Sommaire en anglais
des textes en français

        Overview of research

 

The order of sub-fields below is somewhat arbitrary it reflects however the current focus of my research :

1.       Cognitive Anthropology

2.       Economic Anthropology

3.       Kinship studies

4.       Critique of anthropology

5.       Maritime anthropology

 

1. Cognitive Anthropology

 

A cognitive dimension underlies most of my books and articles mentioned below under the headings Economic Anthropology and Maritime Anthropology(2.1 to 2.4, 2.7) and in particular my book written with Geneviève Delbos, La transmission des savoirs (The transmission of knowledge). In this rubric however I only mention work deriving from a seminal (unpublished) paper entitled « Anthropological insights for Artificial Intelligence » delivered at the Artificial Intelligence Project of Yale University in October 1987. At the time I was just back from my field trips to West Africa, and encouraged by having successfully used algebraic models on kinship issues, I was toying with the idea of doing something similar with "primitive mentality" facts. I tried to answer the following question: « What constraints do you need to impose on logical reasoning to generate so-called "primitive mentality" type statements ' ». It is not that I believed there was anything such as "primitive mentality", but it was clear to me that there exists an essential difference between some Far-Eastern and Western ways of conceptualizing, reflected in particular in the Far-Eastern preference for catalogues, and the Western preference for taxonomies.

The code for the simulation of "primitive mentality" phenomena was written in PROLOG, a programming language which has an underlying formal logic structure. It turned out that two constraints needed to be imposed: 1° that any type of asymmetrical relationship between terms, such as inclusion, or causal link be prohibited, 2° that substitution of terms which elicit similar emotional response (equivalent affect value) be allowed. The conclusions of this research were published in a special issue of Revue Philosophiquedevoted to Lucien Lévy-Bruhl, the journal's founder (here below 1.4).

My training as a psychoanalyst (1971/73, 1974, 1988/92) had prepared me to accept both aspects of the conclusion: the first principle is what presides to free association in the psychoanalytical cure, the second is that at work in what Freud called a « complex » and Jung before him had called a « constellation » - drawing attention with this phrase on emotional values radiating on words associated in the pyche. I published in 1987 a paper entitled « Ce que l'Intel ligence Artificielle devra à Freud » (What Artificial Intelligence will owe to Freud) which was sketching a research program for Artificial Intelligence envisaging thought processes as driven by a dynamics of emotions (here below 1.3). Principes des systèmes intelligents (Principles of Intelligent Systems - 1990) expanded this theme to book length. Views similar to those expressed here became popular in the United States more recently, in particular in Damasio's Descartes' Error: Emotion, reason, and the human brain (1994), and in Le Doux The Emotional Brain (1996).

My research on these themes caught the attention of British Telecom who awarded me an Academic Fellowship in 1988 which allowed me to design and develop ANELLA, an Associative Network with Emergent Logical and Learning Abilities. In the same year 1988, the French cultural radio « France Culture » commissioned from me a set of four two-hour broadcasts on Artificial Intelligence. I mingled interviews with leading researchers from the United States (Paul Smolensky), France and Belgium with short illustrative texts written by me.

In 1993, at the invitation of Maurice Aymard, Administrator of the Maison des Sciences de l'Homme in Paris, I became a Director (along with Jan Kordys, Polish Academy of Sciences) of a research group entitled Théorie et Clinique des Pathologies de la Pensée (Theory and Clinical studies of Pathologies of Thought Processes). Three members of this research group belong to Grenoble University (notorious in France for its technological strength):

1.     Thierry Vincent, psychiatrist, Head of the Clinique Psychiatrique Universitaire de Grenoble, is the author of a highly regarded history of the contribution of psychoanalysis to the treatment of psychoses. Some of his recent research papers are original expansions of themes initially developed in Principes des Systèmes Intelligents.

2.     Vincent Rialle, Cognitive Scientist, Associate Professor at the Faculty of Medicine of Grenoble, has devoted a number of his papers to a popularization of my research in Cognitive Anthropology. Rialle has also been most effective in promoting my work in conferences, collective volumes and special issues of journals.

3.     Jérôme Zeiliger, Artificial Intelligence engineer, of Institut de la Communication Parlée in Grenoble (Institute of Spoken Communication), is currently writing his doctoral dissertation on an AI system inspired by ANELLA (see above). Jérôme and I are working on a joint paper on the hypothesis that syntax should be regarded as the critical points of a multi-dimensional semantic object when projected in the one-dimensional space of discourse.

