The Oxford Research Encyclopedia of African History will be available via subscription on April 26. Visit About to learn more, meet the editorial board, learn about subscriber services.

Show Summary Details

Page of

PRINTED FROM the OXFORD RESEARCH ENCYCLOPEDIA, AFRICAN HISTORY ( (c) Oxford University Press USA, 2019. All Rights Reserved. Personal use only; commercial use is strictly prohibited (for details see Privacy Policy and Legal Notice).

date: 21 April 2019

Pottery chaînes opératoires as Historical Documents

Summary and Keywords

Technical processes—or chaînes opératoires—are heterogeneous cultural aggregates articulating raw materials, tools, knowledge, representations, and agents. The nature and arrangement of these elements stem from a web of social, historical, and ecological relations that not only delineate a sociohistorical framework within which artisans operate, but also determine how individuals shape and give meaning to their daily engagement in the craft. Pottery chaînes opératoires have been the focus of a large body of ethnographical studies in Africa since the beginning of the 20th century, most of them developing within the subfield of ethnoarchaeology. Yet pottery chaînes opératoires may also provide crucial information to historians when analyzed through an approach inspired from historical linguistics and tentatively called “comparative technology.” Pioneered by Haudricourt, this approach combines two levels of comparison. The first consists in a minute comparison of different chaînes opératoires within a given field of activity and geographical area in order to identify similarities and differences in tools, materials, gestures, and the organization of operations. This allows for the identification of specific “technical traditions”; that is, shared ways of doing that stem from a shared set of knowledge. The second level of comparison implies a mapping of the technical traditions (be it whole chaînes opératoires or particular stages or components), with the aim of identifying and characterizing their respective spatial distributions: for example, the effects of aggregation or disintegration, possible boundaries, or interpenetrations.

The relevance of spatial distributions in history-oriented analyses of technical processes is twofold. First, spatial distributions compel us to explore the sociohistorical processes from which they resulted; that is, the set of relations—social, economic, political, and ecological—that determine how artisans interact with each other, share knowledge, use tools and materials, cope with changing situations, or seize new opportunities. Second, spatial distributions may reveal strong and time-enduring connections with various kinds of social identities (e.g., languages, political factionalism, regional affiliation, gender, and ethnicity). When the underlying mechanisms of such connections are appropriately understood, they may be used to formulate hypotheses about past processes, including population movements, the development and evolution of political boundaries, identity negotiations, or socio-economical transformations. Here, chaînes opératoires may prove especially reliable for historians since they are less easily and deliberately manipulated than written or oral documents

Keywords: methods, sources, and historiography in African history, pottery, chaîne opératoire, technology, material culture history, spatial distributions, social and political boundaries, apprenticeship

What Pottery Chaînes Opératoires Are and Why They Matter

The notion of “chaîne opératoire” pertains both to the process of transforming a single (or several) raw material(s) into a finished product, and to the analytical tool used for documenting such a process.1 The latter aspect is critically important. Indeed, each technical endeavor may be envisioned as a sequence—or “chain”—of operations that progressively alter the state of a material. However, providing an accurate description of such chain of operations for comparative purposes is not just a matter of writing down what unfolds under the observer’s eyes: data collection must be systematized, hierarchized, and extended to aspects that are not necessarily visible. This is where the analytical dimension of the chaîne opératoire becomes extremely helpful.

Pottery chaînes opératoires

A pottery chaîne opératoire may include eight major operations, or phases (see Figure 1): (1) clay procurement, (2) clay processing, (3) shaping, (4) decoration, (5) drying, (6) prefiring, (7) firing, and (8) postfiring. Several of these phases can neither be canceled nor postponed and correspond broadly to what Lemonnier calls “strategic operations.”2 Other phases are not “strategic” inasmuch as their presence and position within the chaîne opératoire do not condition its outcome. For instance, there is generally no obligation to prefire the vessels before firing or treat them after firing.

Pottery chaînes opératoires as Historical DocumentsClick to view larger

Figure 1. The different phases and subphases of a pottery chaîne opératoire.

Illustration by O. Gosselain.

Whatever their status, each phase of the chaîne opératoire must be broken down into smaller units for comparative purpose: operations and sequences.3 The operation corresponds to an isolated action or sequence of actions and the sequence to an organized set of operations. In southwestern Niger, for example, the phase of clay processing may include several sequences such as drying, soaking, grog fabrication, and the mixing of grog and clay, that themselves involve several operations (see Figure 2): grinding sherds with an upper and lower grinding stone; winnowing the fragments with a basketwork for obtaining grog, kneading the mixture of clay and grog with both hands, etc. Comparisons between chaînes opératoires are usually made at the level of operations or sequences, the scale of analysis being determined by the degree of precision of the available data and the research agenda (see below). What matters is that comparisons be always made at the same analytical level.

Pottery chaînes opératoires as Historical DocumentsClick to view larger

Figure 2. Analytical distinction between “phase,” “sequence,” and “operation.” In this example, the clay processing phase includes a sequence of grog fabrication, which itself is subdivided into two operations: (1) grinding sherds on a lower grinding stone with an upper grinding stone, and (2) winnowing the crushed sherds with a circular mat in order to sort them into two sizes.

Illustration by O. Gosselain.

There is no place here to detail the sequences and operations involved in each phase of the pottery chaîne opératoire.4 Yet some precision is needed for the shaping phase, which has been the focus of several history-oriented analyses5 and will be discussed extensively below. This phase should indeed be divided into two subphases: (1) roughing out, and (2) preforming.6 During roughing out, potters transform a lump of clay or joint clay segments in order to constitute a hollow volume—the rough shape—whose resulting form does not correspond to that of the finished product. During preforming, potters give the hollow volume its final geometric characteristics through scraping and smoothing operations.

The roughing out substage sometimes proceeds in a single sequence (e.g., spiral-coiling from bottom to top) but is more often broken down into different sequences, according to the part of the vessel to be shaped. Thus, for example, a vessel’s lower body will be roughed out by molding a disk of clay on a convex mold, its shoulder by adding thick coils that are crushed internally and subsequently beaten with a wooden paddle, and its neck by adding a ring-shaped coil subsequently thinned by pinching (see Figure 3). Following the prevailing terminology in lithic studies, the combination of sequences is called a method, while the operations involved in each sequence are called techniques.7 Comparisons made at that level of the chaîne opératoire should ideally focus on methods rather than techniques, since the former are better markers of cultural differences than the latter as we will see below.

Pottery chaînes opératoires as Historical DocumentsClick to view larger

Figure 3. This common West African shaping technique combines molding (1–2), coiling (3, 5), beating (4), and pinching (6).

Photos by O. Gosselain.

Traditions versus Adaptations

Suppose that the comparison of pottery chaînes opératoires within a given region reveals, among other things, two distinct traditions in the processing of clay materials. Suppose also that these traditions cluster in different areas, among potters who respectably use distinct clay sources and clay deposits. How to interpret such technical variations? Could they correspond to sociohistorical constructions, independent from the physical and chemical characteristics of local raw materials, and hence be used as historical documents? Or do they stem from forced adaptations to constraints imposed by the materials (which strongly reduces their historical potential)?

Two decades ago, most scholars would probably have opted for the second alternative, thanks to the popularity of a series of publications that interpreted potters’ behavior in mainly technofunctional terms.8 The prevailing ideas at the time were (1) that pottery chaînes opératoires are fraught with dangers, which forces potters to find the right balance between available materials and processing techniques, and (2) that changes in techniques and materials aim primarily at improving the use performances of the vessels. This interpretative paradigm has been increasingly challenged by ethnographical and archaeological studies developing under the banner of “technological style”9 and “cultural technology”.10 Ethnographical inquiries notably demonstrated that even when potters interpret their techniques in strictly functional terms (as they generally do) and deem it impossible to change or alter them, they could actually rely on a large palette of alternatives. As an illustration, ethnographical fieldwork in Cameroon, combined with laboratory analyses, showed that different raw materials could be processed in exactly the same way or that dozens of processing recipes could theoretically be used for a given category of material.11 Such conclusions extend to other stages of the chaîne opératoire. For example, a systematic monitoring of firing conditions (temperatures, heating rate, and soaking time) in Africa and Asia shows that similar results are achieved with contrasting firing structures and fuels.12 In other words, it is unlikely that the diversity of technical practices observed at any level of the chaîne opératoire stems primarily from technofunctional constraints.

The preceding does not mean that potters are free from any constraints at all. Social, economic, political, or religious requirements notably bear on decision-making and contribute to shape behavior at both the individual and collective levels. First and foremost, potting practices are learned behavior, which means that they are inherited and generally reproduced under specific circumstances and within specific social networks. This “genealogy of practice” contributes not only to determine the overall nature of chaînes opératoires, but also to connect individuals spatially and intergenerationally through the sharing of specific technical traditions. Second, an array of factors may act directly or indirectly on the distribution and evolution of technical traditions, as will be discussed below.

Beyond Convergences

While the making of pottery vessels leaves theoretically much room for technical diversity, options are not infinite at certain stages of the chaîne opératoire and, depending on the analytical level, techniques may even appear monotonous. In sub-Saharan Africa, for example, there exist only eight broad categories of roughing out operations (e.g., coiling, molding, or drawing of a lump), seven categories of firing structures (from bonfire to updraft kiln), and a clear preference for grog and cereal husks as tempering materials.

The reasons beyond such lack of technical diversity probably exceed the set of natural evolutionary principles that Leroi-Gourhan had in mind when seeking to explain what he called “technical tendencies”;13 but whatever their origin, tendencies or convergences may hamper historical reasoning in comparative technology. To take an extreme example, how to interpret the fact that potters in South America, Papua New Guinea, and Central Africa use the coiling technique to rough out their vessels? Surely not by positing the existence of some sort of historical connection between these areas. But if such sharing of roughing out techniques concerned potters living one hundred kilometres apart in a region of Africa, wouldn’t it seem “natural” to envision a connection?

There are three possible ways to overcome the challenge posed by technical convergence in comparative analysis. The first is to confine the analysis to an area that can be explored exhaustively and where the genealogy of practice can be reasonably assessed. A generic technique that would not be worth attention in large-scale comparisons may indeed become locally relevant when connected to a specific learning network. In the Inner Niger Delta of Mali or the Lake Chad area, for example, consistent correspondences between roughing out techniques and learning networks developing within language or ethnic boundaries allow scholars to use techniques such as coiling, molding, or drawing of a lump as indexes of ethnic identities.14 Moreover, their identification in both historical and archaeological contexts allows exploring peopling dynamics and reconstructing ancient social or political boundaries.15

The second way to overcome the problem of technical convergence is to increase the level of details in comparative analyses. This either involves considering operations and sequences more thoroughly or combining different elements of the chaîne opératoire. For example, an in-depth comparison of ethnographic data collected in Southern Niger shows that the pounding technique, used for roughing out vessels in a large portion of the Sahel and previously envisioned as a single tradition, must actually be broken down into at least two distinct techniques: convergent pounding and divergent pounding (see Figure 4).16 Such distinction notably allows identifying the emergence and separate evolution of female and male pottery traditions in the central Sahel.17

Pottery chaînes opératoires as Historical DocumentsClick to view larger

Figure 4. Two variants in the shaping of pottery vessels by pounding a lump of clay on a concave form: 1–4 shows convergent pounding, in which potters start from the rim of the lump and progress toward its center; 5–8 shows divergent pounding, in which potters start from the center of the lump and progress toward its rim.

