of the French colonists, when the latter had secured the Manga from raids and started to build cemented wells. Men still often travel between the Manga and the Kaouar or other regions. Even women and children travel each year to the Kaouar in order to harvest dates.


32The linear distance from Bedouaram (Tedaga: Bidûarem) to Agadem is about 120 km. That means that, by camel the journey will take about three days, whereas a driver of a motorised vehicle can do it in less than one day. As people and animals are numerous in the Manga, the region is not considered to be a navigational challenge. When I asked Moussa, a 34-year old professional Teda driver of the Tezeria clan, to describe the way between

33Agadem and N’gourti, he at first did not understand the task – navigating here seemed too easy for him. Then he explained briefly:

  • 7 Texts from Teda informants translated by the author.

Agadem-N’gourti, there are many wells, […] and also there are trees on the dunes. And then, there are three wells and two like that. All this is there until N’gourti. I have the GPS, but all is direct. I have the coordinates of N’gourti. It is even easy to travel. It is a flat ground. When you come from N’gourti, there are two wells for Arabs. When you pass these two wells, you will arrive at the wells of the Tubu, of the Teda. It is so easy7.

  • 8 N.B.: Moussa described the way from north to south, whereas my description, after Moussa’s text, is (...)

34Let us have a closer look at Moussa’s description (cf. map 1)8. For him, wells and trees play a major role. There are no salient hills or mountains in the Manga which could serve as landmarks. Moussa speaks about two wells of Arabs in the near north of N’gourti. These are Zelege and Bende where Shuwa people live. Then, from Bedouaram northwards, the land is considered as the Teda’s with the main wells of Bedouaram, Grema and Bilaberim. Nevertheless, wells do not really provide a bearing from a far distance as they are not prominent enough and are too much embedded in the landscape. What is it Moussa wants to say when associating wells and the easiness of way-finding?

35Moussa is probably emphasizing the sociality of space – where many Teda people are present, a Teda navigator cannot get lost. Lewis stated a similar reasoning based on sociality around water when discussing the social impact of orientation. He cites an Australian Aboriginal who tells him: “In a strange country I would not note the hills and other places. I would look for waterholes and I would ask people” (1976a: 274).

36Such sociality of space becomes even more visible in the typical behaviour my Teda co-travellers showed when approaching the numerous camels or camel flocks that one encounters in the Manga. When approaching them, they asked the driver to slow down or to stop and tried to recognize the brands of the animals. Through brands, they learn about tribes or families present in a particular place. This knowledge also allows travellers to locate themselves socially in space.

37Moussa mentions trees too. He does not mean big single trees serving as salient landmarks, but wooden vegetation, which changes gradually when travelling on a longitude. Such changing vegetation can serve as an indicator to which degree a traveller has progressed. From the south up to Bedouaram, Leptadenia pyrotechnica (kyzyn) is very abundant; more northwards, this is replaced by Acacia tortilis (tefy). The latter is characteristic until Bilaberim where another species becomes dominant Capparis decidua– (kûsonw).Infact, when travelling here we crossed several depressions with groves of kûsonw, and even an ancient but now abandoned well is named for this tree – Kûsana.

  • 9 The coordinates of these trees are N 15° 58.012‘, E 13° 12.084‘ for A. tortilis and N 16° 24.035’, (...)

38Single trees can also play a role as indicator. On our way northwards, locals travelling with me called my attention to a single exemplar of A. tortilis and, later on, to two trees of C. decidua9. My co-travellers explained that these trees were the most northern exemplars of their species and thus attract their interest.

39In the north of Kûsana, the flora becomes more and more scarce. Teda travelling here now pay attention to surfaces, for example those covered by dark basaltic minerals (bizi), which are emerging to the west of the track. Then, after some time, the mountains of Agadem appear and piloting begins. The village of Agadem is located in the west of the mountains and, before arriving here, one again passes by the Source of Agadem (Šiši Ogorom), which springs up at the foot of big palm-trunks besides the track.

40Moussa underlined in his description that the route is more or less “plain”. This is one reason why the way is “easy” for him. Another one is that the space to be crossed is inhabited by many Teda families, a fact that may strengthen the idea of a space where a traveller cannot get lost. But the main reason for the easiness of the route may be the fact that it goes more or less straight from south to north or vice versa – the route is “direct” as Moussa calls it, and represents a line of direction parallel to longitudes. In order to now establish a circle of position, a traveller can rely on the odometer of his vehicle or on his GPS, or on “traditional” measurements such as camel journeys. The felt-travelled time may serve as an indicator of the distance too. Moreover, the succession of wells, of different vegetation types and finally the emergence of basaltic surfaces can be considered as non-metric measurements of distance. They embody the principle of “circles” of position and tell a traveller up to which degree he has already progressed towards north or south.

