Emil Adamec - Persisting Technological Boundaries 2023

MgA. Emil Adamec

PERSISTING TECHNOLOGICAL BOUNDARIES

ethnographic studies about technological boundaries and cognitive biases which resultate polarization

Social Interactions, Cognitive Correlations and Polarization

Persistence of Technological Boundaries and Techniques Polarizing Social Groups

December 18, 2023 at 10:30 a.m. in room C335

INSTITUTE OF CLASSICAL ARCHAEOLOGY & INSTITUTE OF ETHNOLOGY

Faculty of Arts

CHARLES UNIVERSITY

1 What is about

2 What are the contributions

3 Introduction

4 Methods

5 Data

6 Results

7 Conclusions

8 Limitations

9 Future works

10 Literature survey

11 Practical implications

12 Related Papers

Contents lists available at ScienceDirect

Journal of Anthropological Archaeology

journal homepage:

www.elsevier.com/locate/jaa

Persisting technological boundaries: Social interactions, cognitive

correlations and polarization

Valentine Roux

⁎

, Blandine Bril

, Jessie Cauliez

, Anne-Lise Goujon

, Catherine Lara

Claire Manen

, Geoﬀroy de Saulieu

, Etienne Zangato

Préhistoire & Technologie, CNRS, Nanterre, France

EHESS, Paris, France

Traces, CNRS, Toulouse, France

IRD, Marseille, France

ARTICLE INFO

Keywords:

Technological boundaries

Non-diﬀusion

Ceramic traditions

Cognitive bias

Learning process

Polarization

Social interactions

Ethnoarchaeology

ABSTRACT

In this paper, we address the question of the conditions for persistence of technological boundaries. We use ﬁeld

studies to test the predictions generated by a theoretical model in analytical sociology and examine the micro-

processes at stake in the non-diﬀusion of techniques: to which extent techniques contributes to a sharp dis-

agreement between groups and promote polarization? The ultimate goal is to provide archaeologists with an

empirically tested model to explain spatial distribution of technological clusters and maintenance of technolo-

gical boundaries. Field studies examine ethnographic situations in four countries where social groups using

diﬀerent ceramic techniques for making utilitarian vessels live in close geographical proximity. Two situations

enable us to examine the conditions under which technological boundaries persist, while two others enable us to

analyze, through a boundary-making perspective, how diﬀerences in craft techniques contribute to polarization.

Our data suggest that in a context where diﬀerent techniques are used for diﬀerent types of object there is a

cognitive bias which fosters technological polarization. This cognitive bias develops in the course of interactions

between actors living in close geographical proximity. Polarization increases when technological standards are

used by diﬀerent social groups, thereby favoring negative inﬂuence and persistent technological boundaries.

1. Introduction

In archaeology, the study of technological changes has two sides:

the adoption (invention and spreading) of new techniques and the non-

diﬀusion of techniques (or non-borrowing whether considering the

actor or the technique). Diﬀusion participates in homogenization of

technical traits, whereas non-diﬀusion maintains technical diversity.

The understanding that technical features can diﬀuse across geo-

graphical or social boundaries (as shown in history, geography, so-

ciology, anthropology) makes the persistence of diversity problematic

because movements of people, objects, ideas will tend to homogenize

material culture, in particular when substantial beneﬁts are at stake.

One salient example of this problematic is the non-diﬀusion of the

potter’s wheel in the Mediterranean for hundreds of years and still

today in several parts of the globe even in highly competitive situations

and despite the advantage of the wheel in terms of time manufacturing

(

Roux, 2013; Roux and Jeﬀra, 2016).

Non-diﬀusion of cultural traits is particular visible at boundaries

between cultural clusters distributed in geographical space.

Ethnoarchaeological and anthropological studies have investigated the

social dimension of these boundaries (e.g. Dietler and Herbich, 1998;

Gosselain, 2002, 2000; Hodder, 1985; Latour and Lemonnier, 1994;

Lemonnier, 1993, 1992; Stark, 1998; Stark et al., 2008; Wiessner,

1983

). They have highlighted that technological traditions have a

tendency to superimpose on social boundaries and to be more resistant

to change than easily transmissible traits such as style (shapes and

decor of objects) (Gallay, 2007; Gelbert, 2003; Gosselain, 2000;

Hegmon, 1998; Mayor, 2010; Roux, 2015; Stark et al., 2000). This

diﬀerence between the two cultural traits has been explained in terms

of learning modalities. Techniques are socially learned and culturally

transmitted (e.g.

Kuhn, 2004; O’Brien and Bentley, 2011). Their mas-

tery requires learning through long lasting contacts, generally with

socially close relatives (e.g. Bril, 2002; Gosselain, 2000; Shennan, 2013;

Shennan and Steele, 1999). As a result, the generated population level

pattern links technological traditions with producers’ social group and

are strong identity markers. Stylistic patterns relate not only to

http://dx.doi.org/10.1016/j.jaa.2017.09.004

Received 13 January 2017; Received in revised form 9 September 2017

⁎

Corresponding author.

E-mail address: valentine.roux@cnrs.fr (V. Roux).

Jour nal of Anthropological Archaeol ogy 48 (2017) 320–335

Avai l abl e onl i ne 23 Oct ober 2017

producers’, but also to object and consumers’ history. Stylistic bound-

aries are therefore more ﬂuctuating and their link with social bound-

aries, less marked (David and Kramer, 2001). Since the mechanisms

underlying the diﬀusion and persistence of cultural traits may vary

according to the nature of these traits, each trait should be considered

separately; in this article, only the persistence of technical traits is ex-

amined.

In anthropology and sociology, the co-existence of distinct techno-

logical systems, without exchange of traits, has been mainly explicated

in terms of social aﬃliation/ diﬀerentiation. Under the ‘technological

choice’ approach, at the group level, technical practices are considered

as social facts, made in accordance with social strategies and meanings,

their underlying and embedded representations ﬁtting into a wider

symbolic system (Dobres, 2000; Latour and Lemonnier, 1994;

Lemonnier, 1992, 1993). Consequently, at boundaries, they resist the

homogenizing eﬀects of interactions through the process of aﬃliation/

diﬀerentiation. At the individual level, aﬃliation with the group

practicing the same way has been shown to be the result of social

learning which implies a learner and a tutor, with the outcome of

learners practicing the same way than the tutors (

Bril, 2002; Dietler and

Herbich, 1998; Tehrani and Riede, 2008), having preferentially close

ties with them and therefore developing a sense of aﬃliation with the

tutor’s group (Gosselain, 2011). These studies partially draw on Lave

and Wenger’s work (Lave and Wenger, 1991) on communities of

practice. The latter are made up of individuals who exert or have ex-

erted together their craft, given or not family relationships; shared

learning process acts as a mechanism of aﬃliation to the group. Under

the evolutionary approach, aﬃliation has been explained by the psy-

chological bias according to which, in marked groups (here technolo-

gically), individuals feel close to individuals who are marked the same

way (McElreath et al., 2003, p. 128). As a consequence of aﬃliation,

individuals tend to conform to the norms of the group, while con-

formism exacerbates diﬀerentiation (Henrich and Boyd, 1998;

Moscovici, 1984

). This combined eﬀect of conformism/ diﬀerentiation

would favor non-borrowing of techniques.

Addressing cultural boundaries, the aﬃliation/diﬀerentiation pro-

cess is thus recognized as particularly cogent to explain why non-dif-

fusion of cultural traits may persist and why it should be stronger at

boundary region as suggested for ethnic di

ﬀerentiation (McElreath

al., 2003). In the cultural evolution approach, assumptions are made

using evidence from analytical sociology. Simulation models have

tested the conditions of preservation of cultural diversity. As a ﬁrst step,

diversity has been shown to be preserved when diﬀerences between

cultural groups are too large; in these cases there is a preference to

interact with those who are similar, called homophily (Axelrod, 1997).

Going one step further, Flache and Macy (Flache and Macy, 2011) have

recently proposed that diversity and diﬀerentiation between groups is a

result of polarization through the combined eﬀect of long-range and

short-range ties, the former connecting “local clusters in the network

that are not directly linked otherwise” (Flache and Macy, 2011, p. 147),

the latter describing frequently activated relationships (such as family/

kin ties) (Collar et al., 2015, p. 23). Polarization is deﬁned as the di-

vision of population “into a small number of factions with high internal

consensus and sharp disagreement between them” (Flache and Macy,

2011, p. 149

). Simulation with an agent-based model shows that, in a

small connected world, polarization occurs when “actors who are

connected with a long-range tie are more likely to diﬀer sharply from

each other than are actors connected with a short-range tie” (Flache

and Macy, 2011, p. 172). More precisely, long-range ties bridging

otherwise disconnected clusters can promote cultural integration.

However, these long-range ties can have the opposite eﬀect when

homophily is combined with diﬀerentiation. In this case long-range ties

foster cultural polarization rather than integration. In other words, in

well-connected networks, inter-group interactions can promote cultural

integration but not when the diﬀerences between these groups are

higher than the diﬀerences within the group. In this latter case, the

combination of intra- and inter-group interactions promotes aﬃliation/

diﬀerentiation and polarization.

This sociological model could well explain persistence of technolo-

gical boundaries in the past, technological clusters supposedly corre-

lating with social clusters and therefore technological boundaries with

social boundaries. In these conditions, interactions between technolo-

gically marked groups would favor polarization.

In order to test the assumptions and predictions generated by this

theoretical model, we propose to use ﬁeld studies and examine the

micro-processes at stake in the non-diﬀusion of techniques: to which

extent techniques contributes to a sharp disagreement between groups

and promote polarization? It will be dealt with following a multilevel

analytical framework aimed at relating the individual level (micro-

level, i.e. the role of techniques in diﬀerentiation), the group level

(meso-level, i.e. the polarization phenomenon), and the cultural his-

torical speci

ﬁcs emerging from the later (macro-level, i.e. the persis-

tence

of technological boundaries) (e.g.

Manzo, 2007; Mesoudi, 2007).