Photos by O. Gosselain.

Increasing the level of details may become imperative when researchers have specific questions in mind. For example, Corniquet analyzed the minutiae of collective firings in the Arewa region of Niger with a view to identify technical correlates of intracommunity affinities and antagonisms.18 She notably located them in the way in which potters broke and placed millet stalks (the main fuel) in the firing structure, or the order in which they stacked the vessels. Here, seemingly trivial differences were actually invested as bounding marks between participants; that is, as micro practices liable to be adopted or rejected by members of firing parties, depending on changing social relations. Another example concerns the roughing out and preforming of vessels among Ari potters of Ethiopia. Interested in the interplay between community-based and individual-based skills, Kaneko identified dozens of specific movements of fingers involved in the two shaping subphases, which were either shared at a community level or developed independently by potters in the course of practice.19 These technical idiosyncrasies (that may be likened to “signatures”) had an impact on the weight and thickness of the vessels and, consequently, on customers’ choices.

The third way to overcome the problem of technical convergence in comparative analyses is to combine elements of the chaîne opératoire that are not necessarily good sociohistorical markers in themselves but may become so when considered in association. In a region encompassing southern Burkina Faso, northern Togo, southern Niger, and northern Nigeria, for example, different communities of male (and sometimes female) potters with no linguistic or ethnic ties shape their vessels with the divergent pounding technique, decorate them with braided strip roulettes, and fire them in earthen ovens.20 Each of these elements appears independently in other communities and, with the possible exception of braided strip roulettes, can hardly be used alone for formulating historical hypotheses.21 Yet their association as a “technical bundle” is certainly not fortuitous, especially since many potters concerned are men (a rare occurrence in sub-Saharan Africa), indicating the possible existence of a distinct learning network.22

Markers of Identity

Relying on field inquiries and a large-scale comparison of data collected across sub-Saharan Africa, a tentative model has been proposed for explaining why some components of pottery chaînes opératoires tend to be associated with specific facets of social identities.23 The model sorts the phases of the manufacturing process according to their social salience and resistance to change, through comparing their degree of visibility on finished products and the type of skills involved.24 Three groupings are envisioned. The first includes preforming, decoration, prefiring, and postfiring, all phases (or a subphase in the case of preforming) whose characteristics are (1) to have an effect on the appearance of finished products (shape, decoration, and color), and (2) to involve unspecialized skills. Techniques associated with these phases could thus be easily modified after initial learning and socially, politically, or symbolically invested, which makes them potential markers of situational identities.

The second grouping of technical phases includes clay extraction, clay processing, and firing; three phases that also involve unspecialized skills but do no leave visible marks on the finished products—at least for untrained eyes. As potters often conduct such phases on a communal basis and usually follow locally shared norms, techniques associated with them may become markers of cooperation networks. A good example comes from the Arewa region of Niger, where a dozen clay sources are exploited by some thirty villages: field inquiries reveal that each group of villages related to a single clay source generally displays a distinct clay processing recipe and that such situations stem from potters’ interaction and exchanges of information at the extraction site.25

The third grouping only includes the shaping phase, and more precisely the roughing out subphase. With a reliance on specialized skills gradually acquired under the supervision of more skilled (and often socially related) potters, a resistance to change due to the rooting of motor habits, and an invisibility on finished products, techniques associated with this subphase would be reliable markers of the most stable and rooted facets of identity.26 Field observations show indeed that pottery-learning networks tend to develop within homogeneous social boundaries, with the consequence that potters and their apprentices usually are of similar gender, speak the same language, and share an ethnic, clanic, and socioprofessional affiliation. Since the cost of changing motor habits is supposedly high, while the benefits are low—socially at least— potters are thought to conserve their techniques when confronted with other traditions, even when integrating foreign communities.

While the model described fits with many situations observed throughout Africa, some of its underlying assumptions are problematic. For example, explanatory mechanisms do not pay attention to the agency of potters or the sociohistorical contexts within which practice and identity are produced.27 The connection between technical actions and social identities seems indeed to develop outside the conscious experience of actors, whose historical role is limited to the transmission and reproduction of practice. Recognizing the problem, Gokee has reformulated the model, drawing on the semiotics of C. S. Peirce and a conception of pottery vessels as “constellations of possible significations” anchored in their perceptive qualities.28 In order to accentuate their identities or render them ambiguous, people would either reproduce or transform qualities that have become meaningful indexes of identity, such potential being related both to the context in which steps of the chaîne opératoire are conducted (from public performance to the intimacy of domestic workshops) and the traces left on the finished products. Sticking with the threefold framework described, Gokee thus opposes stages such as shaping, unlikely to be exploited in public contexts of identity negotiation (and thus liable to be an index of shared ancestry), to stages such as decoration, liable to “contribute substantially to the semiotic potency of the pot”.29 A good illustration of the latter appears in Guèye’s study of Haalpularen potters of the Middle Senegal Valley. She found that in a context of marked social hierarchies between artisans, the articulation of matrimonial and consumption networks contributed to the use of decorative attributes that herald the social identity of both potters and their clients.30

Gokee’s reformulation is a step in the right direction, but it does not resolve all the problems. For instance, neither model appropriately accounts for the numerous examples where no clear relationships can be established between specific technical behavior and salient expressions of identity, such as ethnic affiliation or language,31 or, even more disturbingly, where supposedly stable and bodily embedded practices such as shaping techniques are swiftly borrowed or transformed.32 Moreover, while pottery chaînes opératoires are “technical aggregates” whose different components may inform us about various facets of identity, the focus remains both on a single component—shaping—and a particular facet of identity—ethnicity—which is old hat for studies devoted to past and present expressions of identity in Africa.33 In-depth ethnographic inquiries show that crucial processes of identity construction and negotiation may also develop irrespectively of ethnic or linguistic boundaries, and that they may concern any component of the chaîne opératoire, independently of its post-manufacturing visibility or the nature of the skills involved.34

Techniques and Identity in Relational Perspective

Following Lave, it might be more fruitful to explore the interplay of techniques and identity through the prism of the relations that shape the “social world of activity”; that is, relations between persons acting and between the social and material worlds.35 These relations do not merely delineate a sociohistorical framework within which artisans operate, but also determine how artisans organize and give meaning to their daily engagement in the craft: how they interact with each other, share knowledge, use tools and materials, cope with changing situations, or seize new opportunities. This is what Lave calls the circumstances of practice.36 Thus, instead of striving to sort the phases of the chaîne opératoire according to their nature and capacity to materialize specific identities, attention should shift toward the types of relations involved in knowledge acquisition and meaning attributions, as both processes weight on the evolution of pottery traditions.

Knowledge Acquisition

Three types of relations may be envisioned, depending on their ability to ensure a reproduction of practice. A first category pertains to direct and sustained interactions, which are typically associated with communities of practice within which artisans engage mutually in a given activity.37 Although apprentices usually learn the full pottery-making repertoire in such contexts, a long-term face-to-face interaction with experienced individuals is mandatory for mastering specialized body techniques, such as those involved in shaping techniques, but also tool handling or the sensorial appreciation of materials (e.g., selecting appropriate clays according to their odor, taste, and texture). In other words, it is only in the context of direct and sustained interactions that the most skilled parts of a pottery repertoire can be handed down from one individual to another.

A second category of relations involved in knowledge acquisition pertains to casual interactions, which develop in shared practice settings (such as clay sources, firing places, or market places) or during temporary sojourns in foreign communities. In such situations, members of distinct communities of practice have the opportunity to witness other ways of doing and exchange information about the nature and location of raw materials, processing recipes, or the tools used at different levels of the chaîne opératoire. Even if these information exchanges take place outside the practice contexts concerned, they may engender technical adjustments or contribute to the development of “dormant repertoires” from which potters may subsequently draw. A telling example comes from southwestern Niger, where potters from distinct communities who met at a weekly marketplace near the capital, Niamey, swiftly modified their clay-processing recipes—stopping to sort grog after grinding—under the influence of a handful of women originating from the most renowned pottery village of the area.38 The reputation of the latter, combined with the reduction of work brought forward by the new processing recipe, ensured that mere discussions at a marketplace could engender a microregional change in technique. It must be stressed that such knowledge transfer requires a technical expertise sufficient enough to allow involved parties to make sense of what is discussed, pondering its potential interest and putting it successfully into practice. Casual interactions thus practically impact portions of existing chaînes opératoires rather than contributing to their reproduction.

A third category of relations involved in knowledge acquisition pertains to mediated interactions, which correspond to situations in which knowledge is acquired through the manipulation of objects produced outside contexts of participation. Such interactions notably allow for a reproduction of formal attributes or ornamental motives, but force potters who lack information about original ways of doing to draw from their own technical repertoire. For example, some northern Cameroonian potters use a stick for reproducing decorations made with a wooden carved roulette and deem it impossible to do otherwise. Similarly, in the Zarmaganda region of Niger, potters who want to imitate the polychrome water jars brought by seasonal migrants or pedlars ignore that the finer motifs are always painted with an iron blade and use a stick instead.39 In this case, the use of different painting tools not only demarcates communities but also interaction modes and learning genealogies.

Meaning Attribution

While closely connected to the contexts of knowledge acquisition, meaning attribution implies a broader set of relations and is liable to evolve throughout the life of an individual or a community. It also concerns a wider range of preoccupations than negotiations of identity.40 Part of this attribution derives from the work of “imagination,”41 through which participants rely on their direct and vicarious knowledge of the world to broaden their perception of practice (e.g., by envisioning historical continuities, connections, or disconnections with other individuals or communities, or possibilities for change). In contexts of direct and sustained interactions, for example, apprentices are compelled to adopt the whole technical repertoire that prevails in the community of practice they want to be part of. At that level, social conformism, obedience, and a belief in technical adequateness are the main factors in meaning attribution.