At University of California, Irvine, Prof. Douglas R. White and myself are currently working on a joint paper, « Laws of thought / Design for the brain . The paper is a follow-up on « An alternative neural network representation for conceptual knowledge » (here below 1.5). Some of our joint contributions to network theory (here below 3.13 and 3.15) are here transposed to provide hypotheses about memory storage.

My main written contributions in this sub-field are

1.     Truth is shared bad faith (with G. Delbos - 1985a). This article questions the validity of Brentano's « functionalist » view that folk psychological categories such as belief, intention, desires, reflect mental states. Starting from the work of Rodney Needham on Belief, language and experience (1972), the expression of believing or of knowingis shown to be independent from any mental state of « certainty ». Asserting a view as a belief or as something we « know » for sure, is nothing but ways of making others know what part of our knowledge we are prepared to revise under their influence. Underlying the mechanism of belief and knowledge is adhesion: the degree in which a locutor identifies him/herself with the contents of the clauses s/he utters.

2.     Le robot pensant (with G. Delbos, The thinking robot - 1985b). Truth and falsehood are categories which along with necessaryand contingent, possible and impossible, etc. were first used as polemical terms. There is still a difference between a « general truth » which functions in the world at large and the type of truth and falsehood which are being built between interlocutors in a conversation as their « shared knowledge ». Analyzed in an Artificial Intelligence perspective « shared knowledge » is then the mathematical intersection of memory networks conceived as directed graphs.

3.     Ce que l'Intelligence Artificielle devra à Freud (What Artificial Intelligence will owe to Freud - 1987). Artificial Intelligence needs to borrow from Freud the concept of thought processes being the operation of a dynamics based on emotion on a substrate of memory traces structured as a network. The paper sets the research program for a new style in Artificial Intelligence integrating the assets of psychoanalytical insights about the functioning of the thinking mind.

4.     IntelligenceArtificielle et mentalité primitive (Artificial Intelligence and "primitive mentality" 1989). A reflection on so-called "primitive mentality" types of reasoning reveals that what we recognize as such are alternative spontaneous ways of apprehending the world typical of Far-Eastern traditional societies. Two modes of operation distinct from those of the West can be singled out: 1) a preference for symmetrical connections between concepts rather than inclusive relations, 2) a preference for clustering concepts through similar emotional response rather than visual resemblance. These two features are typical of the types of connections which we produce in our cultures through free association. It is suggested therefore that what we recognize as "primitive mentality" types of reasoning are probably expressions of the automatic retrieval of memory traces and therefore revealing of the actual structure of memory storage. Lessons for "knowledge representation" in Artificial Intelligence can consequently be drawn.

5.     Principes des systèmes intelligents (Principles of intelligent systems - 1990a). This book, published in a series called "Cognitive Science", aims - as suggested in the title - at defining general properties of « intelligent systems ». The question of the intelligence of a computer system running a piece of software can be considerably simplified if no consideration is given to the concept of a « subject » displaying intelligent behavior. In everyday life we content ourselves with simply assuming that our human interlocutors are inhabited by a « subject »: we never demand any proof of the actual existence of such a subject. Searle's « Chinese room » thought experiment shows a human subject being fluent in Chinese but being unaware of such capacity for being solely aware of manipulating skillfully sets of symbols.

For all practical purpose all persons dealing with the prisoner of the Chinese room would be justified to regard themselves as dealing with a fluent speaker of Chinese. In the circumstance the « subjective » feeling of the subject that he understands or not Chinese is irrelevant to his actual fluency. The same would apply to a computer system: as long as the sentences it produces are indistinguishable from those that a human being would generate, it would justifiably be regarded as intelligent. The question reduces then to that of designing a function and data structure that generates algorithmically specific sentences by traveling through the word-space of a particular language's lexicon.

The first constraint to be observed is that of grammatical acceptability of the clauses generated. The second is that of the plausibility of these sentences within a general context of human cultures (common sense view). The third constraint is that of topicality. The fourth constraint is that of no-contradiction between consecutive clauses. Little by little a picture emerges of a network of memory traces capable of growing in an organized manner associated with an affect dynamics capable of generating from these memory traces, sentences displaying emergent logical features.

6.     An alternative neural network representation for conceptual knowledge (1990b). This paper introduces the P-Graph representation of a neural network as an alternative to the classical « semantic networks » introduced in knowledge representation by Quillian. None of the shortcomings of Quillian-type semantic networks are displayed by it. The P-Graph is a particular type of dual of a graph: memory traces (typically "words") are associated with the edges of the graph, the relations between the memory traces, with the vertices. The P-Graph is the mathematical object underlying ANELLA (Associative Network with Emergent Logical and Learning Abilities). The P-Graph - in particular the way it grows - is shown to be compatible with the architecture of an actual biologicalneural network, its emergent logical and learning abilities are shown on examples borrowed from the working of ANELLA as developed at British Telecom Laboratories in 1988 under a BT Academic Fellowship.