Ways between Bilma and Agadem

41Anour is an experienced 52-year old Teda guide from the Hurtuna clan. I met him in Agadem and asked him to explain to me the way from here to the Kaouar. He answered that I only had to go northwards with a very slight deviation towards west. After about six days on a camel, I would arrive at Bilma. He described the route with the following words:

If you are going to the north, if you are going to the Kaouar, the polar star is on your forehead; your right hand, your right eye, you have it [the star] there; you will not fail a well; it [the star] will bring you there and you will be in the Kaouar. Yala ndedio tigge ndedio hombûnuma teski yedi bûze kubonûma bore sonûma bore tã ginekti taõ yiigehi ndouki tigge negre.

42Bilma is located in the north of Agadem, but on a more western longitude. This is why, when travelling from Agadem northwards to Bilma, the Polar star hast to be not in the middle of the traveller’s forehead but “in his right eye” and “on his right hand” in order to allow a slight western deviation.

Fig 3: Map Erg of Bilma.

Fig 3: Map Erg of Bilma.

43The description given by Anour is very essentialist, which may be due to his long experience in way-finding. Mamadou, a school director from Agadem (about 35 years old; clan of the Teda Gunna), described tracks between the Kaouar and Agadem in a much more detailed way. He originates from Dirkou and has travelled many times by camel and in cars between his working place and his home town. Mamadou gave me two exhaustive descriptions of routes from Bilma to Agadem. The first way is the “traditional” one taken by travellers on camelback (see here below); the second one is a more recent route for motorised transport.

After Bilma […] one goes directly to the south and reaches Moyegaziniy [Tedaga: “Green Aristida sp.”]. When passing by Moyegaziniy, when going directly to the south, one enters in dunes aligned from east to west. In the middle of these dunes, one reaches Kâu Tulo [Kanuri: “One stone”]. When approaching Zow, there is [fi Zow Dow. […] After only seven kilometres, one reaches Zow Baba.
After Zow, one pushes directly forwards to the southeast, and enters in the big dunes of Ači Ačide aligned from east to west, which are difficult to cross. When leaving them […] one passes by Gore Daha Frama [Tedaga: “Mountain with a flat and large head”]. One continues always directly to the south and reaches after two depressions Gore Arkuna Yidaan [Tedaga: “Mountain of
Maerua crassifolia”].
After that, one goes further directly to the south and reaches Emi Čuukoy Yala-an [Tedaga: “Two Stones of the North”]. When then passing directly to the south one reaches Emi Čuukoy Onum-gan [Tedaga: “Two Stones of the South”]. After that, one goes to Gore Zugur-an [Tedaga: “Mountain of the Hyenas”]. After that, when going directly to the south, one reaches Âgâsâ Kâlâsti [Tedaga: “Sharpen the Swords” (on a rock in the neighborhood)].
Once more heading south, one reaches Dibbela under the big hill named Yirêrom. This comes from the word Yirê which means “truth”. It is to say that, when one sees Yirêrom, one has really arrived at Dubulla. There are springs, date palms and palms of
Hyphaene thebaica.
One leaves Dibbela, and then has on his way Čigirêne Či Busahudinyi [Tedaga: “Čigirêne with the Nose of Sand”]. One goes to the south and has there Čigirêne Čugurondaan [“Čigirêne with the Curved Nose”]. From Čigirêne people can turn to the east to go to Mussoley and pasture their camels there and then turn again to the south. After that, one goes again to the south and reaches Gore Barad-an [Tedaga: “Mountain of the Hunters”]. When one passes
Gore Barada-an, the sand becomes red. One goes then to Malka-an, which are big dunes. After Malka-an, the sand becomes a bit white. In the middle are some very white dunes named Daddara-an. […]. One passes in the east or west of Daddara-an and reaches Kusomiyi Elia-an [“Tedaga:
Capparis decidua of Elia”]. When leaving Kusomiyi Elia-an, after marching only some kilometres, one comes down to Agadem.

44Mamadou’s route description consists of three kinds of information: directional indications, waypoints with local toponyms and some measurements of distance. The basic-kind of information seems to be the directional indications, which often introduce a further discourse (“After Bilma, one goes directly to the south…”, “one goes to the south and has there…” etc.). Thus, the most important thing a traveller has to do when starting or continuing his journey is to head towards a cardinal point, which is, in this case, mostly south. The second kind of information given is waypoints. Interestingly, Mamadou does not advise to “head” for them or to “orientate” by them, but he states only that one “reaches” them or “passes” by them. This is why one would not call them “vistas”, which orientate a traveller. They are rather marking a stage of the journey and allow verification that, after some time, a certain point of the southwards leading line of position has been reached. Thus, they could be considered as intersections of the line of position and a partly-represented circle of position. The latter expresses distance not in terms of time or kilometres but in terms of landscape features embodying a distance from one place to another. Metric distances – the third kind of information given in the description – are used much more scarcely (“seven kilometres”, “some kilometres”). The expression of distance in terms of “two depressions” seems finally to be a compromise between metric distances and distance-marking landscape features.