Yet as individuals subsequently redefine their position within a community of practice or face changing circumstances (e.g., when relocating after marriages or migrations, when frequenting new marketplaces, or when faced with new commercial demands or opportunities), they may be drawn to alter parts of their repertoire. In the case evoked where potters from different communities adopted a new clay-processing recipe under the influence of members of a renowned pottery center, casual interactions in the marketplace only created a possibility for change. Change did occur because potters related positively to those who brought the innovation (the “best potters” in the area) and to the innovation itself (reducing the time and energy devoted to clay processing). Without such work of imagination, the new processing recipe would have been either ignored or integrated in a dormant repertoire of practices.

Another process of meaning attribution is the work of “alignment,”42 through which participants coordinate their actions with members of other communities without necessarily engaging with or even knowing them. This form of belonging not only expands repertoires beyond the confines of communities of practice but contributes as well to make them fit within broader structures.43 Regarding the frequent association of shaping techniques with deeper facets of identity, pottery learning networks tend to develop within homogeneous social boundaries, with the consequence that the type of interactions liable to ensure a reproduction of the most skilled parts of the chaîne opératoire usually bring together individuals of similar gender, language, ethnicity, and socioprofessional affiliation. Yet, rather than motor habits, it is the meaning attached to shaping techniques that explains why artisan are less likely to modify them during their lifetime. Mastered less casually than other steps of the chaîne opératoire and under the supervision of a socially related person, the shaping stage is indeed often perceived as both the heart of pottery making and a family inheritance, one of the things transmitted by parents and ancestors in general that becomes constitutive of a person’s identity. The stability of fashioning techniques thus often results from a deliberate conservatism of the potters, especially in contexts where pottery making is a secondary activity, with few considerations liable to transform a potter’s relationship with her or his techniques (e.g., speed and ease of manufacture, standardization, and morphological flexibility).44 Technical conservatism, combined with the social and spatial logics that weigh on the development of learning networks, generates an intergenerational and intercommunity alignment of shaping techniques that ultimately contribute to their connection with macroscale social boundaries such as language or ethnic affiliation.

Of course, the same process of meaning attribution that gives rise to technical conservatism may lead to a shift in techniques when identities collide or are reformulated. In southcentral Niger, for example, Tuareg and Hausa communities have coexisted for the past two centuries. Among Tuaregs, artisans belong to endogamous subgroups whose members bear a low social status and may even be referred to as “slaves,” while among Hausa, craft activities are theoretically open to anyone and it is wealth rather than birth status that determines social hierarchies. Many artisans of Tuareg origin have consequently adopted a Hausa identity, an ongoing process even in the early 21st century. For potters, this not only meant adopting the Hausa language, dress code, or architecture, but also shifting from the diverging pounding technique (locally associated with Tuareg potters) to the molding technique, perceived as “truly Hausa.” All that remains from their former identity are the names given to some of the shaping tools. Thus, if geographical propinquity and intermarriages created appropriate conditions for the shift to take place, it is meaning attribution that actually led former Tuareg potters to align with a new network of potting communities.

Documenting and Contextualizing Pottery chaînes opératoires

The preceding illustrates the many ways in which pottery chaînes opératoires connect with sociohistorical processes and may, in turn, serve as starting points for exploring various aspects of the past. Emphasis has been placed on the theoretical foundations of an interpretative framework. Yet sound interpretations cannot be achieved without good quality data, which pertains both to their degree of detail and systematicity. This is where a chaîne opératoire approach—understood here as an analytical tool—becomes crucial.

The chaîne opératoire Template

Each phase of the chaîne opératoire must be documented by means of a multiple-entry template corresponding to a “checklist.” This template does not only ensure the systematicity of the collected data (which is a prerequisite of any good comparison), but also helps in acquiring a better understanding of technical logics and the factors affecting the spatial and temporal evolution of technical behaviors. The main entries of this template are summarized here.


While most phases of a pottery chaîne opératoire may take place in specialized infrastructures such as workshops or factories, they are often carried out in distinct practice settings. Information about the location and characteristics of such settings provides clues about the organization of the craft, social interactions, and relations with the environment. They include:

  • situation vis-à-vis dwellings and other areas of activity

  • topography (size, physical features liable to be exploited, existing structures, etc.)

  • ownership (individual or collective; associated with a population or a political power)

  • possible sharing (with whom? occasionally or regularly?)

  • relocation capacities (reasons, other sites envisioned, frequency, assessment of adequacy, etc.)

Here, the idea is to differentiate specialized sites from sites used on the spot, whose choice frequently depends on concerns other than pottery making.


Beside master potters, who usually occupy the center stage in practice settings and ethnographic descriptions, other actors may be implicated in technical operations whose commitment to the craft is highly variable.45 Each one should be properly identified, including:

  • main identity (gender, age, language, claimed ethnicity)

  • socioprofessional status (possible membership in a specialist group;46 economic importance of the income derived from the activity)

  • social relations with other actors engaged in the activity (types of relationships; possible contracts; equality, subordination, or authority)

  • role (central or peripheral; occasional or regular involvement)

  • level of competence (position in the process of apprenticeship)

Knowledge and Know-How

Knowledge and know-how are implicitly contained in technical actions: whether they concern the selection of raw materials, the use of tools, sequences of gestures, or any other transforming operations, they all draw on an expertise that takes form and becomes perceptible in the course of practice. Documenting practice is thus the primary way of identifying knowledge and know-how (especially since know-how often remains nonverbalized). However, historical investigations may also benefit from information about the genealogy of practice. Inquiries will thus aim at determining the following:

  • origin of knowledge (acquired from who? how? where? when?)

  • degree of specialization (is knowledge shared or not outside the domain of pottery making?)

  • existence of alternative knowledge or “dormant repertoires” of knowledge and know-how which have been acquired but are not put into practice (acquired where and from whom? If occasionally used: why and in which circumstances?)

Raw Materials

All materials used throughout the chaîne opératoire must be characterized according to the following aspects:

  • physical characteristics (type, composition, condition)

  • origin (natural or human; raw state, by-product of another activity, or recycling; site and conditions of acquisition [see “Location”])

  • modes of selection (types of criteria [including sensory assessments] and justifications; tolerance for variation; emic definitions of “appropriate” and “inappropriate” materials)

  • alternative choices (emic and etic identification of other potentially usable materials)

This last aspect is crucial, for it is in determining the actual latitude of choice in materials (as related both to personal knowledge and environmental potentials) that one may eliminate cases of mere convergence and start investigating the underlying logics of technical actions.


Actions combine gestures—or “elementary actions upon matter”47—and energies (of human, animal, or vegetal origins). Besides an accurate description, field observations should also aim at characterizing their mode of organization and their purpose.48


Besides information pertaining to the physical characteristics and modes of use of tools,49 inquiries should also provide biographical and contextual elements as follows:

  • structure (material, shape, absolute and relative dimensions, weight, balance, method of hafting, etc.)

  • functioning mode (the way in which the tool is used and the resulting effect on the material worked)

  • function(s) (range of purposes of the tool’s use)

  • origin (manufacturer, place of manufacture, previous users)

  • status (specialized or nonspecialized use, personal or shared)

Relationships with Other Activities

This is a crucial aspect of the analysis, since envisioning relationships with other elements of the broader sociotechnical system often helps in identifying historical, economic, or symbolic logics.50 Relationships may not only concern the elements already listed, but also the ways in which actions or finished products are conceived. Information notably involves:

  • actors (possible associations through status [e.g., caste] or activity (e.g., blacksmith and circumciser; potter and midwife)

  • location (other uses of the sites where technical actions are carried out; primary or secondary use)

  • raw materials (possible uses in other activities)

  • tools and actions (potential borrowing from other spheres of activity; similarity of bodily postures and functions51)

  • knowledge and know-how (transfer from one activity to another)


Two aspects should be considered: the position and status of the stage under study within the whole chaîne opératoire, and its position within the technical system to which the actors belong, with a focus on interrelated temporalities and the contingencies of the technical milieu.52 One will notably seek whether the stage under study is mandatory or elective, or whether it can be delayed or not. The stage will also be situated within the calendric cycle of the actors (seasonality, subordination to other activities) as well as the natural cycle (availability of certain materials, accessibility of specific locations, etc.).

Beliefs and Religious Practices

As pottery making centers on the transformation of natural materials into a cultural product, it constitutes an efficient metaphor for explaining natural processes of transformation and structuring cultural ones. The chaîne opératoire is consequently the locus of a dense web of symbols and religious beliefs whose most visible manifestations are the taboos and rituals attached to technical actions.53 Inquiries should aim at identifying their

  • nature

  • scope (persons and steps of the chaîne opératoire concerned)

  • temporality (onset and duration)

  • purpose (in regard to the technical actions, the products, the artisan, the “natural order,” etc.)

Specialized Vocabulary

A potter’s technical repertoire does not only include know-how, recipes, tools, or attitudes toward materials, but also a vocabulary whose components usually range from the specialized (i.e., words only used and understood by people familiar to the craft) to the mundane. The “words and things” method54 used in historical linguistics may be applied to such vocabulary, provided that it is systematically and carefully collected.55 A basic collection should include generic terms such as “pottery vessel,” “to make pottery,” “to ornate a pot,” “to fire pottery,” as well as the names of tools, materials, and finished products. More detailed studies may aim at collecting words for particular actions and gestures, or the transitory states of the materials.

Potters’ Biographies

The multientry framework as described explicitly aims at contextualizing the various components of a pottery chaîne opératoire and, when possible, retracing their trajectory. In regard to knowledge and know-how, for example, a part of the investigation seeks to reconstruct the genealogy of practice: from whom it was acquired, where, in which circumstances, etc. The same applies to specialized tools or infrastructures (ovens or kilns) whose origin and former owners must be systematically identified. Such a biographical approach helps situating individual practices within a broader web of sociohistorical relations and provides a first indication of the social and spatial embeddedness of learning networks.56 Yet other biographical elements should also be collected.

As the process of meaning attribution, and especially the combined work of “imagination” and “alignment,”57 stem from the multiple relations developed by a potter throughout her or his lifetime, it is crucial to document his or her “social space.” Shaped by historical, social, and ecological circumstances, the social space comprises the space of experience and the space known. The first corresponds to places frequented (and thus “experienced”) by individuals through daily activities, social interactions, economic exchanges, or travels, and around which a person’s sense of identity and belonging develops together with practical knowledge and representations. The second category of space is representational. It concerns all the places that a person knows vicariously; for example, from kin, friends, foreigners, or even the media. Second-hand knowledge generally reinforces a person’s sense of belonging, but also enriches her or his cognitive repertoire.