7.     Typologie des savoirs et transmission informatique (Types of knowledge and their IT transmission - 1991). This paper reviews various types of knowledge and explanation, scientific and empirical, and raises the question of their duplication in IT systems. The contrast between symmetrical vs. anti-symmetrical connections between concepts (central to 1.4 and 1.5) is shown to be the template of the signal/correlation vs. causal types of explanation. Recent questioning of the validity of causal explanation by the representatives of the "qualitative physics" movement in Artificial Intelligence is shown to be justified. This leads to suggest that "signal/correlation" types of explanations are more easily justified in every case than "causal" types of explanation.

8.     L'Intelligence Artificielle : au confluent des neurosciences et de l'informatique (Artificial Intelligence, at the meeting point of neuroscience and Information Technology - 1994), Artificial Intelligence betrays the special dispositions and traditions of the fields which constitute its ancestry: neuro-physiology, psychology, logic and mathematics. What is the common thread between the divergent pull of these fields emerges in a model of thought processes as a gradient on a "memory trace" landscape. Paths generated on this landscape are interpretable as sentences displaying emergent logical properties.

9.     La linguistique d'Aristote (Aristotle's linguistics - 1996). The divide between semantics, syntax and logic is absent from Aristotle's linguistics where the progress from concept to discourse is continuous: concepts linked in pairs constitute judgments, judgments linked make clauses, clauses linked amount to discourse. One's degree of adhesion to one's own speech - from simple quotation to expression of personal belief - constitutes a final coating. Aristotle's linguistics ' as it can be abstracted from the six books of the Organon and the Rhetorics - shows the way to an alternative approach to discourse generation dispensing with the obstacles linked with the semantics, syntax, logic divide.

10. Ce qui fait encore cruellement défaut à l'Intelligence Artificielle (What is still badly missing from Artificial Intelligence - 1997a), Artificial Intelligence still lacks an adequate theory of meaning. Maybe the obstacles we observe to the progress of AI are partially imaginary. We suppose for instance that there is essential difference between opening a window « mechanically » because someone has asked us to do so and converting to a religion under the inspiration of a preacher. What if the phenomena were in fact of a similar nature: the power on a mind that a word has through its meaning ' The model of a gradientin memory associations generating discourse with emergent logical properties - proposed by me in earlier publications - suggests we never intend to say the sentences we utter, but simply register - like anyone else - what are the words that our mouth utters. What we hear ourselves saying re-launches the emotional dynamics underpinning our speaking, just as do sentences we hear when uttered by others.

Plato had noticed that the central part of what we call « thinking » is our inner hearing of sentences produced inside ourselves, in what we call our « imagination ». If such is the case then consciousness is hardly more than the time needed by our emotional dynamics to update itself in line with what we hear ourselves saying (either with the « outer » or with the « inner » ear). If such a perspective is accepted the role of consciousness gets deprived of its decision making role in the generation of rational discourse. Consciousness is real as opposed to illusory but its role is ancillary and needs not to be reproduced in a machine meant to mimic intelligent sentence production. This article, published earlier this year, has launched a debate in the French AI community.

11. Jean Pouillon et le mystère de la chambre chinoise(Jean Pouillon and the riddle of the Chinese room - 1997b). This article is part of a special issue of the journal L'Homme meant as a Festschrift to honor Jean Pouillon for his eightieth birthday (Pouillon was secretary of Jean-Paul Sartre, then of Claude Lévi-Strauss, and editor of such journals as Les Temps Modernes, L'Homme, Nouvelle Revue de Psychanalyse, Le Temps de la Réflexion ). In an article published in 1984, Pouillon claimed that a better rendition of a text heard can be achieved when contents is ignored and attention is focused only on syntactic structure. I examine the implications of this statement in the light of Pouillon's earlier Temps et Roman, a theory of the novel published in 1946. The fact is that if we know for certain what meaning to attach to categoremes (content words) we hardly know how to express the meaning of syncategoremes (structure words). What is the part of meaning being conveyed by syntax, what is that part of meaning which is unrelated to contents '

Pouillon claims that understanding a sentence is identical to adhering to it, that is, to fully identify with its contents. But, do we adhere to sentences that we understand, or is it rather that we claim we understand a sentence because we hear ourselves claiming that we adhere to its contents ' (A positive exploration of what is the part of meaning conveyed by syntax, constitutes my current joint research with Jérôme Zeiliger; see above).