The ultimate goal is to provide archaeologists with an empirically

tested model to explain spatial distribution of technological clusters and

maintenance of technological boundaries.

The present research examines ethnographic situations in four

countries (Ethiopia, Cameroon, Ecuador and India) where diﬀerent

social groups live in close geographical proximity and use diﬀerent

ceramic techniques for making utilitarian vessels. Two situations will

enable us to examine the context under which technological boundaries

persist, while two others will enable us to analyze, through a boundary-

making perspective (

Wimmer, 2013), how diﬀerences in craft techni-

ques contribute to polarization. In the four cases, the boundary ceramic

techniques present advantages over each other, thus seriously begging

the question of the non-integrative eﬀect of contacts.

As we shall see, interactions between groups living in close proxi-

mity and using diﬀerent technological standards favor polarization

given “negative” inﬂuence, namely inﬂuence reinforcing diﬀerentiation

rather than integration. Technological standards are here deﬁned as

speciﬁc ways of making speciﬁc ranges of vessels and whose trans-

mission over several generations makes them traditions. A cognitive

bias makes that they contribute directly to a sharp diﬀerentiation be-

tween groups. The main consequence of technological polarization is

the failure of technical traits to spread between technologically marked

groups, even when these groups belong to the same social community

as is the case in Ecuador.

2. Technological boundaries

Analysis of the persistence of technological boundaries ﬁrst implies

understanding how they are generated and how they may be super-

imposed on social boundaries. Two case studies are examined. They

report on situations where ceramic production is in the hands of dif-

ferent groups of potters living in close geographical proximity and

having contact with each other. Two questionnaires were designed. The

ﬁrst one aimed at mapping the pottery traditions (questionnaire on

manufacturing process and range of vessels) of the region, and at re-

cording the identity of the potters, their numbers, the distribution of the

potters’ villages, their interactions, their knowledge of each other’s

techniques, as well as the organization of the production and the dis-

tribution of ceramic crafts. The second questionnaire focused on the

learning and transmission process. Full details are not here reported

systematically, only those related to the issue addressed.

2.1. Transmission and technological boundaries

This section examines how the learning and transmission process of

technological standards participate to their reproduction within one’s

social group, and as a result, create technologically marked clusters that

superimpose on social boundaries.

Two situations of diﬀerent degrees of complexity are considered in

V. Roux et al.

Jour nal of Anthropol ogic al Archaeol ogy 48 (2017) 320–335

321

order to assess the role of matrimonial alliances in the contour of the

technological boundaries:

– One simple situation which involves two ethnic groups and the

transmission of technological standards through women only; it

takes place in Ethiopia.

– One complex situation which involves four ethnic groups and the

transmission of technological standards through men and women; it

takes place in Cameroon.

2.1.1. A “simple” situation: Ethiopia

The Ethiopian case study shows that the learning and transmission

process participates to the superimposition of technological boundaries

on social boundaries despite variability in the modalities of learning.

The investigations have taken place in the Rift Valley, in the

Oromiya region (Fig. 1). This region is inhabited by diﬀerent ethnic

groups (which include the Oromo, Woloyta, Gurage, Sidama, Amhara,

Kambata, and Tigrayan) due to movements of populations in the past

(Freeman and Pankhurst, 2003). Nowadays, pottery is practiced mainly

by two ethnolinguistic groups: the Oromo who speak Afaan Oromo, the

vehicular language in the region, and the Woloyta who, apart from the

vehicular language, speak Woloytania and are migrants having arrived

in the region forty years ago.

The Woloyta and the Oromo are patrilineal (ethnic aﬃliation de-

pends on patrilineal ﬁliation), virilocal (living at husband’s village) and

exogamous communities although in the potter community studied,

inter-ethnic marriages are rare. Pottery production is a specialized

activity conducted by women exclusively on a domestic scale. It is still a

dynamic activity. It represents the main source of income for the

Woloyta. Their rates of production depend on the season: around 200

vessels per month during the dry season against 80 during the wet

season. Among the Oromo, the rates are more variable because pottery

is not their sole economic revenue, though they are generally high,

particularly among divorced women relying mainly on this revenue.

Two places have been investigated. One is Goljoota, a town in-

cluding 31 Woloyta and 4 Oromo families of potters, out of which 20

Woloyta and 2 Oromo have been interviewed. The other place is a

town, Qarsa, where 30 Oromo families of potters live and out of which

Fig. 1. Map of the Oromiya Region (Ethiopia)

locating the two towns surveyed – Goljoota and

Qarsa – in the Arsi and West Arsi Zones.

Fig. 2. Vessels made by the Woloyta and the Oromo. Note- On the left, the Woloyta ﬂat

dish for cooking bread (called Bashe); on the right, the Oromo cafetiere (called Jabana) to

make coﬀee.

V. Roux et al.

Jour nal of Anthropol ogic al Archaeol ogy 48 (2017) 320–335

322

9 have been interviewed. Qarsa is 45 minutes’ drive from Goljoota.

Potters from Qarsa regularly go to Goljoota where a large market takes

place several days a week bringing together potters from diﬀerent

places. The Woloyta sell their production directly to the consumers,

whereas the Oromo sell their production both directly (as in Goljoota)

and indirectly, through middlemen who sell their production in several

marketplaces, some as remote as Addis Abeba (200 km from Qarsa).

Woloyta and Oromo make utilitarian vessels. Among them, there are

types which are common to both groups. However, in the markets fre-

quented by both groups, there are types exclusive to each of them and

particularly successful sales-wise: the Woloyta ﬂatbread dish, and the

Oromo cafetiere (

Fig. 2). The consumers consider each type to be typical of

each ethnic group’sproductionwhereaspottersthemselvesconsidereach

type to be proper to each group, and to be sold as such on the market.

The manufacturing process diﬀers between these vessels (Figs. 3 and

4):

– The making of the Woloyta ﬂatbread dish involves the following

operations: modeling by drawing, burnishing, organic coating (on

the inner face successive coating of a red slip and a mix of water

with banana tree roots, on the outer side coating of the edge with

dung). The dung coating, used by the Woloyta, is highly popular

among consumers because it symbolically gives to the pots the value

attributed to cattle.

– The making of the Oromo tradition cafetiere involves the following

operations: coiling by drawing, polishing with a polish made out of a

red slip mixed with linseed oil and sometimes post ﬁring carbon

coating (smudging). The polish and the carbon coating give to the

cafetiere a shiny black aspect which makes them very attractive to

the consumers. Smudging is found on di ﬀerent types of vessels

(vessels for grilling coﬀee and meat) and is used only by the Oromo.

As a general rule, the Woloyta vessels have the reputation of being

solid, whereas the Oromo vessels have the reputation of being thin and

aesthetic (knowing that slip and smudging aﬀect permeability and

heating eﬀectiveness, but not strength, Longacre et al., 2000).

The learning and transmission process participates directly to the

reproduction of the tradition within each group.

Among the Woloytas, the learning process is pre-marital and the

transmission of the craft takes place within the family. It is mainly

vertical

through the mother to the oﬀspring or, in some exceptional

Fig. 3. The Oromo Chaîne opératoire for making

the cafetiere (called jabana). (A) Starting with a

thick coil. (B) Drawing the walls with upward

discontinuous pressures. (C) Shaping by scraping

with a bone spatula. (D) Closing the bottom. (E)

Adding a coil for making the shoulder of the ca-

fetiere. (F) Shaping the shoulder. (G) Adding a

coil for making the neck. (H) Shaping the neck. (I)

After ﬁnishing the neck, attaching the handle. (J)

Incising decors with a comb on wet surfaces. (K)

Coating the Jabana with a polish made out of red

clay slip mixed with water and vegetable oil or

diesel. (L) Open ﬁring; combustible includes

wood, twigs and dried dung. (M) Coating the

outer surface with carbon by smoking it out; this

is done by plunging the hot cafetiere into straw.

Vertical transmission: from parents to oﬀsprings; oblique transmission: from in-

stitutions or adults others than parents to children; horizontal transmission: within gen-

eration (Berry et al., 2002).

V. Roux et al.

Jour nal of Anthropol ogic al Archaeol ogy 48 (2017) 320–335

323

cases, oblique, from the aunt (when living as a child with her) or the

mother-in-law (when married as a child). According to this process, the

Woloyta girls are taught the Woloyta tradition and only this tradition.

The learning process encompasses several stages. It starts early by

playing and observing informally the adults, followed at the age of 8–11

by their involvement in the preparation of the clay material and the

ﬁnishing operations, and at 12 by their learning of the fashioning

process. Responsibility for sales and ﬁring comes once married. Be-

coming an accomplished artisan is said to require 6–8 years.

Among the Oromos, when the learning process is pre-marital, the

transmission of the craft takes place within the family and is vertical or

oblique. However, half of the interviewed potters were not born in a

potters’ family and did not marry a potter. They learnt the pottery from

family or neighbors, often after marriage, but always from the Oromo

group. More precisely, until recently pottery was practiced exclusively

by one of the Oromo clans and relatively in secret. Therefore, the ones

who decided to learn the craft in order to improve their revenues

started in secret by watching their neighbors, or their relatives who had

learned from watching their neighbors. The pre-marital learning fol-

lows the same steps as described for the Woloyta. The post-marital

learning occurs over shorter periods with the apprenticeship starting

directly at the fashioning stage.

The diﬀerences in the learning process between Woloyta and

Oromo, -whether pre- or post-marital, within or outside the family,

vertical, oblique or horizontal -, illuminate the fact that these do not

play on the transmission process: whatever the modalities of the

learning process, it takes place within the learner’s social group which

favors the reproduction of the tradition within each social group. As a

result, in the Oromiya region, the technological traditions reﬂect the

ethnic aﬃliation of the potter, whether the tutor is from the family or

from the social group in the case of non-potters’ families, technological

boundaries conforming then to social boundaries.

2.1.2. A “complex” situation: Cameroon

In contrast, the Cameroon case study indicates inter-ethnic alliances

between potters’ families. The learning and transmission process ex-

plains that in all cases technological boundaries mirror distinct social

learning networks related, not to the potter’s ethnic group, but to the

tutor’s ethnic ﬁliation.