Doing History with Pottery chaînes opératoires

There are different ways in which information drawn from an analysis of pottery chaînes opératoires may be used in historically oriented studies. The most common pertains to the construction of interpretative models to be subsequently applied to archaeological contexts.58 This classical ethnoarchaeological pursuit requires a careful protocol in order to build strong archaeological inferences.59 Another possible application is the “direct historical approach,”60 in which modern pottery traditions are used as a baseline against which to compare temporally controlled archaeological assemblages.61 Besides historical inferences drawn from a diachronic comparison of pottery techniques, a crucial contribution of the direct historical approach lies in its capacity to identify long-term changes in pottery production and consumption and, relatedly, to demonstrate how sensitive techniques may be to broad contextual transformations (contrarily to the pessimistic conclusions of Adams regarding pottery decoration).62

Still another application is the comparative technology approach. Inspired by comparative linguistics, this approach rests on a comparison of chaînes opératoires within a given geographical area, a mapping of the various technical traditions, and an examination of the resulting spatial distributions in order to make historical inferences. Such a research program follows several steps whose characteristics and main requirements are successively detailed.

Defining a Comparative Framework

Before launching into a comparative analysis of pottery chaînes opératoires, the first imperative is to choose the right analytical framework. Does the analysis aim at examining the history of a particular population or set of populations? Do the research questions concern a particular region or a whole portion of the African continent? Do they concern specific components of pottery chaînes opératoires? Such questions should be carefully pondered, since comparing chaînes opératoires not only aims at determining a “technical content” and degree of technical homogeneity or heterogeneity within a given social or spatial unit, but also aims at locating meaningful interruptions in spatial distributions and evaluating their historical significance. This means considering both the area where an element is attested and the area where it is not, paying specific attention to where exactly the distribution stops.

Consider the following example. A comparison of potting practices in the Kadiolo region of Mali (near the border with the Ivory Coast) led Frank to identify two shaping (i.e., “roughing out”) techniques among Mande-speaking potters: molding and drawing of a lump.63 Relying on potters’ biographies, patronymic names, and the local history of potting communities, she interpreted the former technique as “Mande” and the second as a “foreign” tradition introduced by women who were probably taken as slaves in neighboring populations and locally paired with craft specialists other than blacksmiths (to whom Mande potters are often associated). These women would have diffused the drawing of a lump technique in certain parts of the Mande heartland, thus introducing technical diversity in a formerly homogeneous zone.

However, a comparison of shaping techniques used within the whole Mande language area reveals that the drawing of a lump technique is found throughout the zone, but tends to be especially associated with the eastern and southwestern Mande subgroups.64 According to linguists, these subgroups have been separated from the Mande “core” for the longest time period and, as far as southwestern Mande speakers are concerned, have subsequently developed few contacts with northern Mande.65 Among northern Mande speakers, the molding technique dominates the picture and displays a spatial distribution that coarsely corresponds to the historical boundaries of the Mali Empire, as that of the linguistic Mande subgroup to which it is associated.66 Considering, on the one hand, that most of the potters using the molding technique belong to caste-like socioprofessional groupings—unlike many of those using the drawing of a lump technique—and, on the other hand, that the development of the caste system in the Mande area could be connected to the emergence of the Mali Empire, the drawing of a lump technique may be interpreted as a Mande “proto-technique” (to paraphrase linguists) liable to have been progressively replaced by the molding technique from the beginning of the second millennium ad.67 Depending on whether one focuses on a specific Mande region or on the whole Mande language area, the same shaping technique is thus perceived as “exogenous” or “endogenous,” and its historical and spatial trajectory interpreted along very different lines.

Note that it is not always mandatory to launch into regional or macroregional comparisons to obtain pertinent results. In his study of Dogon pottery traditions, for example, Gallay compared data within a 100 km radius and demonstrated the existence of a “central tradition,” of ancient origin, and several “peripheral traditions” whose occurrences resulted from processes of acculturation or assimilation.68 Since Gallay’s goal was mainly to explore the complex history of the Dogon population through comparative technology, such a narrow focus was perfectly justified.

However, when the comparison aims primarily at assessing the sociohistorical significance of an element liable to be widely shared—such as a generic technique or a tool—and at reconstructing its trajectory through time and space, the geographical focus should definitely be widened so as to provide sufficient analytical hindsight. Such is the case of the roulette decoration techniques,69 whose spatial distribution only becomes interpretable when considered at a continental level.70 While micro regional comparisons led some archaeologists to view the sudden apparition of rouletting tools as an indication of population movements, a broader comparison shows that such tools mainly diffused through contacts between neighboring populations, without implying any migration process (more below).71

Selecting and Organizing the Data

While all components of a pottery chaîne opératoire are liable to yield historical information about those who inherit and put them into practice, they are likely to tell contrasting stories, depending on the kind of relations required to ensure their intergenerational or interindividual reproduction. For instance, the spatial distribution of a specialized body technique involved in clay selection, shaping, decoration, or postfiring is likely to stem from a (possibly long) chain of direct and sustained interactions. Conversely, the distribution of an unspecialized ornamental tool (e.g., a corn cob) or of a clay processing or painting recipe may stem from looser interactions and correspond to temporary or more lasting alignments between unrelated people or communities.

If the transmission modalities of knowledge and know-how can be grossly modeled, they cannot predict by themselves how the various components of pottery chaînes opératoires will concretely evolve in a given sociohistorical context. Meaning attribution may indeed create powerful filters that alter the “normal” (i.e., “mechanical”) course of their trajectory through time and space. It is thus always safer to compare as many components of a chaîne opératoire as possible and to avoid, from the outset, focusing on those that supposedly provide information on contrasting sociohistorical phenomena (i.e., the usual opposition between the shaping and decoration stages).

That being said, and irrespective of the nature of the components considered, some selection and organization of the data must often be undertaken. To start with, an isolated occurrence is rarely useful. Say, for example, that in the area under study a single potter polishes the mineral slip applied before firing with a wristband of plastic beads while all others use a string of baobab seeds for the same purpose: it will be useless to include polishing tools in the comparison since the spatial distribution of the variants identified will not provide any more information than that gathered in the field through interviews. One should therefore prioritize those components of the chaîne opératoire that display reoccurring rather than isolated variants.

Also, it may be useful to organize the data in such a way as to mitigate the effect of technical convergence or prevalence. At the level of clay processing, for example, common tempering materials such as grog, cereal husk, or dung often prove uninformative when considered independently because their spatial distribution partially or fully exceeds that of the geographical framework of comparison. It is consequently more judicious to compare processing practices as a “package” at that level of the chaîne opératoire; that is, to consider simultaneously the various tempering material that a potter may combine as well as all other possible modifications to which the clay is subjected. Each distinct combination of elements will then be considered as an independent recipe (or “tradition”) in the ensuing comparison.

Similarly, roughing out techniques may prove to be poor cultural markers when considered in generic terms (i.e., coiling, drawing of a lump, molding, etc.). A way of overcoming such problems has been to consider them in more detail: for example, distinguishing “convergent pounding” from “divergent pounding” or singling out the numerous variants of coiling (e.g., spiral superposition, ring superposition, and internal or external crushing). Another way of increasing the meaningfulness of roughing out techniques in comparative analyses is to consider their possible arrangement within the roughing out substage; in other words, to shift from “technique” to “method” (see above).

Technical Continuums, Discrepancies, and Filiations

Keeping in mind that one of the main goals of comparative technology is to find technical boundaries and confront them with other kinds of boundaries, the identification of discrete technical traditions constitutes a crucial operation. Part of its success depends on the degree of details and arrangement of the data (see the section “Selecting and Organizing the Data”). Yet some further problems may hamper the identification and preliminary interpretations of technical boundaries. One of them is the possible existence of technical continuums. Although such a phenomenon has not been investigated thoroughly in African ethnographic contexts, it is theoretically conceivable that a given technique be slightly modified when handed down from one individual to the next, with the result that variants observed at both ends of a learning continuum appear as independent traditions.72 Such a phenomenon does not so much correspond to a “standard deviation” in technical practice (to use a mathematical metaphor) as to correspond to the emergence of a new technical tradition through the gradual transformation of a former one. Contrasting historical interpretations may thus be expected, depending on whether the whole technical continuum or only a portion of it is considered.

This brings forward the related problem of sorting out technical discrepancy from technical filiation. In many instances, different practices can be safely considered as unrelated traditions. This is the case, for example, when some potters use the coiling technique for shaping the lower part of the vessel, while others use the drawing of a lump technique. Not only do the actions diverge, but their intrinsic logics do as well, which allows considering them as respectively independent from one another. However, a relation or even a filiation between different technical traditions may be envisioned in certain circumstances. In the case of the Mande shaping techniques previously evoked, one of the elements that led the researchers to postulate a gradual replacement of the drawing of a lump technique by the molding technique is the fact that the latter is combined to coiling and scraping (see Figure 5) throughout the Mande language area, while it is often combined to coiling and beating (see Figure 3) in other parts of West Africa. Since coiling and scraping are typically associated with the drawing of a lump technique (both inside and outside the Mande area), their local association with molding was interpreted as a technical persistence;73 a material index of the historical encounter between the drawing of a lump and the molding techniques.

Pottery chaînes opératoires as Historical DocumentsClick to view larger

Figure 5. The upper part of a vessel whose lower body was molded on an upturned vessel is shaped by adding coils (1) and scraping them with fingers (2–3) and a piece of calabash (4).

Photos by O. Gosselain.

A similar process may be at work in the southeastern Bantu area, where a shaping technique consisting of the drawing of superimposed rings of clay could well result from a local combination of the coiling and the drawing of a single ring technique.74 Here, the interpretation rests both on the spatial distribution of the three shaping traditions (the drawing of superimposed rings appears broadly in the meeting area of its putative parent techniques) and a consideration of their respective technical logics or “grammars” (joining of the clay segments, forming the bottom, scraping the walls, etc.).

This notion of technical grammar may be illustrated by another example. In southern Cameroon, most potters ornate their vessels by rolling a knotted strip roulette on the wet surface.75 In the early 1990s, some were also using hair rollers for the same purpose: while both tools are clearly unrelated, their functioning mode and function are strictly identical.76 The potters thus keep on with their technical grammar while altering a part of their “technical vocabulary.”