45An interesting point is the deviation a traveller may take to Mussoley in order to pasture his camels there. After pasturing, he doesn’t have to return the same way in order to reach the main route, but can go directly southwards in order to re-join the way to Agadem. This is an example of path integration (cf. also Chapelle, who, in his initially quoted comment on how to test a guide, wrote that the latter should be able to “reach a known destination by a new itinerary”). A comparable example of how people take shortcuts is provided by Lewis. An Australian Aboriginal told him: “If I go south 10 miles, then a little east, to get back I must go north 10 miles and a little west. If there are no landmarks, I still know the direction.” (Lewis 1976a: 262-265, here 265).

46The second way Mamadou described is the “modern” track for motorised vehicles. It is congruent with the following description given by Moussa. The latter is not an autonomous text, but the explanation which accompanies a mental map he drew in my notebook (see illustration).

Fig 4: Mental Map A.

Fig 4: Mental Map A.
  • 10 We were in Bilma when Moussa drew the map, and thus his GPS automatically calculated the distances (...)

47When Moussa described the way, he sometimes looked at his cellphone-integrated GPS in order to verify metric distances, which he nevertheless seemed to know by heart. The fact that he considered it important not only to draw and to describe the route, but also to complete this information by metric indications shows the importance he attaches to the principle of a circle of position10.

Here, this is Bilma (1). If you leave Bilma, you will exit a bit to the south (2). East (3) and south, like this, a corner like this […] and you will arrive at Ankri [20 km] (4). A bit southeast. This is Ankri here […]. There are many mountains.
If you push forwards here at Bilma, there is again Moyegaziniy, this is the camel track. […] There is a well here, and there is a well there. On the route of the vehicles, it is named Ankri, on the route of the camels, it is named Moyegaziniy.
In the south of Ankri, there are dunes like this. Dunes, which are aligned to the east, like this. When you go through these dunes here, you will go like this and then you will go like this and then like this (5) and you will enter Zow Dow (6). […] There are many sand dunes. And also some Guši
[Panicum sp.]. At Zow Dow, there are hills and mountains. When you are at Zow Dow, you will continue like this (7), and you will enter Zow Baba [52 km] (8).
When you are at Zow Baba, you will leave towards east. Because of many sand dunes, you will not leave directly for Agadem. You will continue to the east, the west, and you will make zigzags (9). There are also dunes. After that, you will enter to the south (10), to the west (11), and you will continue to the south (12) again. […]. Now, here, this is Daha Frama [73 km] (13). There is a big mountain, a bit large, and you exit here, you will leave like this, and you will leave like this (14), you will leave like this (15), and you will return like this (16) to Dibbela (17).
Here, at Dibbela [128 km], you will advance like this again (18), you will leave for Agadem, all direct [209 km] (19). Here, this is Agadem (20).

48Moussa’s text is more prosaic than Mamadou’s in what concerns landscape patterns, but he is perhaps even more meticulous in mentioning directions. “South”, “west” and “east” appear several times alone or in combination, and sentences or terms like “you leave like this”, “zigzag”, “straight” or “direct” are directional indicators too. In his mental map, he tries thus to depict realistically the route and its turns. His meticulousness in describing every detour may be explained by the fact that a motorised vehicle has often to circumnavigate obstacles (e.g. dunes), whereas Mamadou’s travellers on camelback can advance in a straighter line. This is why, for Moussa, dunes, hills and mountains seem to serve to a much lesser degree in verifying one’s position or to embody a measurement of distance (for that, he may rely on the odometer of his vehicle or his GPS). More, they represent difficult terrain, which has to be circumnavigated when driving a vehicle (cf. also below his description of the track to Agadez).

Kaouar. Alongside the escarpment

  • 11 For many Teda, “Kawar” means the region between Beza and Aney, which is covered by the Teda-canton (...)

49The Kaouar is a chain of about 14 oases, which are located on the western side of a north-south escarpment (cf. LeSourd 1946). These oases are mostly inhabited by Tubu Teda and Kanuri11. A formerly-intensive caravan trade relied on the Kaouar to travel to Agadez and the Aïr mountains in the west (cf. Fuchs 1989: 18-20; Ritter 1992; Spittler 1984). Caravans also crossed the Kaouar on a north-south route, travelling from Tripolitania through the Fezzan to reach Bornu and Inner Africa (cf. Fuchs 1983: 120, 131-132). The local economy of the Kaouar is based mostly on salt exploitation and date culture – products that explain the interest of caravan traders. However, raids from Teda, Tuareg and Arabs had a severe impact on local economies before the French colonial army “pacified” the region (cf. Chapelle 1982: 112).