The investigations have taken place in the Tikar plain (North-West

of Cameroon) (Fig. 5). This region is nowadays inhabited by eleven

diﬀerent ethnic groups given numerous movements of populations in

the past (

Blench, 1993; Williamson, 1971). Pottery is still practiced but

as a minor secondary activity - ceramic production taking place only

4 months a year. It comes mainly as a complement to other economic

activities, chieﬂy agriculture. The demand is for storage pots for liquids

Fig. 4. The Woloyta Chaîne opératoire for making

the ﬂatbread dish (called bashe). (A) Mixing of

clay. (B) Modeling from a lump of clay by hor-

izontal drawing. (C) Addition of clay in the centre

of the dish. (D) Shaping the rim with continuous

pressures. (E) The dish after shaping. (F) Shaving

the outer surface with a metal knife. (G)

Smoothing the inner face with a piece of cloth

soaked in water. (H) Sun drying before pre-ﬁring.

(I) Burnishing the inner face with a small pebble

for facilitating the adherence of coatings. (J) Pre-

ﬁring. (K) Open ﬁring; combustible includes wood

and straw. (L) Coating the inner face with a red

slip. (M) Coating the inner face with a mixture of

water and Abyssinian banana tree roots. (N)

Coating the edge of the dish with animal dung.

V. Roux et al.

Jour nal of Anthropol ogic al Archaeol ogy 48 (2017) 320–335

324

(water) and solids (ﬂour, grains, meat, medicines, and spices), very

rarely for cooking pots. The potters are specialized and carry out their

activity on a domestic scale. They sell or exchange their products di-

rectly to consumers, either in villages or at markets, rarely through

intermediary salespersons.

All in all, 21 potters were interviewed (

Table 1). They live in nine

diﬀerent villages, all located along the main road over a 30 km distance

(

Fig. 5). The potters are between 40 and 77 years old with an average of

61 years. They are the remaining present-day community of the potters

of the region. The rate of production varies between ethnic groups. It is

medium (between 10 and 20 vessels per week, 4 months a year) or low

(less than 10 vessels per week, 4 months a year).

The potters of the Tikar plain belong to six ethnic groups: Yamba,

Mambila, Tikar, Kwanja, Tumu and Bamiléké, though, as we shall see,

they were taught their craft from tutors of Yamba, Mambila, Tikar and

Tumu ﬁliation only. These groups are organized in lineages, each car-

rying the name of the founding ancestor. They are exogamous, patri-

lineal, patrilocal or virilocal (living at father’s village for men, and at

husband’s village for women). From a linguistic point of view, they are

divided between the Ubangian languages (Tikar), belonging to the

Niger-Congo family, and the Bantoïdes languages (Yamba, Bamiléké)

that encompass the Eastern and Grassﬁeld Mambiloïdes languages

(Mambila, Kwanja, Tumu). Alongside these local languages, there is a

lingua franca namely the Fula or Fulfulbé, which had spread by the

mid-19 th century with the arrival of the Fulani herdsmen and the

spread of the Islamic religion.

The main salient technical features diﬀerentiating the main tech-

nological traditions practiced in the Tikar plain are the clay processing

and the forming technique. Four forming techniques are used (de-

scribed in detail in Gosselain, 2002)(Fig. 6).

1. Modeling by pinching, abbreviated Mp (also called “pinching”). A

clay mass is transformed into a hollow volume by punctual dis-

continuous digital pressures.

2. Coiling by pinching, abbreviated Cp (also called “coiling by vertical

superposition”). Coils are superimposed on top of each other and

joined by punctual discontinuous digital pressures.

3. Coiling by spreading, abbreviated Cs (also called “coiling by internal

apposition”). Short coils are placed against the inner wall and

spread in a discontinuous horizontal translation movement.

4. Modeling by drawing, abbreviated Md (also called “drawing from a

lump of clay”). Clay mass is transformed in a hollow volume by

vertical upwards discontinuous inter-digital pressures.

Each forming technique is performed traditionally by a separate

ethnic group. Modeling by pinching is practiced by Yamba men, coiling

Fig. 5. Map of the Tikar plain (Cameroon) lo-

cating the nine villages surveyed.

Table 1

Subjects interviewed in the Tikar Plain (Cameroon) distributed per village, ethnic group

(husband’saﬃliation), technological tradition, age and sex.

Village Ethnic group Tradition Age Subjects Sex

Bankim Kwanja T4 76 ELH-4 F

Bankim Tikar T3 55 ELG-3 F

Bankim Tikar T4 55 MAD-4 F

Bankim Tikar T4 68 MAR-4 F

Bankim Tikar T3 65 MAM-3 F

Bankim Tikar T4 65 VIC-4 F

Dicki Yamba T1 77 MOU-1 M

Kinguizom Mambila T2 59 HAD-2 F

Kinguizom Mambila T2 56 FAT-2 F

Kinguizom Mambila T2 46 MAE-2 F

Lam Yamba T2 50 MAY-2 F

Mbéto Tikar T3 62 YVO-3 F

Mbéto Yamba T1 68 HAM-1 M

Mgbanji Tikar T3 40 CHR-3 F

Mvou Mambila T2 71 LEC-2 F

Nyamboya Mambila T4 63 HAB-4 F

Nyamboya Mambila T4 70 TOB-4 F

Nyamboya Mambila T2 60 MAIM-2 F

Tiké Mambila T2 57 FAN-2 F

Tiké Mambila T2 62 FAA-2 F

Tiké Yamba T2 73 SAB-2 F

V. Roux et al.

Jour nal of Anthropol ogic al Ar chaeol ogy 48 (2017) 320–335

325

by pinching is practiced by Mambila women and men, coiling by

spreading by Tikar women and men, and modeling by drawing is

practiced by Tumu ﬁliation women. These forming techniques are dif-

ferent in terms of skills and time manufacturing. The Cp is the slowest

one and the Md and the Cs the fastest (Fig. 7). The Md is of particular

interest because it is practiced by a higher number of potters than Cs,

which could have been a factor in favor of a positive inﬂuence for its

borrowing. This applies, in particular, to the Mambila who have the

highest rate of production and at the same time the slowest forming

technique (Cp).

Regarding the clay processing, three diﬀerent recipes are used de-

pending also on the ethnic groups: the Mambila use one clay material

only, the Tikar and the Tumu mix two clay materials, and the Yamba

mix clay material with termite mound earth.

Two situations of craft transmission were encountered: either the

craft is practiced within the family and the transmission is vertical,

from parents to oﬀspring (children or grand-children), or the craft is not

practiced in the family and the transmission is horizontal, done at the

post-marital stage by members of the family in-law.

When the craft is practiced within the family the tendency is that

the oﬀspring practices the tradition taught by the tutor who can be the

mother or the father. As an example, Tumu women who perform the

Md, married men from diﬀerent ethnic groups: three married Tikar, one

married a Kwanja, two other ones Mambila. In these diﬀerent cases,

whoever they married, they kept teaching the Md to their children. As a

result, Md is practiced in diﬀerent ethnic groups (Tikar, Kwanja,

Mambila, Bamiléké). In this region at least (Md has a wide expansion in

West Africa and its diﬀusion may have followed diﬀerent modalities

Fig. 6. The four forming techniques used in the Tikar plain.

(A) Modeling by pinching (Mp, Yamba). (B) Coiling by

pinching (Cp, Mambila). (C) Coiling by spreading (Cs,

Tikar). (D) Modeling by drawing (Md, Tumu).

Fig. 7. Manufacturing time per pot depending on forming techniques. The potters made 6 pots of similar size for three diﬀerent shapes. N1 is a common shape. N23 and N19 are two

unusual shapes with diﬀerent ratios H/DM corresponding to diﬀerent degrees of diﬃculty. Coiling by spreading (Cs) and modeling by drawing (Md) are faster than coiling by pinching

(Cp) and modeling by pinching (Mp).

V. Roux et al.

Jour nal of Anthropol ogic al Ar chaeol ogy 48 (2017) 320–335

326

which are not discussed here), Md remains thus strongly tied with the

Tumu ﬁliation as shown by its transmission over more than three

generations.

When there are two tutors, the mother and the father,

originating from two diﬀerent ethnic groups and therefore using dif-

ferent pottery techniques, the two techniques can be taught succes-

sively and one individual can practice two traditions. This is rare but it

can happen during one life time. In this case, however, only one tra-

dition is taught to the next generation, usually the tradition taught ﬁrst.

For example, one woman learnt pottery ﬁrst from her grand-mother

who was Tumu (Md), and second from her grand-father who was Tikar

(Cs) (when she was 15 years old). After marrying a Tikar who was not a

potter, she taught only the Tumu technique (Md) to her children (three

daughters and one son). In the same way, there are cases where one

subject is taught the craft before marriage and arrives in a new family

where another tradition is practiced. The subject then may come to

learn two traditions. However, when it comes to teaching the oﬀspring,

only one tradition is taught. This is a case of a Bamiléké woman whose

mother was Tumu and who was taught the Md. She married a Mambila

who was a potter practicing the Cp; the potter taught her his tradition

so that she could help him. However, she taught her daughter (Mambila

by her father) the Md and when her husband died, she abandoned the

Cp deﬁnitely.

When women originate from families who are not potters or who pass

the craft on to men only, they are taught the craft after marriage by the in-

laws. The technique they practice corresponds then to their in-laws tra-

dition. Thus a Yamba woman had not been taught the craft because,

among the Yamba, pottery-making is a male activity. When she married a

Mambila, she learnt the Cp (Mambila) which she passed on her oﬀspring.