Mapping the Data

Having defined the framework of comparison and selected the elements to be compared, the next step is to map them. Ideally, all information should be geo-referenced beforehand in order to increase the accuracy of the comparison. Yet this is not always possible, especially when using secondhand information from archives or the ethnographic literature. In such cases, one may either attempt to locate the sites of inquiries more precisely or position them in the vicinity of identifiable localities or landscape features.77

As discussed, various components of a pottery chaînes opératoire should ideally be taken into consideration in the course of comparison. But while an appraisal of their respective spatial distribution is part of the analysis, they should never be brought together on the same map, since too large an amount of information considerably hampers the reading and interpretation of spatial distributions. The easiest way is to make a map for each technical component taken into consideration.78

On the maps, the points or symbols corresponding to the occurrences of specific technical elements may be distributed randomly, dispersed (more or less regularly), or aggregated (in which case they are considered “discrete distributions”). Rarely revelatory on their own, such patterns of distributions may nevertheless open the way to interpretations or at least channel the research questions. Suppose, for example, that the spatial distribution of a particular tool or practice takes the shape of two microregional clusters between which other variants occur. Questions to be addressed will include whether some connections are conceivable between the two clusters, and if so, why they are currently separated. As with linguistic geography, the spatial distribution of technical elements sometimes provides clues about the existence and articulation of successive strata of technical traditions.79

Identifying and Interpreting Spatial Dependencies

The core of the interpretation in comparative technology consists in identifying processes of spatial dependency; that is, the possible overlap between the distribution of a technical feature and that of a wide array of related or unrelated elements (see Figure 6).80 These notably include structuring elements in the landscape: lines of communication, physical barriers, ecological contrasts, variations in the density of human settlement, the presence of urban centers or of markets, etc. For example, a minute comparison of clay processing recipes in northern Cameroon revealed four technical clusters corresponding, respectively, to areas of intense social relations (between people that do not necessarily share the same social identity) and separated by zones of low settling density (and thus lower interactions).81 A plausible explanation is that unpeopled areas created a rupture in social interactions networks and, consequently, in the circulation of knowledge pertaining to clay processing.82

Pottery chaînes opératoires as Historical DocumentsClick to view larger

Figure 6. Identifying possible correlations between the spatial distribution of technical practices and that of the claimed identity of potters in southern Niger: (a) incising motives with a comb made of thorns inserted in a clay pellet; (b) adding natron to the mineral slip; (c) Kanuri (Manga and Mober); (d) Hausa “Beri Beri” (i.e., Hausa-ized Kanuri). Both pottery techniques are clearly associated with the Kanuri sphere in an area that corresponds broadly to a former province of the Kanem-Bornu empire. Their presence outside this area is partially explained by the migration of individual potters.

Illustration by O. Gosselain and A. Livingstone Smith.

Another form of dependency concerns social boundaries: languages, ethnic and geographical affiliations, social status, socioprofessional affiliations, gender, age groups, etc. Such elements are the focus of most comparative analyses of pottery chaînes opératoires. For example, the authors of a comparison of firing techniques in southeastern Botswana state that their “research question is whether differences in the technique of firing ceramic vessels reflect the different language groups to which the potters belong. Alternatively, or additionally, do they reflect other social groupings of potters?”83 Examining both the techniques actually used by potters and those to be used “under ideal circumstances” (which is a reminder of the importance of documenting “dormant repertoires” during field inquiries) and relying on statistical analysis, they conclude that firing techniques do not relate to language groups, gender, or belief systems (as defined by the observance or nonobservance of rituals and taboos) and are only weakly correlated with age groups and learning networks. The determinant factor seems to be the geographical location of the potters. The authors follow the same analytical protocol for other components of the chaîne opératoire, with contrasting results.84

Meaningful historical results do not necessarily entail a matching between technical behavior and specific facets of social identity. For example, a continental comparison of the roulette decoration technique reveals that the two broad categories of tools used in Africa—fiber roulettes and carved roulettes—are distributed in a continuous way over a large portion of the continent, irrespective of any social boundary.85 Such a pattern of distribution clearly stems from a peer-to-peer diffusion process in which casual interactions between potters ensured the gradual propagation of rouletting tools. A subsequent examination of archaeological data confirms this scenario and opens the way to other historical implications.86 For instance, the fact that only a limited portion of Bantu-speaking potters use rouletting tools in the early 21st century means that their remote ancestors did not use them when living in northwestern Cameroon, the place of origin of Bantu speakers: if not, rouletting tools would be found throughout the Bantu language area. It also means that if the Bantu expansion involved population movements, these movements must have occurred before the arrival of rouletting tools in northeastern Cameroon and adjacent areas; that is, before 2500–2000 bp.

In other circumstances, an absence of spatial dependency between technical traditions and social boundaries may result from a process of social assimilation or mixing—as in the case of the Mande and Dogon potters87—from individual movements in space,88 or from more or less large borrowing processes.89 All these situations are obviously interesting from a historical point of view.

Whether spatial dependencies are identified or not, the last, and more arduous, step of the analysis consists in identifying underlying logics. This step is where the data collected by means of a chaîne opératoire approach prove essential. First, one should determine whether such logics directly or indirectly pertain to technical factors. For instance, an uneven distribution of raw materials or the existence of marked ecological contrasts can directly affect the distribution of technical behaviors. Regarding postfiring treatments, and specifically the coating of vessels with organic mixtures, the fact that no potters in the Sahel area use the bark of Parkia filicoidea or Syzygium rowlandii for preparing the mixture is merely because both tree species grow in wet evergreen or semideciduous forests. In the latter environments, the use or nonuse of such tree species necessarily involves factors other than ecological ones. Economic and functional considerations may also directly affect technical behaviors. For example, some scholars interpreted the large-scale diffusion of the pounding technique (used for shaping vessels) in relation to the fact that it is fast and allows for the making of light, thin-walled pots—two factors that supposedly ensured its swift adoption.90 Another example concerns microregional processes of technical standardization around commercial centers where artisans and customers from diverse backgrounds meet and interact with each other.91 In all these examples, pottery chaînes opératoires are the main components affected by an element of the surrounding context, and changes and continuities in technical practice depend on the agency of the potters.

Yet underlying logics often pertain to indirect relations. To make sense of a particular configuration of technical traits, one must therefore identify the elements liable to impact the potters’ social world of activity and, ultimately, their behavior. Such elements potentially concern a vast array of domains, and notably religious or symbolic ones.92 In southcentral Niger, for example, male Hausa potters stopped tempering clay with crushed donkey dung at the turn of the 21th century as a response to the preaching of a new generation of Muslim clerics who condemned donkeys (and their excretions) as “impure”. Although potters still perceive donkey dung positively in strictly technical terms, both their religious faith and their fear of alienating local Muslim clients led them to abandon a tradition inherited from previous generations of potters. In this case, new Muslim orthodoxy and the condemnation of donkeys developed independently from potting practice, but they impacted the potters’ social world of activity, leading artisans to ultimately abandon a part of their technical repertoire.

Other examples of indirect relations may concern political boundaries. As an illustration, Livingstone Smith shows that the ancient polities of the Katanga Province (southeastern Democratic Republic of the Congo) could have had an influence on the local distribution of extant pottery traditions.93 He indeed observes a spatial correspondence between the geographical distribution of five pottery repertoires (defined according to roughing out and decoration techniques) and that of the putative boundaries of Katanga kingdoms around 1800 ad (see Figure 7).94 The question, of course, is how these polities—which notably controlled the exploitation of salt and copper mines, and whose actual boundaries remain much debated—could have had an impact on the circulation of potting knowledge; that is, on the dynamics of a female sphere of activity much remote from political struggles, warfare, or long-distance trade. Livingstone Smith’s answer is to consider each technical area as the product of a long-term aggregation of individual social spaces.95 If we conceive the social space as a combination of spaces experienced and known and, more broadly, as a constellation of places where concrete relations take place between persons, things, materials, representations, and the environment, it is not difficult to envision how (or a least why) they should be shaped by historical, social, and ecological circumstances.96 Likewise, since the ancient polities of Katanga must have impacted the space of experience and the space known of the people living in the area (including female potters), it is only understandable that they would have also impacted their circulation and, among many other elements, the circulation of pottery knowledge.

Pottery chaînes opératoires as Historical DocumentsClick to view larger

Figure 7. The spatial distribution of pottery repertoires based on roughing out and decoration techniques (capital letters) as a putative afterglow of ancient political boundaries in southcentral Africa.97 These political boundaries partially match that of the main salt mines (black dots) and their distribution zones (shaded areas). Arrows indicate the post-learning mobility of individual potters long after the disappearing of ancient polities.

Map by A. Livingstone Smith.

This example brings forward a last crucial element pertaining to the comparative technology approach: the fact that the spatial distribution of a given technique or tradition sometimes constitutes an afterglow of past historical processes. Indeed, the ancient political boundaries of Katanga did not survive the socioeconomic disruptions of the 19th century and the colonial era; yet, as stated by Livingstone Smith, the “mesh of relationships” linked to the social spaces they contributed to shape “may have survived for other purposes or, more surreptitiously, linger in words, gestures or tools.”98 The irony here is that a mundane element such as pottery making, a priori unrelated to the political sphere, can ultimately become a key element for approaching the tangible reality of political boundaries.

As with historical linguistics, this persistence of the past into the present is one of the main assets of pottery chaînes opératoire and the raison d’être of comparative technology.

Discussion of the Literature

The comparative technology approach described in this article is one of the founding principles of the French “Techniques & Culture” research program.99 Its potential in historical analyses remains largely unrecognized, despite promising results.100 Most technological comparisons carried out by anthropologists are indeed synchronic, aiming at characterizing the techniques and technical systems of given societies and at untangling their underlying logics. In such instances, the chaîne opératoire primarily serves as an analytical tool: a template allowing the systematic collection of technical data and their ordering in view of subsequent analyses.

Pottery manufacturing processes started to be documented in Africa by the turn of the 20th century.101 When published, information appeared either as curios, specific sections in ethnographical monographs, complementary data for museum collections, or reference material intended for the interpretation of archaeological materials. The latter grew in importance after the 1950s—together with the development of ethnoarchaeology102—until becoming the primary reason for documenting pottery production and consumption in Africa. Yet even if archaeologists carry out most inquiries and remain the principal beneficiaries, they rarely consider the historical dimension of the practices they document, some going as far as denying its relevance for ethnoarchaeological purposes.103 Unsurprisingly, early attempts at approaching local pottery traditions in a diachronic perspective104—or at drawing historical conclusions from a regional or continental comparison of technical “traits”105—were made by other specialists, among whom were anthropologists, art historians, and historians. These contributions remained scarce and were occasionally undermined by methodological weaknesses.106

It is only during the last decades of the 20th century that pottery chaînes opératoires started to be more consistently exploited as historical documents, mainly among researchers engaged in direct historical approaches.107 This change arose concomitantly with a shift in the interpretative paradigms of pottery techniques, and especially with an awareness that there were consistent relationships between techniques and identities (language, ethnicity, gender, intravillage groupings, etc.). Such relationships are the focus of most diachronic comparisons of pottery chaînes opératoires in Africa, the quest for technical markers of identity developing as a reaction to the “information exchange theory”108 that viewed style as residing only in the most visible parts of the vessels—especially decoration—and stemming from a deliberate expression of ethnic belonging aimed at foreign groups.109 Long-term ethnographic studies reveal not only that style may correspond to a reification of worldviews aimed primarily at those who make and use the vessels,110 but also that various forms of identity belonging may be read at different levels of the artifacts, including chaînes opératoires.