 

2. Economic Anthropology

 

The longer period of my main anthropological fieldwork on the Island of Houat -- off the Southern Coast of Brittany in France -- lasted from February 1973 to May 1974. I spent shorter periods of time in Houat in the Fall of 1974, in 1975, 1977 and 1978. In t he first six months of my stay I was employed half-time in the local Lobster Hatchery, spending the rest of the day copying genealogical data from the parish records. Then I became an extra on the boats in the local fishery. Overall I spent 33 days at sea in 1973/74 as a deck-hand. I spent time on the boats also during my later stays, this time essentially asking questions to complete my record and taking pictures. I managed to maintain a month by month record of crew composition over a five year period.

My first stay was made possible by a five-year research grant from the Belgian Fund for Scientific Research (FNRS). I had been offered the grant in the Spring of 1969 before I presented my finals. In 1975/76, then again in 1977/78 I was a recipient of grants from the Wiener-Anspach Foundation which allowed me to write up my Brussels' Ph.D. dissertation while being a graduate student at Cambridge University.

My Ph.D. dissertation (presented in December 1976 - degree obtained in January 1977) was entitled Anthropologie économique de l'Ile de Houat (Economic Anthropology of the Island of Houat). It led to my 1983 book with a similar title. In the book a portrait is drawn of a small fishing community with an emphasis on the interlocking of demographic and economic constraints. Apart from participant observation, my main sources of data were my book-keeping of twelve boats over one year, personal fishing diaries dating from the 1950s and 1960s and the demographic facts abstracted from the genealogical data found in the parish records.

I became an "Assistant Lecturer" in the Department of Social Anthropology at Cambridge University in 1979. From 1981 to 1983 I spent holidays and sabbaticals collecting historical data, this time on the Breton mussel and oyster fisheries and on the sardine fisheries in the period between the end of the First World War and the 1960s. This additional material complemented my earlier records and that collected separately by Geneviève Delbos -- my first wife -- on contemporary oyster-breeding and on the traditional salt-industry past and present. Our joint work was financially supported by a grant from the French Ministry of Culture. Our joint book, La transmission des savoirs (The transmission of knowledge) was published in 1984; it went into a second print in 1991.

The interview material I had obtained on the 1920/60 sardine fisheries had drawn my attention back on anomalies in price formation which had first struck me while in Houat. I started writing specifically on price formation this time. An integrated theory of price formation would fall into place once I had collected similar material in the West African fisheries.

In 1984, along with the other two colleagues who had been appointed in 1979, I lost my position at Cambridge University. At that time I was negotiating with the Food and Agriculture (FAO) division of the United Nations in order to obtain grant money for graduate students. I was offered by my contacts to become a UN expert in the Fisheries as part of a project on maritime fisheries in West Africa, stationed in Bénin (formerly Dahomey).

The circumstances for investigation were close to ideal: Cyriaque Atti Mama, a sociologist from the University of Bénin was attached to me as a research assistant; in each of the six villages in our study, data collection on the daily catch was performed by « statisticians » we recruited among young local fishermen. Also we were provided with facilities to travel anywhere along the coast of Bénin on a daily basis and to visit any of the twelve countries being part of the development project. I published two FAO reports (here below 2.10, 2.11) and had one article published from material on migration and seasonal moves by fishermen which the UN did not wish to publish (here below 2.12).

In 1988 and 1989 I was the recipient of grants from IFREMER (French research institute for maritime studies) and once again from the French Ministry of Culture. I furthered the interviewing about the Breton sardine fisheries initiated in 1981, this time in a different part of Brittany and also started taking part in the over the counter fish auction taking place at 5 a.m. in Lorient's fishing harbor.

In 1990 the orientation of my field collection of price formation data changed radically when I became a « trader » on the international futuresmarkets. My capacity for programming and my innovative views about price formation led to further employment in the financial software industry, first in designing and developing automated trading systems, then in designing and developing risk management systems.

My training as an anthropologist has allowed me to shed new light on financial techniques. A book I finished writing in June 1997, entitled Le Prix (Price), contains in particular an analytic catalogue of financial instruments based on the two categories of « rent » and « sharecropping » which I studied extensively in the traditional environment of the small-scale fisheries. This completed manuscript brings together my work of the past ten years on price formation. It contains my first systematic treatment of observations and data dating from the 1984/86 West African field trips. It is currently being reviewed by French publishers.