In summary, and in line with Gosselain’s observations (2000), in the

Tikar plain, the technological traditions reﬂect, as a general rule, the

ethnic ﬁliation of the tutor, knowing that the tutor can be the mother,

the grand-mother, the father, the grand-father, or the in-laws in the case

of post-marital learning, and knowing also that the tutor and the learner

can be aﬃliated to diﬀerent ethnic groups (in the case of inter-ethnic

marriages). In general, the tutors teach the technique they were taught

initially, and not the technique they might have learned later in the

course of their life. This tendency points to the importance of the

learning process in the transmission of the tradition. According to this

process the potter progressively integrates the knowledge and the skills

involved in the making of the vessels taught by his (her) tutor. At the

end of the learning process, the technological standards proper to his

(her) tutor’s group are deeply embodied as suggested by the Tumu

potters who keep transmitting the initial technique they learnt despite

marriages with men practicing other techniques and children whose

ethnic aﬃliation depends on the father’saﬃliation.

To conclude, the examples of Cameroon and Ethiopia show well that

because learners learn from tutors, and because tutors are always se-

lected from close social circles, technological traditions are perpetuated

within circumscribed social circles, delimiting technological clusters

made up of socially close individuals.

2.2. Interactions and technological boundaries

The conditions under which technological boundaries persist and tech-

nological traits do not diﬀuse can be deﬁned in terms of connectedness and

interactions between the potters located within and on either side of the

technological boundaries. As we shall see, connectedness within each ethnic

group is strong, while degrees of interactions between ethnic groups vary

depending on regions, Ethiopia or Cameroon.

2.2.1. Ethiopia

In Goljoota, the Woloyta know each other well. They are all related

by blood ties or marriage (they are distributed between four main fa-

milies). Moreover they are neighbors, living in houses located in the

same neighborhood along the same route. In this regard, they represent

a tightly interconnected cluster of potters sharing family relationships

and located in the same area of space. They interact on a daily basis,

though they say that they do not exchange information about their craft

and do not buy from each other.

In Qarsa, the Oromo are less related through blood ties or marriage.

They live in scattered houses. However, they are not isolated. They

have developed friendship relationships and often meet in everyday

spaces (streets and ﬁelds) and

through the meetings of two existing

cooperative (aimed at helping families in need). They buy vessels from

each other (the types they do not make they buy directly at each other’s

homes or on the market). They talk to each other about their craft,

mainly for discussing sources of clay rather than forming or ﬁnishing

techniques. In this regard, even though the Oromo potters are not

connected by family relationships they are still tightly connected by

social and friendly relationships.

Regarding interactions between the Woloyta and the Oromo, the two

groups are in contact on a regular basis in the local markets. The market is a

public space where they meet a few times a week (Fig. 8). They stand at

distinct places. They talk to each other but rather brieﬂy. They mainly talk

about the sales or when asking for change. In some cases, Oromo would

question the Woloyta about their techniques. But as a general rule, they do

not discuss their craft. They can buy from each other. Besides, Woloyta and

Oromo are neighbors in Goljoota. They have friendly relationships, they

visit each other and casually discuss over a cup of coﬀee.

These interactions explain that Woloyta and Oromo know each

other’s productions well as shown also by their capacity to recognize

who manufactured what when we showed them photographs of the

diﬀerent types of vessels made by both groups. When asked about not

copying stylistic and technical traits from each other in order to im-

prove their revenues, Woloyta and Oromo give the same sorts of ex-

planations (technical [our raw material will be inappropriate; our techni-

ques are better for what we do; we do not have the knowledge],

organizational [our working organization is not the same and therefore we

cannot do the same way than the others] and economic [consumer prefer

the original to the copy; it will take too much time to make them]) justifying

the pointlessness of borrowing.

In summary, in the Oromiya region, potters from each ethnic group

entertain close relationships, given family or social ties. In contrast, the

connections between the two ethnic groups are rather loose. However,

they do exist and imply that each group is well acquainted with each

other’s ceramic production and technological features.

2.2.2. Cameroon

In the Tikar plain, the general situation is that all the potters are

connected to each other, either directly or indirectly, thus being aware of

each other’s technological traditions. These interactions are based on fa-

mily ties (blood ties or marriage) or on friendly and social relationships.

Family ties are met within each ethnic group. Friendly and social

relationships are encountered within and between the ethnic groups. In

the latter case, they are visible in the sharing of clay sources and dec-

oration tools. Interestingly, the pattern of connections does not privilege

links at the group level (intra-group interactions) or at the technological

tradition level (intra-tradition interactions) as we shall see with the ex-

ample of the interactions between the potters practicing the Md tech-

nique, - the fastest forming technique and therefore, the technique likely

to be borrowed -, and the other potters using slow forming techniques.

The large majority of the potters practicing the Md live in the same

village, Bankim (4 out of 6). Among them, MAR-Md

has regular direct

As an example, a Tumu woman married a Kwanja (not a potter); she taught her

daughter the Md who also married a Kwanja (not a potter) and who taught her children

the Md in turn (2 daughters and one son).

Abbreviation of the ﬁrst name followed by the abbreviation of the technique used.

V. Roux et al.

Jour nal of Anthropol ogic al Ar chaeol ogy 48 (2017) 320–335

327

interactions with potters using the Md, the Cs and the Cp forming

techniques. Her professional and friendly interactions are mainly with

EHL-Md and MAM-Cs living in Bankim, and FAN-Cp and SAB-Cp living

in Tiké; these occur when these potters visit each other’s homes and

collect the clay material on MAR-Md’s land. MAR-Md is also connected

directly with MAY-Cp, living in Lam, with whom she exchanges dec-

oration tools (“roulettes”) and with MAD-Md, VIC-Md, and ELG-Cs,

three co-wives living in Bankim and who are neighbors. HAB-Md, from

Nyamboya, has also direct interactions with Cp potters among whom

MAY-Cp, collecting clay on her land, and MAIM-Cp who is a neighbor.

Furthermore, indirect interactions between potters using the Md

and the other potters exist through the Cp potters who have direct in-

teractions on the one side with Md potters and, on the other side with

most of the Mp and Cp potters. This is the case of SAB-Cp, from Tiké,

who has direct interactions with MAR-Md and informs most of the

potters about clay sources (FAN-Cp, FAA-Cp and SUZ-Cp her daughters;

FAT-Cp, MAE-Cp and HAD-Cp, from the same household in Konguizom;

LEC-Cp, from Mvou; and MOU-Mp, from Dicki). This is also the case

with Cs potters who have both interactions with Md, Cs and Mp potters

(MAM-Cs, a close friend of MAR-Md, is YVO-Cs’s sister, living in Mbéto

and in close interactions, either for clay or decoration tool sharing, with

HAM-Mp, a neighbor).

In summary, in the Tikar plain, there are strong interactions on the

one hand between the potters belonging to the same ethnic group given

family ties, and on the other hand between the potters belonging to

diﬀerent ethnic groups given friendly and social relationships.

2.3. Cognitive correlations and technological boundaries

In the previous examples, the groups of potters diﬀer socially and in

their craft. In the polarization issue, one of the questions is about the

features participating to both a high consensus within the groups and a

sharp disagreement between them. Social origin is one. Technological

standard is another one that we propose now to examine in the light of

human cognitive limitations. The cognitive limitation considered here

is the one in which individuals use a “covariation principle” to assess

causality. The “covariation principle” states that “if event A accom-

panies outcome B, and if event A is absent when outcome B is absent,

then people tend to attribute A as the cause of B” (Carley, 2001, p. 181;

Kelley, 1967, 1973). According to this principle, if technique A ac-

companies product B, and if technique A is absent when product B is

absent, then people may tend to attribute technique A as the cause of

product B, or product B as the cause of technique A.

As we shall see, this cognition principle has major repercussions,

among which polarization even within a same social group. This will be

ﬁrst exempliﬁed by a case study in Ecuador where ethnographic in-

vestigations were conducted according to the same questionnaires de-

signed for Ethiopia and Cameroon. A second case study in India will

enable us to explain how this “covariation principle” comes into play

when technological standards are used by diﬀerent social groups.

2.3.1. Ecuador

The situation in Ecuador is a case in point because it shows that, all

other things being equal, technological standards have contributed di-

rectly to polarization.

The case study takes place in the canton of San Miguel de Porotos

(south of the Cañar province) (

Fig. 9). This region is inhabited by

potters who are of indigenous origin (cañari) and speak Spanish, but

whose parents or grandparents used to speak Quichua (language of pre-

Hispanic origin related to the Quechua of Peru). Pottery is a specialized

activity conducted by women or men on a domestic scale during half of

the year (

Sjöman, 1992). It is a source of income in addition to agri-

culture.

The potters

’ eleven houses in the canton of San Miguel de Porotos

are

distributed between three hamlets: Pacchapamba (5 potters), Chico

Ingapirca (3 potters) and San Juan Bosco (3 potters). Nine potters were

interviewed (Table 2). In each hamlet transmission is vertical, from

parents to their oﬀspring.

The three hamlets are separated by only a few hundred meters

(10–20 min walking distance). Each hamlet is home to families who

have no kinship ties; whereas members of a same family or members

with strong ties can be living in diﬀerent hamlets (example of one fa-

mily whose members live in San Juan Bosco and in Chico Ingapirca;

example of a godfather living in Chico Ingapirca and his godchild in

Pacchapamba).

Regarding ceramic production, the potters from Pacchapamba and

San Juan Bosco are specialized in the manufacture of utilitarian wares

(cooking pots, jars, tortilla dishes), whereas the potters from Chico

Fig. 8. Woloyta and Oromo potters meeting regularly in

market places (market of Goljoota – Arsi Zone, Oromiya

Region). (A) In the foreground the Oromo Potters, in the

background the Woloyta Potters. (B) In the foreground the

Woloyta Potters, in the background the Oromo Potters.

V. Roux et al.

Jour nal of Anthropol ogic al Ar chaeol ogy 48 (2017) 320–335

328

Ingapirca are specialized in ornamental or specialized wares (birthday

pots, plates for ﬂower pots). The annual production is slightly lower in

Pacchapamba (550 items a year against 750 in San Juan Bosco and

Chico Ingapirca).