50Orientation in the Kaouar is relatively easy, as the mountainous escarpment in its east generates a kind of longitudinal line of position. When travelling along this line, a navigator has just to bear in mind the distance from one place to another in order to establish a circle of position. Distances separating the respective oases are, in fact, very present in the mind of almost every inhabitant of the Kaouar. When, for example, people from Dirkou wanted to describe to me an oasis or to tell me something about its history, they never forgot to “introduce” the respective place by information about its distance from our current location. For example (from Dirkou): “Chemidour is very near in the south, three kilometres only”. Or: “Achenouma, this is a village 15 km northwards from here”.

Fig 5: Map Kaouar.

Fig 5: Map Kaouar.

51When speaking about a village and explaining how to get there, people often mention not only distance and direction, but add also information about mountains in its vicinity. In fact, for every village of the Kaouar, one or two mountains are associated which are salient in shape or colour patterns and which bear a proper name in Teda, Kanuri or even Tubu Daza language. The following text on villages and mountains was compiled with information obtained from Hachemi (53 years; Tumagura), the chief of Dirkou, and Agi, a former deputy and sub-prefect of Bilma (61 years; Gunna):

When travelling from the south towards Bilma, 20 kilometres before the town, the mountain Moya-tim (see “Moyegaziniy” in Mamadou’s description above) which is the only one in this place appears slightly in the east: “You can see only this sole mountain here, nothing more”. Later, still before Bilma and still in the east, appears the mountain Silema, the surface of which is very dark coloured and the summit of which is flattened. Having passed Bilma, the traveller arrives at the village of Aguer with its mountain Tima-mi. The next village, Beza, is located westwards of two salient mountains: Nguso-mi (Kanuri: “The House of the Hyena”) in the south and Emi Mada (Tedaga: “Red mountain”) in the north. Three kilometres before Dirkou, the traveller arrives at Chemidour with, in its east, the mountain Kâu Sǝmo Indí-a (Kanuri: “Stone with Two Ears”). The next town is Dirkou with the mountain Mugudum. After again seven kilometres, the traveler arrives at the villages of Argui with its mountain Têgema (Tedaga: “Castle”). More northwards, after 15 kilometres from Dirkou, follows Achenouma with the mountain Tima, a village inhabited by former Teda slaves. In order to arrive at Emi Tchouma, a place 36 kilometres in the north of Dirkou, one has to pass by the mountains Emi Tirkema (Tedaga: “Camel Saddle Mountain”) and Šafarom. Some kilometres in the north of Emi Tchouma follows then the village of Aney. Finally, in order to get from Aney to Seguedine (or: Sûuw), the traveller has to pass by a wide space without major settlements. Two mountains mark this way: Gre Zugra (Tedaga: “Hill of the Hyena”) and Yegeba. Seguedine is then reached after having passed the Emi Zumri (“Splendid Mountain”).

52The characteristic mountains nearby a village make piloting possible when approaching the latter. Ferguson and Hegarty speak in this context about a hierarchical organization of a natural environment, with landmarks considered as “anchors” (1994: 456). Moreover, as was the case in the Manga and the Erg of Bilma, landscape patterns – in this case mountains – can serve a traveller by updating his path knowledge. That means, when travelling alongside the escarpment, they constitute intersections of the line of position, on which a traveller moves northor southwards with a circle of position they partly embody. In this way, the traveller can determine his own position.

53Four further mountains are located in the Tenere desert west of the Kaouar: Nguso-mi (Kanuri: “House of the Hyena”), Kafran (the name of a Nigerien Corporal who got lost here), Fasana and Tigizalan (the Tamasheq-name of the latter attesting its importance for former Tuareg travellers). They constitute some sole landmarks for desert-travellers and allow them to compute their position by combining cardinal points with the angle from which they perceive the mountain. Nevertheless, from many places these mountains are not visible, and travellers base their way-finding exclusively on dead reckoning. The deputy Agi narrated, for instance, the case of caravans which wanted to travel directly from Fachi to Djado. When, at the periphery of Fachi, they prepared to enter the desert, they headed northnortheast without any landmark. Only towards the end of their journey (about six days later), was Mount Zumri emerging which confirmed to them that the village of Seguedine, the door to the Djado, was close.

54However, it was not only caravans which took such shortcuts. Raiders were even more mobile, and their abilities of path-integration allowed them to stay hidden in the desert until they were approaching an oasis, when they could rapidly attack the village and then retire immediately to uninhabited places where they continued on their way.

Through foreign land: the Tenere

  • 12 Tubu call such places awi, a term which designates the absence of vegetation, water and people – a (...)

55The way from Agadez to the Kaouar is by crossing the Tenere desert. “Tenere” is a Tuareg name which means “desert” or “empty land”12. Different caravan tracks exist in the Tenere, some of them going to Fachi, others, more northern ones, directly to Bilma and the Kaouar (cf. Ritter 1992). Caravan trade was monopolized by Tuareg of the Aïr Mountains who brought particular products of animals and cultivation to the Kaouar and who bought or exchanged there salt and dates (cf. Spittler 1984). Teda did not take part in these caravans as they often maintained hostile relations with Tuareg (cf. Fuchs 1983: 126-132).