The explicit or implicit implementation of direct historical approaches centered on pottery chaînes opératoires has clearly improved our understanding of archaeological and historical contexts in different parts of the African continent.111 Yet the potentials of comparative technology in which extant pottery traditions are not exploited as baselines against which to compare archaeological data, but as actual historical sources (inasmuch as they are shaped by a web of sociohistorical relations), still remain underexploited. This article aims at redressing this imbalance.

Primary Sources

Given the variety of profiles of those who collected information on pottery production in Africa since the beginning of the 20th century, data are scattered in a wide range of publications. Up to the 1970s, interested scholars should especially peruse journals and monographs in the fields of anthropology and ethnology; from then on, they will probably find more materials in archaeological, art history, and museum publications. Since the 1980s to 1990s, master’s theses achieved in the humanities departments of African universities also provide reliable firsthand data.

While most published studies document chaînes opératoires in a limited number of potting communities, there exists some broader comparisons or compilations of data, developing either at a continental, subcontinental, or regional level.112 None of the data compilations should be considered exhaustive or definitive, however, especially since dozens of new studies are liable to appear each year. Researchers should also be aware that published data are only the tip of the iceberg as far as pottery making (and any other craft) is concerned. Besides public and private pottery collections, whose close examination may provide technical information, there exists a large body of gray literature, especially in museums associated with the colonial enterprise. Much remains to be done as regards the inventory and exploitation of such material.

Further Reading

Bromberger, Christian, and Alain Morel. Limites floues, frontières vives. Des variations culturelles en France et en Europe. Paris: Éditions de la Maison des Sciences de l’Homme, 2001.Find this resource:

Frank, Barbara E. Mande Potters and Leatherworkers. Washington, DC: Smithsonian Institution Press, 1998.Find this resource:

Gosselain, Olivier P. “A Tradition in Nine Maps. Un-layering Niger River Polychrome Water Jars.” In Balkan Dialogues. Negotiating Identity Between Prehistory and the Present. Edited by Maja Gori and Maria Ivanova, 85–108. London: Routledge, 2017Find this resource:

Langlois, Olivier. “Distribution des techniques actuelles de façonnage céramique au sud du bassin tchadien: un outil pour la recherche historique régionale.” Journal des Africanistes 71, no. 1 (2001): 225–256.Find this resource:

Lave, Jean, and Etienne Wenger. Situated Learning. Legitimate Peripheral Participation. Cambridge, UK: Cambridge University Press, 1991.Find this resource:

Lemonnier, Pierre. Elements for an Anthropology of Technology. Ann Arbor: Museum of Anthropology, University of Michigan, 1992.Find this resource:

Mayor, Anne. “Ceramic Traditions and Ethnicity in the Niger Bend, West Africa.” Ethnoarchaeology 2, no. 1 (2010): 5–48.Find this resource:

Sall, Moustapha. Traditions céramiques, identités et peuplement en Sénégambie. Ethnographie comparée et essai de reconstitution historique. Oxford: Bar International Series, Cambridge Monographs in African Archaeology 63, 2005.Find this resource:

Smith, Alexandre Livingstone. “Pottery and Politics: Making Sense of Pottery Traditions in Central Africa.” Cambridge Archaeological Journal 26, no. 3 (2016): 471–491.Find this resource:

Sterner, Judy, and Nicholas David. “Action on Matter: The History of the Uniquely African Tamper and Concave Anvil Pot‐Forming Technique.” Journal of African Archaeology 1, no. 1 (2003): 3–38.Find this resource:


(1.) Jean Lave, “The Practice of Learning,” in Understanding Practice. Perspectives on Activity and Context, ed. Seth Chaiklin and Jean Lave (Cambridge, UK: Cambridge University Press, 1996), 3–32; Heather Lechtman, “Style in Technology. Some Early Thoughts,” in Material Culture. Style, Organization, and Dynamics of Technology, ed. Heather Lechtman and Robert S. Merrill (Saint Paul, MN: West Publishing Company, 1977), 3–20; and Pierre Lemonnier, Elements for an Anthropology of Technology (Ann Arbor: Museum of Anthropology, University of Michigan, 1992).

(2.) Lemonnier, Elements, 21.

(3.) Hélène Balfet, ed., Observer l’action technique. Des chaînes opératoires pour quoi faire? (Paris: Éditions du CNRS, 1991), 12–17.

(4.) Regarding Africa, readers will find details in Dietrich Drost, Töpferei in Afrika: Technologie (Leipzig: Akademieverlag, 1967); Olivier P. Gosselain, “Ceramics in Africa,” in Encyclopaedia of the History of Science, Technology, and Medicine in Non-Western Cultures, ed. Helen Selin (New York: Springer, 2014), 464–477; Anne C. Lawton, “Bantu Pottery of Southern Africa,” Annals of the South African Museum 49 (1967): 1–434; and Alexandre Livingstone Smith, Chaînes opératoires de la poterie. Références ethnographiques, analyse et reconstitution (Tervuren, Belgium: Musée Royal de l'Afrique Centrale, 2007).

(5.) Barbara E. Frank, “Reconstructing the History of an African Ceramic Tradition. Technology, Slavery and Agency in the Region of Kadiolo (Mali),” Cahiers d'Études Africaines 33, no. 3 (1993): 381–401; Barbara E. Frank, Mande Potters and Leatherworkers (Washington, DC: Smithsonian Institution Press, 1998); Alain Gallay, “Sociétés englobées et traditions céramiques: le cas du Pays dogon (Mali) depuis le 13e siècle,” in Terre cuite et société: la céramique, document technique, économique, culturel, ed. Françoise Audouze and Didier Binder (Juan-les-Pins, France: Editions APDCA, 1994); Alain Gallay, Youssouf Kalapo, and Elisée Guindo, “Les traditions céramiques Dogon,” Etudes Maliennes 65 (2006): 127–144; Olivier P. Gosselain, “Poterie, société et histoire chez les Koma Ndera du Cameroun,” Cahiers d'Études Africaines 153, no. 39 (1999): 73–105; Olivier P. Gosselain, “Ethnographie comparée des trousses à outils de potiers au sud du Niger,” Bulletin de la Société préhistorique française 107, no. 4 (2010): 667–689; Olivier P. Gosselain, “Roads, Markets, Migrants. The Historical Trajectory of a Male Hausa Pottery Tradition in Southern Niger,” in The Distribution of Technological Knowledge in the Production of Ancient Mediterranean Pottery, ed. Walter Gauß, Gudrun Klebinder-Gauß, and Constance von Rüden (Vienna: Sonderschriften ÖAI, 2015), 277–296; Olivier P. Gosselain, “A Tradition in Nine Maps. Un-layering Niger River Polychrome Water Jars,” in Balkan Dialogues. Negotiating Identity Between Prehistory and the Present, ed. Maja Gori and Maria Ivanova (London: Routledge, 2017), 85–108; Eric Huysecom, “Iron Age Terracotta Pestles in the Sahel Area: An Ethnoarchaeological Approach,” in Interregional Contacts in the Later Prehistory of Northeastern Africa, ed. Lech Krzyzaniak, Karla Kroeper, and Michal Kobusiewicz (Poznan, Poland: Archaeological Museum, Studies in African Archaeology 5, 1996); Alexandre Livingstone Smith, “Pottery and Politics: Making Sense of Pottery Traditions in Central Africa,” Cambridge Archaeological Journal 26, no. 3 (2016): 471–491; Anne Mayor, “Ceramic Traditions and Ethnicity in the Niger Bend, West Africa,” Ethnoarchaeology 2, no. 1 (2010): 5–48; Anne Mayor, Traditions céramiques dans la boucle du Niger: ethnoarchéologie et histoire du peuplement au temps des empires précoloniaux (Frankfurt am Main: Africa Magna Verlag, Journal of African Archaeology Monograph Series 7, 2011); Anne Mayor, “Impressions de vannerie et technique du martelage sur forme concave: anthropologie et histoire d’une technique dans la boucle du Niger,” Azania. Archaeological Research in Africa 46, no. 1 (2011): 88–109; Bruno Pinçon and Dominique Ngoie-Ngalla, “L’unité culturelle Kongo à la fin du XIXème siècle. L’apport des études céramologiques,” Cahiers d’Études Africaines 118, no. 30–32 (1990): 157–178; and Moustapha Sall, Traditions céramiques, identités et peuplement en Sénégambie. Ethnographie comparée et essai de reconstitution historique (Oxford: Bar International Series, Cambridge Monographs in African Archaeology 63, 2005).

(6.) Valentine Roux, “La technique du tournage: définition et reconnaissance par les macrotraces,” in Terre cuite et société. La céramique, document technique, économique, culturel, ed. Françoise Audouze and Didier Binder (Juan-les-Pins, France: Editions APDCA, 1994), 45–58, 46–47.

(7.) For example, Marie-Louise Inizan, Michèle Reduron, Hélène Roche, and Jacques Tixier, Technologie de la pierre taillée (Meudon, France: Cercle de Recherches et d’Etudes Préhistoriques, 1995).

(8.) For example, David P. Braun, “Pots as Tools,” in Archaeological Hammers and Theories, ed. James A. Moore and Arthur S. Keene (New York: Academic Press, 1983), 107–134; Prudence M. Rice, Pottery Analysis: A Sourcebook (Chicago: University of Chicago Press, 1987); and Michael S. Tite, “Pottery Production, Distribution, and Consumption. The Contribution of the Physical Sciences,” Journal of Archaeological Method and Theory 6, no. 3 (1999): 181–233.

(9.) Lechtman, “Style in Technology.”

(10.) Lemonnier, Element.

(11.) Olivier P. Gosselain, Poteries du Cameroun méridional. Styles techniques et rapports à l'identité (Paris: CNRS Editions, 2002); and Alexandre Livingstone Smith, “Processing Clay for Pottery in Northern Cameroon: Social and Technical Requirements,” Archaeometry 42, no. 1 (2000): 21–42.

(12.) Alexandre Livingstone Smith, “Bonfire 2. The Return of Pottery Firing Temperatures,” Journal of Archaeological Science 28, no. 9 (2001): 99–117.

(13.) André Leroi-Gourhan, L’Homme et la matière (Paris: Albin Michel, 1971 [1943]).

(14.) Alain Gallay, Eric Huysecom, and Anne Mayor, Peuples et céramiques du delta intérieur du Niger (Mali): un bilan de cinq années de missions 1988–1993 (Mainz, Germany: Verlag Philipp von Zabern, Terra Archaeologica 3, 1998); Mayor, “Ceramic Traditions and Ethnicity”; Mayor, Traditions céramiques; and Olivier Langlois, “Distribution des techniques actuelles de façonnage céramique au sud du bassin tchadien: un outil pour la recherche historique régionale,” Journal des Africanistes 71, no. 1 (2001): 225–256.