Now, in the canton of San Miguel de Porotos three open ﬁ ring

structures coexist: open ﬁring in Pacchapamba, walled ﬁring in San

Juan Bosco and kiln in Chico Ingapirca (

Fig. 10). The kiln presents

objective advantages when considered in terms of thermal performance

characteristics - maximum temperature achieved, temperature varia-

tion within the kiln load and control of the rate of temperature increase

– and in terms of control of ﬁring atmosphere - protecting the load from

gusts of wind and reducing the incidence of ﬁre clouds in oxidized

pottery kilns (Arnold, 1991; Pool, 2000; Rice, 1984). The kiln presents

also advantages in terms of combustible and working hours (Table 3).

The presence of the kiln in Chico Ingapirca has been attested for a

long time. There is no memory of its introduction. On the basis of oral

narratives the kiln was already used in the sixties by the three house-

holds of potters who were then in activity.

The potters of Pacchapamba and San Juan Bosco never borrowed

the kiln because at ﬁrst they say that the open or the walled ﬁring is

their method, that it works well and therefore that they have no interest

in changing it. But they also give other reasons (technical, organiza-

tional and economic) among which the more frequent is that the kiln is

appropriate only for decorative items and inappropriate for utilitarian

vessels.

As potters say: “The kiln is used by Francisco whose production is

famous and diﬀerent from ours; we do not make the same items and

therefore we do not need the kiln” (a potter from San Juan Bosco); or

“Kilns are for potters who make special items like the “decorated ﬁg-

urines” made by Francisco or the “birthday vessels” made by Edelina”

(a potter from Pacchapamba); or “Our pots are well ﬁred, with a nice

color and sell well” (a potter from San Juan Bosco).

With reference to the “co-variation principle”, these explanations

given for the non-borrowing of the kiln highlight that the potters make

a strong cognitive correlation between the ﬁring structures and the

products to be ﬁred: they link the kiln with the ﬁring of decorative

items and no other items than these ones. Inter-marriages between

hamlets are particularly telling examples of this cognitive correlation.

Thus, two potters of San Juan Bosco married men of Chico Ingapirca.

When they moved there, they left behind the walled ﬁring and adopted

the kiln along with the manufacture of small decorative items under the

guidance of the mother-in-law. They never made any more big vessels

because, as they say, the kiln is for

ﬁring small items only. Another

example

is provided by a woman from Pacchapamba who married a

man from Chico Ingapirca and then adopted the kiln along with the

manufacture of small decorative items. When her husband died, she

returned to Pacchapamba, gave up the kiln and the small decorative

item production, and used open ﬁring again which she liked better

because, as she said, it allows her to ﬁre bigger vessels.

As a result of this cognitive correlation, technological boundaries

were formed diﬀerentiating the three hamlets: the kiln is found only in

Chico Ingapirca and is used to ﬁre decorative items only; open ﬁring

and walled ﬁring are found respectively in Pacchapamba and San Juan

Bosco and are used to ﬁre utilitarian vessels only.

In summary, in San Miguel de Porotos, at ﬁrst the three hamlets

were making the same types of vessels that they ﬁre using two ﬁring

structures (the story of the walled ﬁring is not known). Then, in the

course of adopting the kiln the kiln users specialized in the production

of speciﬁc types of objects. As a consequence of the principle of ra-

tionality favoring correlations between techniques and ﬁnished pro-

duct, groups with no family tie got formed, each of them using diﬀerent

technological standards. These diﬀerences in technological standards

have divided the potters into three groups, fostering polarization to the

extent that on the one hand the kiln has not been borrowed despite its

advantages over the open and walled ﬁrings, on the other hand potters

of San Miguel do not feel they belong to the same community, as shown

Fig. 9. Location map of San Miguel de Porotos (Ecuador).

Table 2

Subjects interviewed in the three hamlets of the canton of San Miguel de Porotos

(Ecuador).

Hamlet Family Subjects

Pacchapamba Pérez Zimbaña María Pérez

María Zimbaña (María’s daughter)

Fernández Margarita Fernández

María Fernández (Margarita’s sister)

Chico Ingapirca Suquinagua Edelina Suquinagua

Inga Francisco Inga (Godfather of

Margarita’s son)

San Juan Bosco Maríana Inga (Francisco’s sister)

Tenenpohuay

González

Teolinda Tenenpohuay

Remigio Gonzalez (Teolinda’s son-in-

law)

V. Roux et al.

Jour nal of Anthropol ogic al Ar chaeol ogy 48 (2017) 320–335

329

by their refusal to form an association which required a minimum of ten

potters and which would have helped them in increasing their sales.

2.3.2. India

The case study in India (North West India, Bulandshar district,

Fig. 11) illustrates not only the intrinsic value of technological stan-

dards in the polarization process, but also how this process integrates

progressively social diﬀerentiation. The principle of rationality corre-

lates, in this case, potter’s tools and preforming techniques and potter’s

tools and social groups.

The ethnographic situation has already been described elsewhere

(

Roux, 2013). The craft production is distributed between Muslim and

Hindu potters, two endogamous groups (Kramer, 1991) who live in the

same villages or in diﬀerent villages. Each of them uses diﬀerent

technological systems which present strong diﬀerences in terms of cost

and beneﬁt. The Muslim potters use the foot wheel, the kiln and glaze;

the Hindu potters use the ﬂy wheel, the open ﬁring and do not glaze

their vessels. Like the kiln as opposed to open ﬁring, the foot wheel

presents clear advantages over the ﬂy wheel. The foot wheel is operated

with the foot and enables the control of the speed all throughout the

forming process. In contrast, the ﬂy wheel is launched with a stick and

the speed of the ﬂy wheel cannot be controlled. It decreases progres-

sively once launched. As for the glazing technique, it is mainly applied

on cooking pots, which are then less permeable and have better thermal

properties than the unglazed ones. However, neither the kiln, nor the

foot wheel and glaze have ever been borrowed by the Hindu commu-

nity, except for a few cases not discussed here. This is all the more

noticeable since Hindu potters interact regularly with Muslim potters in

a professional (sharing customer orders) and friendly framework (vis-

iting each other), and therefore know well the Muslim techniques (they

could draw the Muslim kiln and explain how it works).

In order to investigate how technological diﬀerences enhanced po-

larization we interviewed 50 Hindu and Muslim male and female

teenagers of diﬀerent age groups (from 8 to 19 years old) living in nine

diﬀerent villages (30 male teenager potters including 17 Hindus and 13

Muslims; 20 female teenager potters, including 9 Hindus and 11

Muslims, Table 4). The interviews included questions about manu-

facturing process, properties of the techniques and tools, technological

system used by the other group, advantages or disadvantages of each

technological system, and at last who uses which system. For this

purpose, the subjects were shown photographs of each community’s

Fig. 10. Firing structures in San Miguel de Porotos. (A)

Open ﬁring. (B) Vertical updraft kiln. (C) Walled ﬁring.

Table 3

Comparative costs between kiln, open ﬁring and walled ﬁring in Ecuador (in 2014).

Capacity (100

pots)

Fuel Working hours to

prepare the ﬁring

Cost of

the kiln

Kiln 1 bundle of wood and 6

muleloads of straw

2 410 $

Open ﬁring 7 muleloads of wood and 20

muleloads of straw/

eucalyptus leaf

4 –

Walled ﬁring 5 muleloads of wood and 3

muleloads of straw

46$

V. Roux et al.

Jour nal of Anthropol ogic al Ar chaeol ogy 48 (2017) 320–335

330

instruments so that they could make precise comments (Fig. 12). We

will focus on the comments given about the potter’s wheels. Two main

results obtain: (a) cognitive correlations evolve depending on one’s

experience and contribute progressively to link technique and social

group; (b) interactions between actors using diﬀerent techniques are

necessary for correlation between technique and social group.

The ﬁrst result reports on the correlation between instruments and

ﬁnished products, and between instruments and preforming techniques.

We asked the male potters (30) about the diﬀerences between the ﬂy

wheel and the foot wheel. Out of 30 subjects (13 Muslims and 17

Hindus), 24 knew about both instruments, 13 Muslims and 11 Hindus.

When they knew, they could attribute each instrument to each com-

munity.

In the Muslim community, when asked about the advantages of the

foot wheel over the ﬂy wheel, 6 out of 13 indicated that the foot wheel

is good for big vessels, and the ﬂy wheel for small vessels only. They

include three boys aged 10–11 from Shirkapur and Jahangirabad where

both communities are settled and the Hindu potters specialized in the

making of small vessels; and three teenagers aged between 15 and 18

from Gesupur, Aurangabad and Mirzapur. The two latter villages are

inhabited by Muslim potters only, but the teenagers had seen Hindu

potters from Bulansdshar (a nearby town) specialized also in the

making of small vessels. The Muslim teenagers who did not correlate

the wheel with speciﬁc ﬁnished products considered the ﬂy wheel not

as good as the foot wheel because it could not be operated with the foot,

it needed to be restarted every time the rotation slows down or it might

be tiring for the arms.

In the Hindu community (11), one potter (17 years old) correlated

the ﬂy wheel with small vessels and the foot wheel with big vessels; he

is from Shirkapur where the Muslim potters make bigger vessels than

the Hindus. Another (18 years old) correlated the foot wheel with the

throwing of big lumps of clay. He is from Nagara where the Hindu

potters make a wide range of vessels and are not specialized in small

vessels. These two Hindu potters have never seen the foot wheel but

have heard of it. Seven others, who are younger, estimated that the foot

wheel was not as good as theirs because it was operated with the foot

which might be tiring. Two older ones estimated that the use of the foot

wheel would bring them bad luck because the wheel is the chakra of

Vishnu and should not be touched with the foot.

In addition, we asked a few Hindu senior potters why they did not

adopt the foot wheel. The majority of them outlined that the ﬂywheel is

very well suited for what they produce, namely thick vessels whose

body is then paddled and can reach up to 50 liters capacity. In contrast

they said that the Muslim potters throw thinner vessels whose body is

not paddled but whose bottom is pounded, and therefore require a more

stable wheel. In other words, senior potters correlate the type of wheel

and the ﬁnished products under the rational explanation that the ﬂy

wheel is good for thick vessels and the foot wheel is good for thin

vessels because the ﬂy wheel is less stable and therefore less appro-

priate to throw thin vessels. There are also senior potters who invoked

that foot wheels were not good instruments because wheels should not

be touched with the foot.