56Today, caravan trade has considerably diminished, and motorised vehicles are more and more replacing camels. “Smuggling” goods and migrants between Agadez (Niger) and Sabha (Libya) has now become a very lucrative business with which many young Teda are engaging, leaving even Tuareg far behind them in a space where formerly the latter where the sole masters.

Fig 6: Map Tenere.

Fig 6: Map Tenere.

57Teda drivers consider the route from Agadez to the Kaouar as very easy. In fact, many vehicles use it, and the Nigerian army provides a weekly convoy between Agadez and Dirkou in order to protect travellers against attacks from bandits and terrorists. This is why numerous skid marks “draw” the route. These marks serve a traveller as line of position he only has to follow. In case the marks should not be visible, most of the young Teda drivers have their own cellphone-integrated GPS bought in Libya. Nevertheless, most Teda drivers participate in the convoy only until the well of Assheggur, as “controls around Dirkou are too numerous and that would become too expensive for us”. When they arrive at Assheggur, they head straight northeastwards to take a shortcut towards Sabha. Moussa again agreed to draw a map for me representing the track from Dirkou to Agadez. As he did previously when drawing the map from Bilma to Agadem, Moussa consulted his GPS in order to verify distances and he explained his drawing by the following text:

Here, this is Dirkou (1). When you leave Dirkou, you pass by Assheggur. After about 122 kilometres. It is here, Assheggur (2). When you pass by Assheggur, it is Dune 400 (3). Dune 400, this is a bit long, a bit dirty, with a house [a roofed cemented well]. […] Here there is a hill; its name is Kafran (4).
Here, there is a dune, which enters into the route (5). […] It is 52 kilometres from Dune 400.
You will leave. [Here is] the beginning of the hill of Agadez (6). The name [of the place here] is Augmenter Air, “the place on which you add air to the tyres of the vehicle”. One has to add air, because this is a stony ground. […] When you leave here, there is a bit zigzag (7), like this, and you enter at Kori-n-Kantana (8). At Kori-n-Kantana, you will continue to Turayt (9).
There [around Turayt] is bush, which has names, there are also wells, but I do not know their names. This is not important. On the route, there are herders. This is not interesting. I will only leave for Agadez (10) (566 km) […].

Fig 7: Mental Map B.

Fig 7: Mental Map B.

58Most of the places Moussa drew on the map and described in his text are no prominent landmarks. Rather they have important implications for safe and secure motorised travelling. Thus, the points called Dune 400 (Puits de l’espoir), Dune 52 or Augmenter air (“Add air [to the tires]”) show the presence of dunes to be surmounted or circumnavigated or the beginning of a rocky terrain, which needs well-filled tyres. The post of Turayt, the valley Kori-n-Kantana and the well of Assheggur are also not landmarks, but places to pass controls or to rest and to stay in contact with other drivers. Only the spike in the east, symbolizing the mount Kafran, can be fully considered as a landmark that is important for piloting. In fact, many vehicles from several directions (Seguedine, Dirkou, Bilma, Fachi) cross here, and the alignment of skid marks could be confusing. A prominent landmark, which is visible even when driving at high speed, can thus be useful for updating path information and for piloting.

59In order to establish a circle of position, driver’s measure distances by means of their vehicle’s odometer or their GPS. But even without technical instruments drivers can compute distances as they know the approximate speed of their car with which they can calculate distances by means of time spent travelling. Waypoints, which Moussa marked on the map, can also partly indicate distance, as already shown in the preceding cases (Manga, Erg of Bilma, Kaouar), and thereby a circle of position.

60At the end of his text, Moussa displays complete lack of interest towards places, wells and peoples alongside the track: “This is not interesting”. Such lack of interest is opposed to the importance Moussa attributed to wells (and implicitly to people) when driving in the Manga. In fact, between Agadez and Assheggur, a Teda is “not at home”. For several hundred years this has been Tuareg country, and young Teda drivers only recently appropriated the track for their lucrative business. The space thus remains “foreign” to the Teda travellers.

61It is only around Assheggur (a well that still has a Tamasheq name) that the Teda’s country really begins. Eli, (49 years; Mada clan), a school director from Bilma who has never worked as a guide or driver but who has traveled many times between Agadez and the Kaouar by camel and in vehicles, showed a very fine knowledge of ground patterns of this “Teda space”:

When one approaches Assheggur [from the west], one can observe depressions on the bottom of which there is lime, a white powder […] and there are shells too […]. When one comes even closer to Assheggur, one notes the reddish or brown colour of the ground, and small stones. […]
When one then leaves Assheggur [eastwards], uphills and downhills are beginning, similar to the depressions of the Manga. Then, when mounting and descending, mounting and descending, even if you don’t see anything, you can feel it, even if you are in a vehicle, according to the uphills and downhills, even if you close your eyes, you can know, if the place is familiar to you, you can state that you are there.
Some ten or so kilometres after, the ground becomes flatter, stones are appearing on the earth, and the earth becomes darker. […]. Stones everywhere. So you can recognize that you are on a ground where you drive more easily, even if your eyes are closed, according to the speed of the vehicle, you can recognize in what place you are.
Once at Kafran, you push forwards a slight uphill, so you know that you are at Kafran. And after Kafran […] the ground becomes softer, uphills and downhills appear again, uphills and downhills again, after that some stones.
[Then] again appear uphills and downhills, depressions and, when approaching Dirkou, you will see the vegetation at the foot of the escarpment, which is in all the region of the Kaouar, and so you can know that you are not far from Dirkou.