(15.) See especially: Mayor, Traditions céramiques; and Mayor, “Impressions de vannerie.”

(17.) Gosselain, “Roads, Markets, Migrants.”

(18.) Claire Corniquet, “Cadres de pratiques et circulation des connaissances chez les potières de l’Arewa (Niger),” Cahiers d’Etudes Africaines 51, no. 201 (2011): 87–114; and Claire Corniquet, Ancrage social, ancrage spatial. Circulation des savoirs céramiques chez les potières de l’Arewa et du Kurfey (Niger) (Brussels: Université libre de Bruxelles, unpublished PhD diss., 2014).

(19.) Morie Kaneko, “Variations in Pottery Making by Ari potters in Southwestern Ethiopia: Analysis of the Finger Movement Patterns Used in Forming Pots,” Nilo-Ethiopian Studies 11 (2007): 1–15.

(20.) See Anne Haour, Katie Manning, Noemie Arazi, Olivier P. Gosselain, Sokhna Guèye, Daouda Keita, Alexandre Livingstone Smith, Kevin MacDonald, Anne Mayor, Susan McIntosh, and Robert Vernet, African Pottery Roulettes, Past and Present. Techniques, Identification and Distribution (Oxford: Oxbow Books, 2010), 70–74.

(21.) Mayor, Traditions céramiques; Mayor, “Impressions de vannerie.”

(22.) Although this case requires further consideration, it seems probable that male itinerant potters contributed to the current distribution of this ‘technical bundle’ in pre- or early colonial times, as they did in Niger and Nigeria; see Gosselain, “Roads, Markets, Migrants.”

(23.) Olivier P. Gosselain, “Materializing Identities: An African Perspective,” Journal of Archaeological Method and Theory 7, no. 3 (2000): 187–217.

(24.) For a visual rendition of the model, see Ken D. Fowler, “Zulu Pottery Production in the Lower Thukela Basin, KwaZulu-Natal, South Africa,” Southern African Humanities 20 (2008): 477–511, 484.

(25.) Corniquet, “Cadres de pratiques.”

(26.) Dean R. Arnold, “A Model for the Identification of Non-Local Ceramic Distribution: View from the Present,” in Production and Distribution: A Ceramic Viewpoint, ed. Hilary Howard and Elaine Morris (Oxford: BAR International Series 120, 1981), 31–44.

(27.) For example, Diane Lyons and Andrea Freeman, “I’m Not Evil: Materialising Identities of Marginalised Potters in Tigray Region, Ethiopia,” Azania: Archaeological Research in Africa 44, no. 1 (2009): 75–93.

(28.) Cameron Gokee, “Shapen Signs: Pottery Techniques, Indexicality, and Ethnic Identity in the Saluum, Senegambia (ca. 1700–1950),” in Ethnic Ambiguity and the African Past, ed. François G. Richard and Kevin C. MacDonald (Walnut Creek, CA: Left Coast Press, 2015), 55–86.

(29.) Gokee, “Shapen Signs,” 63.

(30.) Sokhna Guèye, “Dis-moi quel pot tu as et je te dirai qui tu es! Matérialiser les identités sociales dans les décors céramiques de la moyenne vallée du fleuve Sénégal (nord du Sénégal),” Azania. Archaeological Research in Africa 46, no. 1 (2011): 20–35.

(31.) See, for example, Maria Dores Cruz, “‘Pots Are Pots, Not People’: Material Culture and Ethnic Identity in the Banda Area (Ghana), Nineteenth and Twentieth Centuries,” Azania: Archaeological Research in Africa 46, no. 3 (2011): 336–357; Scott A. MacEachern, “Scale, Style, and Cultural Variation: Technological Traditions in the Northern Mandara Mountains,” in The Archaeology of Social Boundaries, ed. Miriam Stark (Washington, DC: Smithsonian Institution Press, 1998), 107–131; and Phenio C. Thebe and Karim Sadr, “Firing Pots in Contemporary South-Eastern Botswana,” Ethnoarchaeology 9, no. 2 (2017): 146–165.

(32.) Agnès Gelbert, Ceramic Traditions and Technical Borrowings in the Senegal Valley (Paris: Editions de la MSH & Editions Epistèmes, 2003); and Olivier P. Gosselain, “The World Is Like a Beanstalk: Historicizing Potting Practice and Social Relations in the Niger River Area,” in Knowledge in Motion: Constellations of Learning Across Time and Place, ed. Andrew P. Roddick and Ann Brower Stahl (Tucson: University of Arizona Press, 2016), 36–66; see note 23, Sterner and David, “Action on Matter.”

(33.) MacEachern, “Scale, Style, and Cultural Variation”; and Innocent Pikirayi, “Ceramics and Group Identities. Toward a Social Archaeology in Southern African Iron Age Ceramic Studies,” Journal of Social Archaeology 7, no. 3 (2007): 286–301.

(34.) Ingrid Herbich, “Learning Patterns, Potter Interaction and Ceramic Style among the Luo of Kenya,” The African Archaeological Review 5 (1987): 153–204; Fowler, “Zulu Pottery Production”; Frank, “Reconstructing the History”; Frank, Mande Potters; Lyons and Freeman, “I’m Not Evil”; Thebe and Sadr, “Firing Pots”; Phenio C. Thebe and Karim Sadr, “Forming and Shaping Pottery Boundaries in Contemporary South-Eastern Botswana,” African Archaeological Review 34 (2017): 75–92; and Nathalie Tobert, “Potters of El-Fasher: One Technique Practised by Two Ethnic Groups,” in Earthenware in Asia and Africa, ed. John Picton (London: Percival David Foundation, 1984), 219–237.

(35.) Lave, “The Practice of Learning”; and Jean Lave, Apprenticeship in Critical Ethnographic Practice (Chicago: University of Chicago Press, 2011).

(36.) Lave, Apprenticeship, 143.

(37.) Jean Lave and Etienne Wenger, Situated Learning. Legitimate Peripheral Participation (Cambridge, UK: Cambridge University Press, 1991); and Etienne Wenger, Communities of Practice. Learning, Meaning, and Identity (New York: Cambridge University Press, 1998).

(38.) Gosselain, “The World Is Like a Beanstalk,” 55.

(39.) Gosselain, “A Tradition in Nine Maps,” 94–95.

(40.) See notably Liza Gijanto, “Exchange, Interaction, and Change in Local Ceramic Production in the Niumi Commercial Center on the Gambia River,” Journal of Social Archaeology 11, no. 1 (2011): 21–48.

(41.) Wenger, Communities of Practice, 175–178.

(42.) Wenger, Communities of Practice, 178–181.

(43.) Wenger, Communities of Practice, 174.

(44.) See examples in Gelbert, Ceramic Traditions.

(45.) Corniquet, Ancrage social, ancrage spatial.

(46.) Ismail Rashid, “Class, Caste and Social Inequality in West African History,” in Themes in West Africa’s History, ed. Emmanuel K. Akyaempong (Oxford: James Currey, 2006), 118–140; and Tal Tamari, “The Development of Caste System in West Africa,” Journal of African History 32, no. 2 (1991): 221–250.

(47.) See details in Leroi-Gourhan, L’Homme et la matière, 43–113.

(48.) For example, Kaneko, “Variations in Pottery Making.”

(49.) François Sigaut, “Quelques remarques sur la nomenclature des outils,” in Outils aratoires en Afrique. Innovations, normes et traces, ed. Yves Marzouk, Christian Seignobos, and François Sigaut (Paris: Karthala, 2000), 369–375.

(50.) Lemonnier, Elements; Marie-Claude Mahias, Le barattage du monde. Essai d’anthropologie des techniques en Inde (Paris: Editions de la Maison des Sciences de l’Homme, 2002).

(51.) Amanda L. Logan and M. Dores Cruz, “Gendered Taskscapes: Food, Farming, and Craft Production in Banda, Ghana in the Eighteenth to Twenty-First Centuries,” African Archaeological Review 31 (2014): 203–231.

(52.) Logan and Cruz, “Gendered Taskscapes”; and André Leroi-Gourhan, Milieu et technique (Paris: Albin Michel, 1973 [1945]).

(53.) Nigel Barley, Smashing Pots. Feats of Clay from Africa (London: The British Museum Press, 1994); Per Ditlef Fredriksen, Material Knowledge, Thermodynamic Spaces and the Moloko Sequence of the Late Iron Age (AD 1300–1840) in Southern Africa (Oxford: BAR International Series 2387/Cambridge Monograph in African Archaeology 80, 2012); Olivier P. Gosselain, “In Pots We Trust. The Processing of Clay and Symbols in Sub-Saharan Africa,” Journal of Material Culture 4, no. 2 (1999): 205–230; and Malcolm D. McLeod, “Akan Terracotta,” in Earthenware in Asia and Africa, ed. John Picton (London: Percival David Foundation, 1984), 365–381.

(54.) For example, Birgit Riquier, “The ‘Words and Things’ Method,” in Field Manual for African Archaeology, ed. Alexandre Livingstone Smith, Els Cornelissen, Olivier P. Gosselain, and Scott MacEachern (Tervuren, Belgium: Royal Museum for Central Africa, Documents on Social Sciences and Humanities, 2017), 261–263; and David L. Schoenbrun, The Historical Reconstruction of Great Lakes Bantu Cultural Vocabulary (Cologne: Ruediger Koepp Press, 1997).

(55.) Koen Bostoen, “What Comparative Bantu Pottery Vocabulary May Tell Us About Early Human Settlement in the Inner Congo Basin,” Afrique & Histoire 5 (2006): 221–263.

(56.) As advocated by Igor Kopytoff, “The Cultural Biography of Things: Commodization as a Process,” in The Social Life of Things, ed. Arjun Appadurai (New York: Cambridge University Press, 1986), 64–91.

(57.) Wenger, Communities of Practice.

(58.) See a particularly fruitful application in Barbara van Doosselaere, Le roi et le potier. Etude technologique de l’assemblage céramique de Koumbi Saleh, Mauritanie (5e/6e – 17e siècles AD) (Frankfurt am Main: Africa Magna Verlag/Reports in African Archaeology 5, 2005).

(59.) Diane Lyons, “Ethnoarchaeology,” in Field Manual for African Archaeology, ed. Alexandre Livingstone Smith, Els Cornelissen, Olivier P. Gosselain, and Scott MacEachern (Tervuren, Belgium: Royal Museum for Central Africa, Documents on Social Sciences and Humanities, 2017), 270–274.

(60.) Ann B. Stahl, “The Direct Historical Approach,” in Field Manual for African Archaeology, ed. Alexandre Livingstone Smith, Els Cornelissen, Olivier P. Gosselain, and Scott MacEachern (Tervuren, Belgium: Royal Museum for Central Africa, Documents on Social Sciences and Humanities, 2017), 250–252.