From these answers, it can be seen that the young teenagers explain

the diﬀerences of properties of the two instruments either in terms of

the skills they require (with the foot, with the arm), or in terms of the

type of vessels made by each community. The answers given by the

Hindu seniors correlate the instruments with the preforming techniques

(paddling versus pounding): in their words, paddling require thick

vessels whose throwing needs less stability than thin pounded vessels.

Paddling is a preforming technique used by the Hindus, pounding by

the Muslims. In this regard, the fashioning chaînes opératoires di

ﬀer-

entiate both

groups with the result that, given the principle of ration-

ality, the two groups perceive the properties of their instruments in the

light of these chaînes opératoires, and only these.

These results highlight that perception of the properties of the tools

is constructed in the course of learning but is not taught as shown by the

teenagers’ answers which focus on correlations between tools and body

postures or between tools and ﬁnished products. Only seniors propose

correlations between tools and chaînes opératoires. Only the older

teenagers and seniors attribute a symbolic value to their wheel. This

shows on the one hand how the cognitive bias is exerted depending on

one’s experience and one’s age, how then it inﬂuences the perception of

a technique’s properties and how, eventually, it disfavors transfers of

techniques from one technological system to the other. On the other

hand, it shows that correlations between tools/techniques and social

groups is not straightforward; the explicit correlation is mainly between

ﬁnished products and social groups (they are the Hindus versus the

Muslim who make small versus big vessels); in this regard, it does not

play directly on the perception of the properties of the techniques; it

comes after to enhance diﬀerentiation between both technological

systems (“in our group, we do like that” as opposed to “in their group,

they do like that”).

Fig. 11. Location map of the villages surveyed in

Bulandshar district (Uttar Pradesh, NW India).

V. Roux et al.

Jour nal of Anthropol ogic al Ar chaeol ogy 48 (2017) 320–335

331

The second result indicates that the individuals who do not have

interactions with the other community do not know about the other’s

technological system. This is the case of most of the girls who know

about the whole manufacturing process which takes place within the

household and in which they participate (helping in clay preparation,

painting and ﬁring the vessels), but who are not used to leaving their

neighbourhood and therefore are not exposed to other ways of working

(

Fig. 13). This is also the case of six Hindu potters (out of 17; they are

10, 12, 15, 17 and 19 years old) who are not involved in pottery sales.

Consequently, they do not know who uses what and therefore do not

make any correlation between the instruments and the ﬁnished pro-

ducts or the social groups. In contrast, all the Muslim potters knew

about both technological systems because, unlike the Hindus, they have

all been involved in the craft since childhood, accompanying their fa-

ther to sell or buy the pots, preferring, as a general rule, to learn pottery

rather than to go to school. In brief, it is not enough to belong to the

same village and have parents socializing with each other’s commu-

nities, one also has to interact, directly or indirectly, for diﬀerentiation

to develop.

3. Discussion

We started by raising the question of persisting technological

boundaries and diversity on the basis of ethnographic cases. These cases

include social groups who use techniques presenting diﬀerent proper-

ties in terms of cost and beneﬁts.

3.1. Context of persisting technological boundaries

At ﬁrst, data from Ethiopia and Cameroon illustrate how technolo-

gical boundaries get formed and are superimposed on social bound-

aries. In both cases the learning and transmission process explains that

craft learning requires a tutor, and that the tutor is always selected

within one’s social group. Comparison between both countries shows

that this process of the tutor’s selection within one’s group occurs

whatever the learning modalities (pre- vs. post-marital), the potters’

gender (male vs. female) or the matrimonial alliances (alliances be-

tween technologically marked groups). As a result, technological tra-

ditions reﬂect social groups and technological boundaries conform to

social boundaries, i.e. the tutor’s social group boundaries, or put it

another way the learning network boundaries. The conformity between

technological boundaries and learning networks has been acknowl-

edged for some time now and explains why in archaeology technolo-

gical traditions are cogent proxies to identify past social groups (e.g.

Bowser and Patton, 2008; Dietler and Herbich, 1998; Gosselain, 2000;

Hegmon, 1998; Herbich, 1987; Longacre, 1991; Stark et al., 2000) and

why they can be regarded as coherent units amenable to evolutionary

analyses (e.g., implying a phylogenetic link between cultural groups)

(Manem, 2008; Shennan, 2013; Shennan et al., 2015).

As a second step, the context into which technological boundaries

persist has been deﬁned in terms of connectedness and interactions

between the potters located within and on either side of the technolo-

gical boundaries. Data show that technological boundaries persist in a

context where the diﬀerent groups interact on a regular basis, no matter

the nature of the interactions, distant (Ethiopia) or close (Cameroon), or

the nature of the ties within each group, social (Ethiopia) or family

(Cameroon) relationships. We add that the persistence of technological

boundaries takes place in two diﬀerent demographic situations. In

Ethiopia, pottery activity is still very alive and potters are numerous.

On the contrary, in Cameroon, pottery activity is dying and potters are

very few in number.

3.2. Technological standards as diﬀ

erentiation markers

As a

third step, we took a boundary-making perspective in order to

assess better the polarizing eﬀect of interactions between groups using

Table 4

Subjects interviewed in India distributed per village, caste, sex and age.

Village Caste Sex Age

Aurangabad MK M 16

Aurangabad MK M 12

Aurangabad MK M 10

Dibai MK M 16

Dibai MK M 12

Dibai PR M 13

Dibai PR M 12

Dibai PR M 10

Gesupur MK F 19

Gesupur MK F 9

Gesupur MK M 15

Gesupur MK M 8

Gesupur PR F 12

Gesupur PR M 12

Gulaothi MK F 15

Gulaothi MK F 10

Gulaothi MK M 14

Jahangirabad MK F 19

Jahangirabad MK F 15

Jahangirabad MK F 14

Jahangirabad MK F 11

Jahangirabad MK M 11

Jahangirabad PR F 15

Jahangirabad PR M 15

Jahangirabad PR M 14

Jahangirabad PR M 11

Jahangirabad PR M 9

Mirzapur MK F 10

Mirzapur MK M 18

Nagada PR F 15

Nagada PR M 18

Shikarpur MK M 10

Shikarpur PR F 19

Shikarpur PR F 15

Shikarpur PR M 19

Shirkapur PR F 18

Shirkapur PR F 14

Shirkapur PR F 10

Shirkapur PR F 9

Shirkapur PR M 17

Siyana PR F 9

Siyana PR M 15

Siyana PR M 10

Fig. 12. Young potters commenting on photographs showing Muslim and Hindu kilns.

V. Roux et al.

Jour nal of Anthropol ogic al Ar chaeol ogy 48 (2017) 320–335

332

diﬀerent technological standards.

Our data show that technological standards participate in polar-

ization given ﬁrst a cognitive bias correlating techniques and ﬁnished

products. In Ecuador, this cognitive bias makes that the kiln is attrib-

uted to small decorative items and the open ﬁring to big vessels. The

case of Ecuador is not an isolated one. In India, it has been observed too

that the covariation principle linking ﬁnished products and kiln inﬂu-

ences the potters’ perception of the functional properties of the ﬁring

structures. Thus, in the region of Jodhpur, in the town of Pachpadra

inhabited by Hindu potters, the kiln was introduced in 1998 to make

small glazed items and capture a new market (Roux and Gabriellini,

2016). The kiln was built in the house of one of the potters involved in

the project. But the latter failed. It took then seven years for the potter

to think enlarging the kiln and using it for ﬁring his own production

composed of plain utilitarian vessels. For seven years, he perceived the

kiln as only good for ﬁring small glazed items. The inﬂuence of the

ﬁnished product in the perception of the properties of the kiln has also

been observed in two other Indian places (Roux and Lara, 2016 ). In one

of them (Jahangirabad, Bulandshar dist.), a few Hindu families use the

kiln for big vessels, while ﬁring small vessels in open ﬁrings. The Hindu

potter who ﬁrst adopted the kiln borrowed it from a Muslim friend who

was making only big vessels. As a consequence, he considered that the

kiln was unsuitable to ﬁre small vessels. The other Hindu families living

in the same town (same sub-caste) are not interested in the kiln because

they say they make only small vessels and that small vessels cannot be

ﬁred in kilns. In a nearby village, Siyana, ﬁve Hindu families use also a

kiln. But, unlike in Jahangirabad, the kiln is small and used for small

vessels only (the history of the adoption of the kiln is diﬀerent); the big

vessels are ﬁred in open ﬁrings. The other Hindu families (same sub-

caste) use open ﬁ rings only, explaining that they make big vessels and

that big vessels cannot be ﬁred in kilns. In summary, in Northern India,

the same type of vertical updraft kiln is perceived as good for ﬁring big

or small vessels depending on the type of vessels ﬁred in the kiln which

was originally adopted. This relative perception of the kiln properties

makes that potters do not transfer the new ﬁring technique even to their

own production.

The ﬁeld experiments in India explain how this cognitive bias is

constructed at the individual level in the course of learning: young

potters develop progressively, by reference to their experience, their

own understanding of the properties of the techniques in relationship

with the properties of the ﬁnished products (I use the kiln because I make

decorative items

; I use the ﬂy

wheel because I throw big pots). They use

obvious indicators of what they think should be causing their practice.

This causality process is studied in social psychology and has been well

described within the framework of the attribution theory which is about

how people make causal explanations (

Kelley, 1967, 1973). These

causal explanations prevent individuals from transferring their skills to

new situations because they do not understand the properties of the

task. In experimental psychology, capacities of transfer of skills to new

situations is used to characterize degrees of expertise (e.g. Bril et al.,

2010).

In Ecuador, where two technological standards are used, this cog-

nitive bias has promoted polarization despite absence of social diﬀer-

entiation - individuals belong to the same social group - and close ties

between the potters. Consequently, the potter community is distributed

between those who use the kiln and make decorative items and those

who use open ﬁrings and make utilitarian vessels. It perceives itself as

composed of two technologically marked groups and for this reason,

refuses to form an association despite economic advantages.