62In his description, Eli concentrates on ground patterns, which are not useful for piloting, but for verifying one’s position and for updating route information. Sentences such as “you can state that you are here”, “even if your eyes are closed, according to the vehicle’s speed, you recognize in which place you are” or “you can know that you are at Kafran”, show a process of orientation somehow similar to orientation by changing vegetation and ground patterns that I described in Manga’s case. Navigating and updating path information by the recognition of ground features is also used in other topographic and cultural contexts. For instance, hutchins describes how navigators on the Pacific use similar “ground”-marks: “Deep below, the presence of submerged reefs changes the apparent colour of the water. The surface of the sea undulates with swells born in distant weather systems, and the interaction of the swells with island produces distinctive swell patterns in the vicinity of land” (1996: 67; cf. also Gladwin 1970: 153, 168-69). Not least, contemporary marine navigation uses a mode of orientation, which is called “Depth-Contour Matching” and which is based on changes in patterns of the sea bottom (Hutchins 1996: 56).

63Eli’s attention to different grounds combined with the vehicle’s speed (“according to the speed of the vehicle, you can recognize in what place you are”) is of particular interest, as it computes a current position relative to speed/distance and direction. Later, when approaching the Kaouar, the escarpment furnishes landmarks for taking a bearing, and piloting takes over.

64Despite their excellent abilities in way-finding, even Teda can lose orientation. On the road from Agadez to Dirkou, our driver Ismael (about 30 years, Mada clan), took a wrong dune corridor, which narrowed more and more and then led to a dead end. He did not return immediately, but stopped, left the cabin and walked around in the sand for some time. I do not know if he was searching for some signs. Perhaps he tried to regain concentration by a “pause” similar to a procedure, which Chapelle (1982: 175) has already described: When a Teda guide loses his orientation, he stops, closes his eyes, turns several times around himself, lays down and waits that the dizziness leaves. Then he gets up and looks at the stars and the landscape “as if this was something new”. He thus recaptures orientation and continues.


65There is a debate as to whether people orientate in space by practical mastery or by mental mapping. As underlined by several authors (cf. Istomin and Dwyer 2009; Widlok 1997), both procedures seem to be involved. Probably route knowledge is more important when orientating in familiar spaces, whereas survey knowledge intervenes in less familiar or unknown ones. Nevertheless, as for the Teda, both procedures seem to be only of secondary importance in the orientation process. The basic means is dead reckoning, i.e. orientating by direction and distance. Dead reckoning as the Teda’s basic principle of orientation is used in both large-scale or less familiar (even unknown) environments such as the Tenere desert, and also in more familiar ones, such as the Manga. Two other procedures, piloting and updating, complete the way-finding process.

66Landscape features play an important role for way-finding. They are polyvalent. A mountain, for example, can allow piloting, but can also be used for updating the information of the travelled way. A less salient mark, such as a ground pattern, is primarily used for updating but can, if more prominent marks are lacking, also be used – of course to a much lesser degree – for piloting. The two constraints of dead reckoning – the line and the circle of position – can also be represented in different ways. In a more abstract one, the line is established by means of cardinal points. But landscape can also provide a line of position – for instance mountains constituting a north-south escarpment or vegetation and ground patterns representing a longitudinal line along which the traveller navigates. The same is true for the circle of position: it can be expressed by metric distances, but a traveller can also deduce distance from daytime or changing environments, as, for example, different vegetation and ground patterns. Again, one can state the multiple use of landscape patterns in the orientation process, which can shift from markers of a line of position to points establishing a circle of position and vice-versa. Chapelle ascribed to Teda the capacity of using “the finest references” for orientation (1982: 174). It seems, however, that more than their fineness, it is the polyvalence of such references which makes notable the particularity of the Teda’s orientation system, or even the combination of complex polyvalent landscape patterns with the simplicity of dead reckoning based only on a circle and a line.

Haut de page


Baroin Catherine, 1986, Organisation territoriale, organisation sociale : la logique du système toubou, Journal des africanistes 56 : 7-27.

Bourdieu Pierre, 1977, Outline of a Theory of Practice, Cambridge, Cambridge University Press.

Bryant David J., 1997, Representing Space in Language and Perception. Mind & Language 12 (3/4): 239-264.