Having developed a good "feel" for trading on the futures markets I started building in 1996 a simulation model of price formation on such markets. Taking advantage of the Object Oriented features of Visual C++, I am in the process of trying to mimic price formation as the consequence of the interaction of various groups of investors having differential access to market information (time lag), different assets, different time horizons and different stop-loss and profit taking strategies.

My main published contributions to Economic Anthropology are

1.     Adjuration du hasard et maîtrise du destin (see 2.2. -1976a). The quality of being a « good fisherman » is shown to be largely independent from economic performance. The four qualities mentioned in conversations are endurance (courage), persistence (patience), flair (cognitive alertness) and luck, the latter accounting for all possible accidents which are independent of an individual's personal qualities.

2.     To be a good fisherman you do not need any fish(1976b). English translation of 2.1.

3.     Marks and rabbit furs. Location and sharing of grounds in coastal fishing (1978). It is shown in this paper that contrary to the intuition of the non-fisherman that the sea is an homogeneous body of water, knowledge of reefs, wrecks, lobster caves, etc. is localized and results in fishing rights according to the principle of « knowing is owning ». Territory rights associated with geographical proximity or historical record of association, when infringed lead to riposte which may culminate in murder.

4.     Les deux concepts fondamentaux de la pêche artisanale: la « saison » et le « métier » (The two fundamental concepts of small-scale fishing: the "season" and the "trade" - 1979). The « season » and the « trade » in off-shore fishing seem to refer respectively to a division of time and to a fishing technique associated with a particular type of gear. In actuality, the season determines also a space and responds to varying ecological circumstances, while the trade has several ramifications, like the by-catch of the fishery and constraints on crew expertise and composition.

5.     All-brother crews in the North Atlantic (1982). The common pattern of fishing crew composition in the North-Atlantic is that of an alternation of father and sons and all-brother crews depending on the stage reached by the household in the family cycle. All-brother crews are notable for their instability. Authors have reported this about Newfoundland and Sweden in particular. Reasons advanced for such instability are always of a psychological nature. It is shown here that the developmental cycle of the family implies that from father and sons to all-bother crews, the ratio of consumers depending on one boat's income to producers rises from single to double. Economic pressure provides therefore a sufficient explanation for stress and ultimate rupture.

6.     Effet attracteur de la performance économique moyenne (The attractive power of the mean economic performance - 1983a). Statistical reasoning is an easy prey to folk interpretations of probability and statistical theory. The independence of draws leading to a normal distribution is difficult to grasp and most people « feel » that if at dice a long series of six has obtained another six becomes less likely. The « attractive power of the arithmetic mean » is one such fallacy (recently reborn under the name of « mean reversion » as a misnomer for anti-persistence). It has been at the center of controversies in the École d'Anthropologie of Paul Broca at the end of the nineteenth century in Paris. Figures from economic performance in a Breton fishing community show how the mechanics of social pressure can contribute at giving the appearance of an « attractive power of the arithmetic mean ».

7.     Les pêcheurs d'Houat (Fishermen of Houat - 1983b). This is the book version of my thesis. It contains the material of 2.1 to 2.5; also an in-depth study of the demography of a community which until the time of the investigation (1973/74) displayed so-called « ancien régime » (pre- 1789) demographic features in terms of birth-control and family size. The book had extensive reviews on the French radio and television (interview of the author) as well as in daily papers (Libération, Le Figaro). The Times Literary Supplement devoted to the book a half page review signed Malcolm Chapman, then at the Institute of Anthropology of Oxford University.

8.     Chayanov should be right: Testing Chayanov's Rule in a French fishing community (1984a). The object of this paper is to test the hypothesis of an « invisible hand » operating within small-scale traditional communities such as the fishermen of the Island of Houat in Brittany. Statistical evidence is used to show that in Houat « it all happens as if » producers modulated their work effort so as to ensure a similar income level per consumer. Indeed income is more highly correlated to a consumer/producer index than to any other variable such as tonnage of the boat or number of workers (correlation between workers and income: 0.462, correlation between dependent consumers and income: 0.826). A scattogram shows the strong correlation between the consumer/producer index and the actual number of crab pots effectively used on a daily basis out of a set of 300 available.

Also, comparative demographic figures from Houat and Saint-Molf a neighboring community of traditional salt-producers are used to show an effective - although unconscious - strategy of shaping the family according to the ideal size of the working unit. Actual figures for Houat turn out to be very close to what would be expected in terms of family size if there were a strategy of having three sons (constituting with the father an economically optimal crew), while actual figures for Saint-Molf are very close to what would be expected if there were a strategy of having one son only (optimal for the running and transmission of a salt-production unit).