(61.) See, for example: Leonard B. Crossland and Merrick Posnansky, “Pottery, People and Trade at Begho,” in The Spatial Organisation of Culture, ed. Ian Hodder (London: Duckworth, 1978), 77–89; M. Dores Cruz, Shaping Quotidian Worlds: Ceramic Production and Consumption in Banda, Ghana c. 1780–1994 (PhD diss., Binghamton University [SUNY], 2003); Kwaku Effah-Gyamfi, “Traditional Pottery Technology at Krobo Takyiman (Techniman), Ghana: An Ethnoarchaeological Study,” West African Journal of Archaeology 10 (1980): 103–116; Fredriksen, Material Knowledge; Tom N. Huffman, “Shona Pottery From Pumula Township, Bulawayo, Rhodesia,” South African Archaeological Bulletin 27 (1972): 66–81; Mayor, Traditions céramiques; Alice Mezop Temgoua-Noumissie, “Pottery and Oral History in the Faro (Northern Cameroon),” in Field Manual for African Archaeology, ed. Alexandre Livingstone Smith, Els Cornelissen, Olivier P. Gosselain, and Scott MacEachern Tervuren, Belgium: Royal Museum for Central Africa, Documents on Social Sciences and Humanities, 2017), 275–279; and Ann B. Stahl, Making History in Banda: Anthropological Visions of Africa’s Past (Cambridge, UK: Cambridge University Press, 2001).

(62.) William Yewdale Adams, “On the Argument From Ceramics to History: A Challenge Based on Evidence from Medieval Nubia,” Current Anthropology 20, no. 4 (1979): 727–744.

(63.) Frank “Reconstructing the History”; Frank, Mande Potters.

(64.) Nicolas Nikis, Louis Champion, and Els Cranshof, “Putting Together the Mande Puzzle. Mapping Pottery Techniques and Identities in West Africa” (unpublished paper presented at the Society of Africanist Archaeologists 21st Biennial Meeting, Toronto, Canada, June 20–23, 2012).

(65.) Claire Grégoire and Bernard de Halleux, “Étude lexicostatistique de quarante-trois langues et dialectes mande,” in Africana Linguistica XI (Tervuren, Belgium: Musée Royal de l'Afrique Centrale, Annales du MRAC, Sciences Humaines 142, 1994), 53–70.

(66.) Georges E. Brooks, Western Africa to c. 1860 A.D.: A Provisional Historical Schema Based on Climate Periods (Bloomington: African Studies Program, Indiana University, Working Papers no. 1, 1985).

(67.) Brooks, Western Africa; Frank, Mande Potters; and Tamari, “The Development of Caste System.”

(68.) Gallay, “Sociétés englobées”; and Gallay et al., “Traditions céramiques Dogon.”

(69.) Haour et al., African Pottery Roulettes.

(70.) Gosselain, “Marterializing Identities”; and Alexandre Livingstone Smith, “Histoire du décor à la roulette en Afrique subsaharienne,” Journal of African Archaeology 5, no. 2 (2007): 189–216.

(71.) For example, Christiane Desmedt, “Poteries anciennes décorées à la roulette dans la Région des Grands Lacs,” The African Archaeological Review 9 (1991): 161–196; for a critical discussion, see Kearsley A. Stewart, “Iron Age Ceramic Studies in Great Lakes Eastern Africa: A Critical and Historiographical Review,” The African Archaeological Review 11 (1993): 21–37.

(72.) Except for some experiments made under dubious conditions; for example, Hélène Wallaert, “Learning How to Make the Right Pots. Apprenticeship Strategies and Material Culture: A Case Study in Handmade Pottery From Cameroon,” Journal of Anthropological Research 57 (2001): 471–493.

(73.) Nikis et al., “Putting Together the Mande Puzzle.”

(74.) Alexandre Livingstone Smith, “Pottery Traditions in Katanga (DRC). A Comparative Examination of Roughing Out Technologies,” Anthropos 105, no. 1 (2010): 179–190; and Livingstone Smith; “Pottery and Politics.”

(75.) Haour et al., African Pottery Roulettes, 62–68.

(76.) After Sigaut, “Quelques remarques.”

(77.) With the use of free websites such as GeoNames.

(78.) See Gosselain, “A Tradition in Nine Maps.”

(79.) For example, Hannah J. Haynie, “Geography and Spatial Analysis in Historical Linguistics,” Language and Linguistics Compass 8 (2014): 344–357.

(81.) Smith, “Processing Clay for Pottery.”

(82.) Other examples of the way in which structuring elements in the landscape weigh on the evolution of pottery traditions are given in Gosselain, “The World Is Like a Beanstalk,” and Gosselain, “A Tradition in Nine Maps.”

(83.) Thebe and Sadr, “Firing Pots,” 153.

(84.) Thebe and Sadr, “Forming and Shaping Pottery”; Phenio C. Thebe and Karim Sadr, “Pottery Decoration in Contemporary Southeastern Botswana,” Azania. Archaeological Research in Africa 52, no. 3 (2017): 305–323.

(85.) Haour et al., African Pottery Roulettes; their spatial distribution does indeed crosscut that of three language families and a dozen of their respective subdivisions, without ever matching them entirely (see Gosselain, “Materializing Identities,” 199).

(86.) Livingstone Smith, “Histoire du décor à la roulette.”

(87.) Frank “Reconstructing the History”; Frank, Mande Potters; Gallay, “Sociétés englobées”; and Gallay et al., “Traditions céramiques Dogon.”

(88.) Gosselain (2002), Poteries du Cameroun méridional; Gosselain, “Roads, Markets, Migrants”; Huysecom, “Iron Age Terracotta Pestles”; Livingstone Smith, “Pottery Traditions in Katanga”; and Livingstone Smith, “Pottery and Politics.”

(89.) Agnès Gelbert, “Ethnoarchaeological Study of Ceramic Borrowings: A New Methodological Approach Applied in the Middle and Upper Valleys of the Senegal River,” in Ethnoarchaeology and Its Transfers, ed. Sylvie Beyries and Pierre Pétrequin (Oxford: BAR International Series 983, 2001), 81–94; and Gelbert, Ceramic Traditions.

(90.) Sterner and David, “Action on Matter”; note, however, that other factors are conceivable here, notably the long-distance migrations of specialist potters, population movements, and/or marriage prescriptions within the socioprofessional subgroups concerned (see Gosselain, “Roads, Markets, Migrants”; Huysecom, “Iron Age Terracotta Pestles”; and MacEachern, “Scale, Style, and Cultural Variation”).

(91.) Gosselain, “The World Is Like a Beanstalk”; Gosselain, “A Tradition in Nine Maps”; and Guèye, “Dis-moi quel pot tu as.”

(92.) See examples in Barley, Smashing Pots; Gosselain, “In Pots We Trust.”

(93.) Livingstone Smith, “Pottery and Politics.”

(94.) According to a map published by Jan Vansina, Les anciens royaumes de la savane (Léopoldville: Université de Lovanium, Institut de Recherches Economiques et Sociales, 1965).

(95.) Livingstone Smith, “Pottery and Politics,” 15.

(96.) Gosselain, “The World Is Like a Beanstalk.”

(97.) After Livingstone Smith, “Pottery and Politics,” 15.

(98.) Livingstone Smith, “Pottery and Politics,” 16.

(99.) Robert Creswell, “Transfert de techniques et chaînes opératoires,” Techniques & Culture 2 (1983): 143–163; Lemonnier, Elements.

(100.) Bromberger and Morel, Limites floues; and André-Georges Haudricourt, La technologie science humaine. Recherches d’histoire et d’ethnologie des techniques (Paris: Editions de la Maison des Sciences de l’Homme, 1987).

(101.) For a historiography of pottery studies in Africa, see Olivier P. Gosselain and Alexandre Livingstone Smith, “A Century of Ceramic Studies in Africa,” in The Oxford Handbook of African Archaeology, ed. Peter Mitchell and Paul Lane (Oxford: Oxford University Press, 2013), 117–129.

(102.) Nicholas David and Carol Kramer, Ethnoarchaeology in Action (Cambridge, UK: Cambridge University Press, 2001).

(103.) Valentine Roux, “Ethnoarchaeology. A Non-Historical Science of Reference Necessary for Interpreting the Past,” Journal of Archaeological Method and Theory 14, no. 2 (2007): 153–178.

(104.) For example, A. C. Quarcoo and Marion Johnson, “Shai Pots. The Pottery Industry of the Shai People of Ghana,” Baessler-Archiv, Beiträge zur Völkerkunde 16 (1968): 47–88; and Enid Schildkrout, Jill Hellman, and Curtis A. Keim, “Mangbetu Pottery: Tradition and Innovation in Northeast Zaïre,” African Arts 22, no. 2 (1989): 38–47.

(105.) Drost, Töpferei; and Lawton, “Bantu Pottery.”

(106.) See especially Drost, Töpferei.

(107.) For a presentation of the method and a good case study, see Ann B. Stahl, Making History in Banda: Anthropological Visions of Africa’s Past (Cambridge, UK: Cambridge University Press, 2001).

(108.) Martin H. Wobst, “Stylistic Behavior and Information Exchange,” in For the Director: Research Essays in Honor of James B. Griffin, ed. Charles E. Cleland (Ann Arbor: University of Michigan Museum of Anthropology [Memoir 13], 1977), 317–342.

(109.) Ian Hodder, “Economic and Social Stress and Material Culture Patterning,” American Antiquity 44 (1979): 446–454; and Ian Hodder, Symbols in Actions. Ethnoarchaeological Studies of Material Culture (Cambridge, UK: Cambridge University Press, 1982).

(110.) Nicholas David, Judy Sterner, and Kodzo Gavua, “Why Pots Are Decorated,” Current Anthropology 29, no. 3 (1988): 365–389; and Judy Sterner, “Who Is Signalling Whom? Ceramic Style, Ethnicity and Taphonomy Among the Sirak Bulahay,” Antiquity 63 (1989): 451–459.

(111.) Examples include: Crossland and Posnansky, “Pottery, People and Trade at Begho”; Cruz, Shaping Quotidian Worlds; Effah-Gyamfi, “Traditional Pottery Technology at Krobo Takyiman”; Fredriksen, Material Knowledge; Huffman, “Shona Pottery”; Mayor, Traditions céramiques; Mezop Temgoua-Noumissie, “Pottery and Oral History”; and Stahl, Making History in Banda.

(112.) Drost, Töferei; Gosselain, Poterie du Cameroun méridional; Gosselain, “Ceramics in Africa”; Haour et al., African Pottery Roulettes; Huysecom, “Iron Age Terracotta Pestles”; and Sterner and David, “Action on Matter”; for example, Gallay et al., Peuples et céramiques; Lawton, “Bantu Pottery”; and Mayor, Traditions céramiques.