Chaliand Gérard, 1981, Réponse à Robert Buijtenhuis. Guerre de guérilla et révolution en Afrique noire. Les leçons du Tchad, Politique africaine 1 : 23-33.

Chapelle Jean, 1982 [1958], Nomades noirs du Sahara, Paris, L’Harmattan.

Cline Walter, 1950, The Teda of Tibesti, Borku, and Kaouar in the Eastern Sahara, Menasha, Wisconsin.

Collett Matthew, Collett Thomas S., Wehner Rüdiger, 1999, Calibration of Vector Navigation in Desert Ants, Current Biology 9: 1031-1034.

Comité pour le développement de la langue tedaga, 2011, Dictionnaire multilingue. Tibesti.

Cyffer Norbert, 1994, English-Kanuri Dictionnary, Cologne, Rüdiger Köppe Verlag.

Cyffer Norbert, Hutchison John (eds.), 1990, Dictionary of the Kanuri Language, Dordrecht, Foris Publications.

Denis Michel, Zimmer, Hubert D., 1992, Analog Properties of Cognitive Maps Constructed from Verbal Description, Psychological Research/ Psychologische Forschung 54: 286-298.

Ferrandi (lieutenant J.), 1910, Les oasis et les nomades du Sahara oriental, Renseignements coloniaux 1 : 3-8, 38-46.

Foo Patrick, Warren William H., Duchon Andrew, Tarr Michael J., 2005, Do humans integrate routes into a cognitive map? Map-versus Landmark-Based Navigation of Novel Shortcuts, Journal of Experimental Psychology: Learning, Memory and Cognition 31 (2): 195-215.

Frisch Max von, 1965, Tanzsprache und Orientierung der Bienen, Berlin, Springer.

Fuchs Peter, 1989, Fachi. Sahara Stadt der Kanuri, Wiesbaden, Franz Steiner.

— 1983, Das Brot der Wüste. Sozio-Ökonomie der Sahara-Kanuri von Fachi, Wiesbaden, Franz Steiner.

Gallistel C.R., 1990, The Organization of Learning, Cambridge, MIT Press.

Gell Alfred, 1985, How to Read a Map: Remarks on the Practical Logic of Navigation, Man 20 (2): 271-286.

Gladwin Thomas, 1970, East is a Big Bird. Navigation and Logic on Puluwat Atoll. Cambridge, Harvard University Press.

Hutchins Edwin, 1996 (1995), Cognition in the Wild, Cambridge, MIT.

Ingold Tim, 2011 (2000), The Perception of the Environment. Essays on Livelihood, Dwelling and Skill, London, Routledge.

Istomin Kirill V., Dwyer Mark J., 2009, Finding the Way. A Critical Discussion of Anthropological Theories of Human Spatial Orientation with Reference to Reindeer Herders of Northeastern Europe and Western Siberia, Current Anthropology 50 (1): 29-49.

Le Cœur Charles, 1950, Dictionnaire ethnographique Téda, Paris, Larose.

Le Cœur Charles et Marguerite, 1956, Grammaire et textes Teda-Daza, Dakar, IFAN.

Le Sourd (lieutenant), 1946, Tarikh el Kaouar, Bulletin de l’Institut français d’Afrique noire, VIII (1-4) : 1-54.

Lewis David, 1976a, Observations on Route Finding and Spatial Orientation among the Aboriginal Peoples of the Western Desert Region of Central Australia, Oceania 46 (4): 249-282.

— 1976b. Route Finding by Desert Aborigines in Australia. The Journal of Navigation 29: 21-38.

Loomis Jack M., Klatzky Roberta L., Golledge Reginald G., Philbeck John W., 1999, Human Navigation by Path Integration, in Reginald G. Golledge (ed.), Wayfinding behavior. Cognitive mapping and other spatial processes, Baltimore, The John Hopkins University Press: 125-151.

MacDonald John, 1998, The Arctic Sky. Inuit Astronomy, Star Lore, and Legend, Toronto, Royal Ontario Museum/Nunavut Research Institute. mallot Hanspeter A., Gillner Sabine, 2000, Route Navigating without Place Recognition: What is Recognized in Recognition-Triggered Responses? Perception 29: 43-55.

McNaughton B.L., Chen L.L., Markus E.J. 1991, “Dead Reckoning”, Landmark Learning and the Sense of Direction: A Neurophysiological and Computational Hypothesis, Journal of Cognitive Neuroscience 3 (2): 190-202.

Müller M., Wehner R., 1988, Path Integration in Desert Ants, Cataglyphis fortis, Proceedings of the National Academy of Sciences of the USA 85: 5278-5290.

Ritter Hans, 1992, Der Karawanenweg nach Bilma bei S.W. Koelle und H. Barth in der Mitte des 19. Jahrhunderts: Ein historisches Beispiel zur Problematik der Informantenbefragung, Paideuma 38: 143-162.