Diﬀerentiation created by technological standards increases when

they are in the hands of diﬀerent social groups. The ﬁeld experiments in

India show how young potters correlate progressively techniques and

social groups: correlation between techniques and ﬁnished products

gets combined with the correlation linking ﬁnished products and social

groups. The result is twofold: a strong link between techniques and

social group (technique t is correlated to product p, product p to group

Fig. 13. Number of potters distributed per gender (F, M) and between those

who know about the foot-wheel (1) and those who don't (0).

V. Roux et al.

Jour nal of Anthropol ogic al Ar chaeol ogy 48 (2017) 320–335

333

G, then technique t to group G), and a sharp disagreement between the

social groups - not the same technique, not the same ﬁnished products,

not the same social group – as seen in Ethiopia, Cameroon and India.

This disagreement promotes diﬀerentiation while, at the same time,

aﬃliation is reinforced in the course of learning, communities of

practice and cultural process of identity construction shaping each

other (

Lave, 1991, p. 80).

Interactions are of a major importance in the polarization process.

Indeed, interactions are required for actors to correlate techniques and

social groups as shown by Indian teenagers. These interactions can be of

diﬀerent nature, as exempliﬁed by the four case studies, but they al-

ways take place between actors who live in close geographical proxi-

mity and who in this regard are strongly connected.

3.3. Conditions for persistence of technological boundaries

The four case studies show that independently of the organization of

production, interactions between actors belonging to diﬀerent well

connected groups are not suﬃcient for positive inﬂuence when the

diﬀerences between the groups are higher than within the groups. The

technological standards are strong markers of diﬀerences, creating

inter-group disagreement (even within a same social group). As a

consequence, these technological standards contribute directly to en-

hance and add diﬀerentiation between interacting communities, espe-

cially if those belong to diﬀerent social groups. It explains that persis-

tence of technological boundaries has been systematically observed in

situations where diﬀerent communities live in close geographical

proximity and use diﬀerent technological standards. Whatever the

cultural context and the local social situation, these interactions have

promoted negative inﬂuence.

Lastly, let us note that there may be exceptions to this regularity.

One example takes place in Cameroon where one Mambila potter (FAN-

Cp, whose mother SAB-Cp taught her Cp) adopted recently Md, after

visiting MAR-Md and ELH-Md and practicing with them, from the

preparation of the clay paste to the forming of the vessel. As a result,

there has been a positive inﬂuence and an integrative eﬀect. This po-

sitive inﬂuence reveals:

(a) The role of the cognitive bias correlating here clay material and

fashioning technique; the potter did not borrow only the forming

technique, but also the clay material, linking both features which,

in the case of Cameroon, appear to have played a role in the dif-

ferentiation process between the four groups, each of them using

diﬀerent clay materials and recipes.

(b) The role of social learning. Borrowing new practices has implied

social learning between individuals from the Md and Cp social

groups through common practice, from the preparation of the clay

until the forming of the pot. In a situation of technological polar-

ization, such a situation is rare. It goes with exceptional close re-

lationships. This has been observed in Cameroon, and in India, in

the town of Jahangirabad (Bulandshar dist.) where a Hindu potter

and a Muslim potter had developed exceptional friendship followed

by a unique case of borrowing of the kiln by the Hindu potter.

4. Conclusions

Field studies have enabled us to test empirically the predictions of

the experimental model elaborated by

Flache and Macy (2011)

while

examining the

micro-processes at play in the non-borrowing of tech-

niques. They highlight that interactions between groups using diﬀerent

technological standards and living in close geographical proximity re-

present conditions in favor of the persistence of technological bound-

aries. Technological standards add to diﬀerentiation between groups

due to a cognitive bias developed at the individual level. Polarization

along with negative inﬂuence increases when the technologically

marked groups belong to diﬀerent social groups. The former do not

correspond necessarily to distinctive, bounded, homogeneous social

groups; their nature can be expressed by their degree of “groupness”,a

term coined by Brubaker and Cooper while debating identity (Brubaker

and Cooper, 2000, p. 20). By using the term “groupness”, the point is to

take advantage of:

“an analytical idiom sensitive to the multiple forms and degrees of

commonality and connectedness, and to the widely varying ways in

which actors (…) attribute meaning and signiﬁcance to them. This

will enable us to distinguish instances of strongly binding, vehe-

mently felt groupness from more loosely structured, weakly con-

straining forms of aﬃnity and aﬃliation.”

Brubaker and Cooper, 2000, p. 21

Our four case studies illustrate diﬀerent degrees of groupness, from

a weakly constraining form of aﬃnity as in Ecuador, to a strongly sense

of groupness expressed by explicit reference to the other group as in

Ethiopia, Cameroon and India. The sense of groupness is enhanced

when technological boundaries overlap social boundaries, strengthened

then by greater diﬀerentiation between groups. In archaeology con-

textual data should help to correlate technological clusters with dif-

ferent degrees of groupness (e.g. Parkinson, 2006).

Polarization and negative inﬂuence can last for centuries, as ex-

empliﬁed by the long co-existence of wheel made and hand-made

ceramics around the Mediterranean basin (

Gauss et al., 2016).

Acknowledgements

This work was supported by the ANR (The French National Agency

for Research) within the framework of the program CULT

(Metamorphosis of societies –“Emergences and evolution of cultures

and cultural phenomena”), project DIFFCERAM (Dynamics of spreading

of ceramic techniques and style: actualist comparative data and agent-

based modeling) (n°ANR-12-CULT-0001-01).

The data were collected, analyzed and illustrated: in Cameroon by

E. Zangato and G. de Saulieu; in Ecuador, by C. Lara; in Ethiopia by J.

Cauliez, C. Manen and A.-L. Goujon; in India by B. Bril and V. Roux. The

paper has been written by V. Roux.

In Cameroon, we are most grateful to the students Anselme Ossima,

François Ngouoh and Ruth Wa ﬀa, from the University of Yaounde 1,

and Guillaume Delebarre, from the University of Paris Nanterre, who

took part in the ﬁeldwork during the seasons of 2013 and 2014. We

thank also the IRD Yaounde, the department of archaeology of Yaounde

1, the sub-prefect of Bankim and the traditional Tikar chief of Bankim

for their help in the ﬁ

eld. In Ecuador, we thank Tamara Landívar,

curator of

the ethnographic department of the ethnographic museum of

Pumapungo in Cuenca, Ministry of Culture, for her help and her interest

in the project. In Ethiopia, the project was supported by the Fyssen

Foundation and the UMR 5608 – Traces (CNRS, University of Toulouse

Jean-Jaurès, France) and beneﬁted from the help of the Authority for

Research and Conservation of Cultural Heritage (The Federal

Democratic Republic of Ethiopia, Ministry of Culture and Tourism,

Addis Ababa, Ethiopia) and the French Center of Ethiopian Studies

(IFRE 23 and USR 3137, Addis Ababa, Ethiopia). The data have been

collected with the help of Vincent Ard (CNRS, UMR 5608) and

Joséphine Caro (PhD candidate, UMR 5608). In India, the project was

supported by the UMR 7055 – Préhistoire & Technologie (CNRS,

University of Paris Nanterre, France). We thank all the potters in

Cameroon, Ecuador, Ethiopia and India for their availability and their

unfailing kindness. We thank also two anonymous reviewers for their

constructive and useful feedback.

References

Arnold, P.J.I., 1991. Domestic Ceramic Production and Spatial Organization: A Mexican

Case Study in Ethnoarchaeology. Cambridge University Press, Cambridge.

Axelrod, R., 1997. The dissemination of culture a model with local convergence and

V. Roux et al.

Jour nal of Anthropol ogic al Ar chaeol ogy 48 (2017) 320–335

334

global polarization. J. Conﬂ. Resolut. 41, 203–226.

Berry, J.W., Poortinga, Y.H., Segall, M.H., Dasen, P.R., 2002. Acculturation and inter-

cultural relations. Cross-Cult. Psychol. 345–383

Blench, R., 1993. New developments in the classiﬁcation of Bantu languages and their

historical implications. In: Barreteau, D., Graﬀenried, C.V. (Eds.), Datation et

Chronologie dans Le Bassin du Lac Tchad. ORSTOM, Paris, pp. 147–160.

Bowser, J.B., Patton, J.Q., 2008. Learning and transmission of pottery style: women’s life

histories and communities of practice in the Ecuadorian Amazon. In: Stark, M.T.,

Bowser, B.J., Horne, L. (Eds.), Cultural Transmission and Material Culture. Breaking

Down Boundaries. The University of Arizona Press, Tucson, pp. 105–129.

Bril, B., 2002. L’apprentissage de gestes techniques: ordre de contraintes et variations

culturelles. In: Bril, B., Roux, V. (Eds.), Le Geste Technique. Réﬂexions

Méthodologiques et Anthropologiques, Technologies/Idéologies/Pratiques. Editions

érès, Ramonville Saint-Agne, pp. 113–150.

Bril, B., Rein, R., Nonaka, T., Wenban-Smith, F., Dietrich, G., 2010. The role of expertise

in tool use: skill diﬀerences in functional action adaptations to task constraints. J.

Exp. Psychol. Hum. Percept. Perform. 36, 825–839.

Brubaker, R., Cooper, F., 2000. Beyond “Identity”. Theory Soc. 29, 1–47.

Carley, K.M., 2001. Learning and using new ideas: a sociocognitive perspective. In:

Casterline, J.B. (Ed.), Diﬀusion Processes and Fertility Transition: Selected

Perspectives. National Academy Press, Washington, D.C., pp. 179–207

Collar, A., Coward, F., Brughmans, T., Mills, B.J., 2015. Networks in archaeology: phe-

nomena, abstraction, representation. J. Archaeol. Method Theory 22, 1–32.

David, N., Kramer, C., 2001. Ethnoarchaeology in Action. Cambridge University Press,

Cambridge.