9.     La transmission des savoirs (with G. Delbos, The transmission of knowledge - 1984b, 2d print 1991). This book has attracted an important following, not only among French-speaking anthropologists, but also among French-speaking sociologists. For the launching of its second print in 1991, a conference was gathered at the Abbey of Royaumont near Paris by the French Ministry of Culture. The paper I delivered then is referred to above as 1.6. The field material used in the book was gathered by Geneviève Delbos among traditional salt-producers and oyster-breeders and by myself among fishermen and, at the time of the investigation, among mussel-breeders.

Chapter 1, deals with attempts at transmitting empirical knowledge through the school system; this is usually done by deriding empirical knowledge as « anti-scientific » or « superstitious ». It is shown however that often scientific and empirical knowledge do not overlap. The distinction established in IT between "procedural" and "declarative" is shown to correspond closely to the distinct manners in which respectively apprenticeship and schooling attempt at transmitting empirical knowledge.

Chapter 2 contains a detailed analysis of the developmental cycle of the household-centered economic units, it expands arguments present in 2.5., 2.6 and 2.8; it is shown to what large extent what may appear as free decision entailing the future is nothing more than complying with a harsh economic reality.

Chapter 3 shows how little there exists deliberate strategies of transmitting knowledge in the occupations here studied. What is actually transmitted is a workload, knowledge is acquired while working but essentially through a process of identification: knowledge is reinvented much more than it is transmitted.

In Chapter 4, we systematically study the differences between scientific knowledge and empirical knowledge. Examples we analyze have often been quoted; for instance, in a recent issue of the journal Turbulences my discussion of conflicting views about behavior of lobsters by scientists and fishermen is once again being quoted at length.

Chapter 5 deals with change, it is shown how these traditional communities are keen to adopt technological innovation and only resist it when it brings with it tensions disruptive of the community, in particular when it would lead to stratification on the basis of wealth. Clearer distortions to sound knowledge are shown to be those that hope entails: when times are hard rationality gets obscured by the necessity to keep going, « superstition » is but the small price to be paid by a most resilient species for survival (as acknowledged in the book, Prof. Meyer Fortes drew my attention on this latter point).

10. The influence of socio-economic and cultural structures on small-scale coastal fisheries development in Bénin (1985a). This 42 page FAO report draws a quite detailed portrait of the fishing communities living on the 150 km coastline of B énin. The interactions between the permanent or seasonal beach settlements and the lagoon villages of native horticulturalists are described. A comprehensive explanation is presented of the economic organization of the polygynous household where me n fish while women buy from their husbands and resell the fish either fresh or after they have smoked it. The « share system » of family-based crews is described as well as the « company system » common among migrant "Keta" fishermen (from the Anlo peninsula in Ghana).

This report is - to my knowledge - the first that debunks the view that small-scale maritime fisheries in West Africa are unprofitable. It shows that if husbands sell indeed at an overall loss to their wives, the household as a whole (with an average o f 1.4 wife per adult man) is a profitable venture. It is shown also how the village voodoo "churches" by taxing the richer among the fishermen contribute actively at preventing social stratification.

11. Non-monetary distribution of fish as food in Béninois small-scale fishing villages and its importance for self-consumption(1985b). This 26 page FAO report shows the important role played by gift distribution of fish in the self-consumpt ion of both migrant fishermen and their native hosts from lagoon villages. Fish distribution is statistically analyzed both in times of plenty and in times of dearth. During the latter, it appears that the totality of the catch is habitually distributed. Through scrutinizing the daily diet of two families, one living on the lagoon, one living on a lake further inland, it is shown how much fish received as gift contributes to the diet. Fish received as gift is shown to be often further redistributed throug h a network of kin and neighbors.

12. Going out or staying home: Seasonal movements and migration among Xwla and Anlo-Ewe fishermen (1988). A detailed demographic study of fishing encampments on the coast of Bénin reveals a mixture of different circumstances among local fish ing populations: on the one hand, wives, children and old parents of indigenous fishermen currently working in Gabon and in the Congo, on the other hand, Ghanaian fishing « companies » in a seasonal movement of following the fish over a coast le ngth of a few hundreds of miles only. Interviews and questionnaires in Sierra Leone, Liberia, Ghana, Togo, Bénin and the Congo, made it possible to map a complicated pattern of migration and seasonal movements among two ethnic groups holding a majo r share in the moves: the « Keta » of the Anlo peninsula in Ghana and the « Pédah » or « Popo » from the grand-Popo region in Bénin, actually a splinter-group from the Fante who settled in Aneho (Togo) and Grand-Popo (Bénin) in the eighteenth century. Although my capacity of United Nations Officer made possible the extensive traveling required by a study of the type, my conclusions were not published by the UN as the migratory pattern revealed w as undermining accepted views of both local governments and the UN itself.