Spittler Gerd, 1990, Führer und Karawane in der Wüste, in Hans Oswald (ed.), Macht und Recht. Festschrift für Heinrich Popitz zum 75, Geburtstag: 175-195.

— 1998, Hirtenarbeit. Die Welt der Kamelhirten und Ziegenhirtinnen von Timia, Cologne, Rüdiger Köppe Verlag.

— 1984, Karawanenhandel und Kamelrazzia bei den Kel Ewey. Die Kontrolle des Salzund Hirsehandels zwischen Air, Bilma und Kano (1850-1900), Paideuma 30: 139-160.

Wang Ranxiao F., Spelke Elizabeth S., 2002, Human Spatial Representation: Insights from Animals, Trends in Cognitive Sciences 6 (9): 376–382

Widlok Thomas, 1997, The Shared Cognition of Hai‖om Bushpeople, The Journal of the Royal Anthropological Institute 3 (2): 317-332.

Haut de page


1 Translation of French texts into English done by the author.

2 A hypothesis is proposed in which the integration process, an operation that may be difficult for neurons to implement, is replaced by a linear associative mapping, an operation that is at least theoretically easy to implement with neurons. The proposed system makes use of a set of linearly independent vectors representing the combination of the current head direction, and head angular velocity representations to ‘recall’ the resulting head direction. It is then proposed that visual landmarks become incorporated into the directional system, enabling both the correction of cumulative error and, ultimately, the computation of novel, optimal trajectories between locations” (McNaughton et al. 1991: 190).

3 Fieldwork and the present research were financed by the German Research Council (DFG).

4 Not to be confused with the measurement of time. “I stayed two days” is said, diske čû tudugu (diski= “day”).

5 Contemporary Teda seem not to use a large number of stars for orientation, as in the case of “star compasses” (cf. Gladwin 1970) or Inuit navigation systems (cf. MacDonald 1998).

6 This is a very general delimitation, and not everybody agrees with it. For some of my interlocutors, the most northern point of the Manga is Dibbela (Tedaga: Dubbula), for others Bilaberim (Tedaga: Bulabrim). The latter argue that trees becomes scarce or absent in the north of Bilaberim, “and this is why you call it ‘Sahara’ there”. When a Teda of the Kawar says “I am going to the Sahara” (Sahara tede-ê), he means that he will go to the northern Manga.

7 Texts from Teda informants translated by the author.

8 N.B.: Moussa described the way from north to south, whereas my description, after Moussa’s text, is of my own outward journey from south to north, which I travelled with several locals of the Teda Gunna clan.

9 The coordinates of these trees are N 15° 58.012‘, E 13° 12.084‘ for A. tortilis and N 16° 24.035’, E 13° 22.894’ for C. decidua.

10 We were in Bilma when Moussa drew the map, and thus his GPS automatically calculated the distances from here. The fact that Moussa considers it as important to rely on distances when describing the route is significant, even if these GPS distances are linear ones and thus can bias the “real” distance. In Moussa’s text, I added these distances in brackets. I also added numbers in parentheses to his text which correspond to stages on the mental map, in order to facilitate the simultaneous lecture of text and map.

11 For many Teda, “Kawar” means the region between Beza and Aney, which is covered by the Teda-canton of Dirkou. For Teda from Tibesti, the Kawar extends from Bilma to Seguedine and is called tiggey. For people of the Manga, the Kawar sometimes starts as soon as Agadem. This wide defi tries to denominate a space where the date palm grows and fructifi and where the harsh conditions of the surrounding desert do not exist. A country “where you can stay and live”, as one interlocutor named it. For administration, fi , the Kawar covers the whole department of Bilma.

12 Tubu call such places awi, a term which designates the absence of vegetation, water and people – a place hostile for humans.

Haut de page

Table des illustrations

Titre Fig 1: Map Overview
Fichier image/png, 344k
Titre Fig 2 : Map Manga
Fichier image/png, 252k
Titre Fig 3: Map Erg of Bilma.
Fichier image/png, 235k
Titre Fig 4: Mental Map A.
Fichier image/png, 96k
Titre Fig 5: Map Kaouar.
Fichier image/png, 213k
Titre Fig 6: Map Tenere.
Fichier image/png, 311k
Titre Fig 7: Mental Map B.
Fichier image/png, 188k
Haut de page

Pour citer cet article

Référence papier

Tilman Musch, « Six days towards the polar star: orientation among Tubu Teda », Journal des africanistes, 85-1/2 | 2015, 282-310.

Référence électronique

Tilman Musch, « Six days towards the polar star: orientation among Tubu Teda », Journal des africanistes [En ligne], 85-1/2 | 2015, mis en ligne le 06 juillet 2016, consulté le 24 juillet 2017. URL :

Haut de page


Tilman Musch

Ethnologue, Université de Bayreuth

Haut de page

Droits d’auteur

Société des africanistes

Haut de page
  • Logo Société des africanistes