Dietler, M., Herbich, I., 1998. Habitus, techniques, style: an integrated approach to the

social understanding of material culture and boundaries. In: The Archaeology of

Social Boundaries. Smithsonian Institution Press, Washington, DC and London, pp.

232–269.

Dobres, M.A., 2000. Technology and Social Agency. Blackwell Publishers, Oxford.

Flache, A., Macy, M.W., 2011. Small worlds and cultural polarization. J. Math. Sociol. 35,

146–176

Freeman, D., Pankhurst, A., 2003. The Excluded Minorities of Ethiopia. London.

Gallay, A., 2007. The decorated marriage jars of the inner delta of the Niger (Mali): essay

of archaeological demarcation of an ethnic territory. www.thearkeotekjournal.org

1 (1).

Gauss, W., Klebinder-Gauss, G., von Rüden, C. (Eds.), 2016. The Transmission of

Technical Knowledge in the Production of Ancient Mediterranean Pottery.

Österreichisches Archäologisches Institut, Wien.

Gelbert, A., 2003. Traditions céramiques et emprunts techniques dans la vallée du ﬂeuve

Sénégal. Editions de la Maison des sciences de l’

homme, Editions Epistèmes, Paris.

Gosselain, O.,

2002. Poteries du Cameroun méridional styles techniques et rapports à

l’identité, Monographie du CRA. CNRS Editions, Paris

Gosselain, O., 2000. Materializing identities: an African perspective. J. Archaeol. Method

Theory 7, 187–217.

Gosselain, O.P., 2011. Fine if I do, ﬁne if I don’t. Dynamics of technical knowledge in Sub-

Saharan Africa. In: Roberts, B.W., Vander Linden, M. (Eds.), Investigating

Archaeological Cultures: Material Culture, Variability, and Transmission. Springer

Science+Business Media, New York, London, pp. 211–227.

Hegmon, M., 1998. Technology, style, and social practice: archaeological approaches. In:

Stark, M.T. (Ed.), The Archaeology of Social Boundaries. Smithsonian University

Press, Washington D.C., pp. 264–279.

Henrich, J., Boyd, R., 1998. The evolution of conformist transmission and the emergence

of between-group diﬀerences. Evol. Hum. Behav. 19, 215–242.

Herbich, I., 1987. Learning patterns, potter interaction and ceramic style among the Luo

of Kenya. Afr. Archaeol. Rev. 5, 109–136.

Hodder, I., 1985. Boundaries as strategies: an ethnoarchaeological study. In: The

Archaeology of Frontiers and Boundaries. Academic Press, New York.

Kelley, H.H., 1973. The processes of causal attribution. Am. Psychol. 28, 107.

Kelley, H.H., 1967. Attribution theory in social psychology. In: Levine, D. (Ed.), Nebraska

Symposium on Motivation. University of Nebraska Press, Lincoln.

Kramer, C., 1991. Ceramics in two Indian cities. In: Ceramic Ethnoarchaeology. The

University of Arizona Press, Tucson, pp. 205– 230

Kuhn, S., 2004. Evolutionary perspectives on technology and technological change.

World Archaeol. 36, 561–570.

Latour, B., Lemonnier, P., 1994. De la préhistoire aux missiles balistiques. L’intelligence

sociale des techniques. L’intelligence sociale des techniques, Editions La découverte,

Paris.

Lave, J., 1991. Situating learning in communities of practice. Perspect. Socially Shar.

Cogn. 2, 63–82.

Lave, J., Wenger, E., 1991. Situated Learning: Legitimate Peripheral Participation.

Cambridge University Press, Cambridge.

Lemonnier, P., 1993. Technological Choices: Transformation in Material Cultures Since

the Neolithic. Routledge, London and New York.

Lemonnier, P., 1992. Elements for an Anthropology of Technology. University of

Michigan, Ann Arbor

Longacre, W.A., 1991. Sources of ceramic variability among the Kalinga of Northern

Luzon. In: Longacre, W.A. (Ed.), Ceramic Ethnoarchaeology. The University of

Arizona Press, Tucson, pp. 95–110

Longacre, W.A., Xia, J., Yang, T., 2000. I want to buy a black pot. J. Archaeol. Method

Theory 7, 273–293.

Manem, S., 2008. Etude des fondements technologiques de la culture des Duﬀaits (âge du

Bronze moyen). PhD thesis, Paris-X, Nanterre.

Manzo, G., 2007. Variables, mechanisms, and simulations: can the three methods be

synthesized? Rev. Fr. Sociol. 48, 35–71.

Mayor, A., 2010. Ceramic traditions and ethnicity in the Niger bend, West Africa.

Ethnoarchaeology 2, 5–48

McElreath, R.,

Boyd, R., Richerson, P.J., 2003. Shared norms and the evolution of ethnic

markers. Curr. Anthropol. 44, 122–130.

Mesoudi, A., 2007. Using the methods of social psychology to study cultural evolution. J.

Soc. Evol. Cult. Psychol. 1, 35–58.

Moscovici, S., 1984. Psychologie sociale. Presses Universitaires de France, Paris.

O’Brien, M.J., Bentley, R.A., 2011. Stimulated variation and cascades: two processes in

the evolution of complex technological systems. J. Archaeol. Method Theory 18,

309–337

Parkinson, W.A., 2006. Tribal Boundaries: stylistic variability and social boundary

maintenance during the transition to the Copper Age on the Great Hungarian Plain. J.

Anthropol. Archaeol. 25, 33–58.

Pool, C.A., 2000. Why a kiln? Firing technology in the Sierra de Los Tuxtlas, Vera Cruz

(Mexico). Archaeometry 42, 61–76

Rice, P.M., 1984. Pots and Potters. Current Approaches in Ceramic Archaeology. Institute

of Archaeology, University of California, Los Angeles.

Roux, V., 2015. Standardization of ceramic assemblages: transmission mechanisms and

diﬀusion of morpho-functional traits across social boundaries. J. Anthropol.

Archaeol. 40, 1–9.

Roux, V., 2013. Spreading of innovative technical traits and cumulative technical evo-

lution: continuity or discontinuity? J. Archaeol. Method Theory 20, 312–330.

Roux, V., Gabriellini, S., 2016. Structures de cuisson et période de transition au Rajasthan

(Inde, 2005–2015). Des choix instables avant sélection déﬁnitive. In: Müller, C.,

Heintz, M. (Eds.), Transitions Historiques. De Boccard, Paris, pp. 157–168.

Roux, V., Jeﬀra, C., 2016. The spreading of the potter’s wheel in the ancient

Mediterranean. A social context-dependent phenomenon. In: Gauß, W., Klebinder-

Gauß, G., von Rüden, C. (Eds.), The Transmission of Technical Knowledge in the

Production of Ancient Mediterranean Pottery. The Austrian Archaeological Institute,

Vienna, pp. 165–182

Roux, V., Lara, C., 2016. Why potters have not borrowed the kiln? Comparing Narratives

of Indian and Ecuadorian potters. Americae <

http://www.mae.u-paris10.fr/articles-

articulos/why-potters-have-not-borrowed-the-kiln-comparing-narratives-of-indian-

and-ecuadorian-potters/ >.

Shennan, S., 2013. Lineages of cultural transmission. In: Roy, E., Lycett, S.J., Johns, S.E.

(Eds.), Understanding Cultural Transmission in Anthropology: A Critical Synthesis,

Methodology and History in Anthropology. Berghahn Books, Oxford, pp. 346–360.

Shennan, S., Steele, J., 1999. Cultural learning in hominids: a behavioural ecological

approach. In: Box, H., Gibson, K. (Eds.), Mammalian Social Learning. Symposia of the

Zoological Society of London 70. CUP, Cambridge, pp. 367–388.

Shennan, S.J., Crema, E.R., Kerig, T., 2015. Isolation-by-distance, homophily, and “core”

vs. “package” cultural evolution models in Neolithic Europe. Evol. Hum. Behav. 36,

103–109.

Sjöman, L., 1992. Vasijas de barro: la cerámica popular en el Ecuador. Centro

Interamericano de Artesanías y Artes Populares, Cuenca.

Stark, M.T. (Ed.), 1998. The Archaeology of Social Boundaries, Smithsonian Series in

Archaeological Inquiry. Smithsonian Institution Press, Washington, WA; London.

Stark, M.T.,

Bishop, R.L., Miska, E., 2000. Ceramic technology and social boundaries:

cultural practices in Kalinga clay selection and use. J. Archaeol. Method Theory 4,

295–332.

Stark, M.T., Bowser, B.J., Horne, L. (Eds.), 2008. Cultural Transmission and Material

Culture. Breaking down Boundaries. The University Arizona Press, Tucson.

Tehrani, J.J., Riede, F., 2008. Towards an archaeology of pedagogy: learning, teaching

and the generation of material culture traditions. World Archaeol. 40, 316–331.

Wiessner, P., 1983. Style and social information in Kalahari San projectile points. Am.

Antiq. 48, 253–276.

Williamson, K., 1971. The Benue-Congo languages and Ijo. Curr. Trends Linguist. 7,

245–306.

Wimmer, A., 2013. Ethnic Boundary Making: Institutions, Power, Networks. Oxford

University Press, Oxford.

V. Roux et al.

Jour nal of Anthropol ogic al Ar chaeol ogy 48 (2017) 320–335

335

India

Ethiopia

Cameroon

Four Areas of Focus

Tikar plain

Goljoota and Qarsa

Bulandshar district

Cameroon

Ecuador

Four Areas of Focus

San Miguel de Porotos Tikar plain

Ethiopia

Goljoota and Qarsa

Cameroon

Tikar plain

India

Bulandshar district

Ecuador

San Miguel de Porotos

Gender and Foot-wheel

MgA. Emil Adamec

info@adamec.org

info@adamecstudio.com

日

月

。

工

作

室

M C M X C I V

Thank you for your attention!

INSTITUTE OF CLASSICAL ARCHAEOLOGY & INSTITUTE OF ETHNOLOGY

Faculty of Arts

CHARLES UNIVERSITY