13. Hommes, femmes, et l'"intérêt supérieur du ménage" (Men, women, and the "superior interest of the household" - 1989a). This paper analyzes the identification of the wife with the "superior interest of the house hold" in the traditional Breton small-scale fisheries. The control of women over the household's finances, first described in 1977 (see here below 5.1), is shown to have further increased over a fifteen year period as a consequence of new financial practi ces and techniques such as out-sourced book-keeping and payments by check.

14. Le pêcheur rencontre l'économie (The fisherman meets the economy - 1989b). This final report of an IFREMER and French Ministry of Culture investigation contains extensive field material on price formation in the French fishe ries. It contains also my first analysis of why the supply and demand model of price formation is inadequate. Finally it introduces Aristotle's model of price formation in terms of relative status of buyer and seller and shows its relevance both in the Br eton and in the West African fisheries.

15. Déterminants sociaux de la formation des prix de marché. L'exemple de la pêche artisanale (Social determination of market prices. The case of small-scale fisheries -1990). This paper shows that the theoretical framewo rk provided by marginalist economic theory is inadequate in the case of small-scale fisheries. The social determination of price is shown to be overwhelming. The notion of a necessary allocation of at least a « subsistence wage » to all actors involved in the industry seems to be both admitted and enforced. Agreements attained through negotiation between fishermen and buyers such as middle-(wo)men or cannery dealers - although being typical of crisis situations only - are shown to display the distribution of income that applies actually in all circumstances. Price variations are shown to be hardly influenced by variations in demand and supply and reflect a more global power balance between seller and buyer, wherein social status is the determi ning factor. Reciprocal status between deck-hand and skipper and between skipper and middle-(wo)man appears to be the basis whereupon economic surplus is distributed, providing thus unexpected support for a model of price formation first proposed by Arist otle.

16. Le prix comme proportion chez Aristote (Price as a proportion in Aristotle - 1992). In his Nichomachean Ethics, Aristotle devotes a few paragraphs to a model of price formation. This model has been regarded by most commentators as either obscure or « pedestrian » (Schumpeter). In 1957, Polanyi gave a correct interpretation of the model but claimed it was meant to be normativerather than descriptive. Aristotle's model is here fully developed. It is shown th at it needs to be understood as embodying Eudoxos' theory of proportion known nowadays as Book V in Euclid's Elements. The diagonalproportion of price is contrasted by Aristotle with the parallelproportion of « correctiv e justice ». A more central use of the theory of proportion is what Aristotle develops about the syllogism as a continuous proportion in the two parts of the Analytics for what concerns its scientific use and in the Top ics for what concerns its use in court and in the assemblies.

17. L'économique comme science de l'interaction humaine vue sous l'angle du prix (Economics as the science of human interaction in the perspective of price - 1994a). This article shows how Aristotle's theory of price formation can be extended from traditional markets to the economy as a whole, and to finance in particular. In the non-financial part of the economy, buyers and sellers constitute distinctive groups with a somewhat stable composition; in the financial world, sellers and b uyers constitute groups whose composition is being constantly renewed. It remains however that it is the current power balance between buyers and sellerswhich determines the level of price. Because of the speed of transaction, this power balance i s difficult to assess and relies therefore on representations more than on substantive evidence. This renders price formation on the financial markets particularly prone to be influenced by misgivings such as rumors or inappropriate interpretations of mar ket movements. Strangely enough, Aristotle's philia, the striving for the common good under the shape of the survival of the industry is shown to be an active force even in the contemporary financial world.

18. La queue qui remue la chien. Métamorphose de la finance due à son informatisation (The tail wagging the dog. The metamorphosis of finance due to Information Technology -1994b). Finance has been evolving rapidly since the ea rly 1980s when the personal computer was first introduced. The speed in diffusion of news and in transactions has increased. Fair prices for complex financial instruments can now be calculated in a split of a second. The computer has acquired the function of being an intermediary between human beings and price formation. The computer however ignores fear which with human beings - unless it develops into a panic - acts as a factor for self-regulation, dampening the natural tendency of a hyper-critical phen omenon such as price formation to end up in collapse. In return, men cannot refrain from assigning the machine some intentions as they do with fellow-workers, assigning interpretations to the computer's calculations and its displaying of the outcome of hu man activity. The advent of the computer further encourages actors in the financial world to believe that price formation derives from a rational process; in doing so it contributes at the acceleration in price movements (volatility) which makes fi nancial crises more likely.

19. L'économie au quotidien (The economy in every day life - 1995a). The man in the street believes the economy to be working the way the economist tell him it works. That economists are more often wrong than right in their prediction s is puzzling to the man in the street. But the wrong prediction of the economist is only to be expected as theories belong to the information pool that the economy - in its highly interactive functioning - feeds on. Inaccurate economic predictions are ho wever not indifferent to the economy for two reasons. Firstly because the man in the street is quick to replace the part in an explanation which does not seem to work with a conspiracy theory involving in most cases members of minorities; secondly because an unclear understanding of which parts of the economy are self-regulating and which parts are not, results in new legislation whose effects are often counter-productive. Illustrations of both these negative tendencies are provided about the French fishe ries.

20. Statut, rareté et risque (Status, scarcity and risk - 1995b). « Rating » as performed by financial « rating agencies » expresses relative status of institutional debt issuers in terms of the credit risk they represent. It is shown that the equation of status with credit risk can be extended from institutions to persons. The overall validity of Aristotle's model of price formation (as expounded in 2.16) becomes apparent. Since a person's credit risk is determined by 1) risk of having to interrupt one's activity due to death or incapacity, 2) unreliability of returns, 3) level of competition between practitioners of the activity, Aristotle's model of price formation provides simultaneo usly a theoretical frame for the economic and for the social order. It applies to all types of societies having an economy proper, i.e. all societies where goods circulate and are exchanged against currency in quantities called prices.< /LI>

21. Le rapport entre la valeur et le prix (The relationship between value and price - forthcoming). Currently submitted to the Canadian Review of Sociology and Anthropology.

22. Aristotle's theory of price revisited(forthcoming). Currently submitted to Philosophy.

 

 

3. Kinship studies

 

My research work on kinship began in my first months of being an undergraduate student in Social Anthropology at the Free University, Brussels (1964). Three of my teachers had a marked interest for kinship studies: Luc de Heusch, Jacques Maquet and Ann ie Dorsinfang. De Heusch was expanding Lévi-Strauss' formal perspective onto African kinship and mythological material and encouraged first year undergraduates to read Structural Anthropology (1958) and The Savage Mind (1962). The lat ter book contained as a figure, a representation by Guilbaud of the Aranda marriage system as a permutation group. In 1961 Guilbaud had made a presentation of the Ambrym marriage system at Lévi-Strauss' seminar in a similar vein (published as Guilb aud 1970).

In 1969 I became one of Claude Lévi-Strauss' students at the École Pratique des Hautes Etudes (now École des Hautes Etudes en Sciences Sociales) in Paris. I attended his lectures at the Collège de France and to ok part in his seminar. That same year Lévi-Strauss delivered a seminar paper on « complex structures » of kinship, concentrating on some of Margaret Mead's Mundugomor « rope » material (this led to a chapter in Structur al Anthropology II). Lévi-Strauss' lectures in that year were devoted to Native American mythologies, (incorporated as chapters in volume four of the Mythologiques). At the same time I had become a student of Georges-Théophile Gui lbaud, also at the Ecole Pratique des Hautes Etudes, which I remained until 1971.

Guilbaud's seminar was called « Mathematics for Social Scientists ». I was well prepared for this seminar as at the "Athénée" (high school in Belgium) my majors had been Mathematics and Latin. Also, the mathematical training at Free Univer sity, Brussels in sociology was quite extensive and the same as that for economics students, i.e. in the first year sixty hours of classes and sixty hours of discussion sections in mathematics (calculus, combinatorics, matrices, etc.), and in the second y ear the same amount of teaching for statistics and probability theory. Guilbaud's visual presentation of kinship data were in line with Weil's suggestion in an appendix in Lévi-Strauss' The Elementary Structures of Kinship (1949), to represe nt alliance between marriage « classes » as colored arcs (standing for the generators of the group) connecting a marriage to either the marriage of a daughter (say « red ») or a son (say « blue ») at the next generation..

In 1976 I was appointed Associate Professor at the Free University, Brussels. Gisèle De Meur was a mathematical colleague of mine in the Economic, Political and Social Sciences Faculty. One day Gisèle asked me to explain an example of an « Australian marriage system » mentioned in the standard Introduction to Finite Mathematics (1957) by Kemeny, Snell and Thomson. I referred Gisèle to the work of Harrison White and of Guilbaud, which she then read with enthusiasm. This initiated common discussions, leading in the years 1977 to 1980 to three papers that we wrote jointly (here below 3.2, 3.4, 3.5). In October 1979 I became an Assistant Lecturer at Cambridge University and collaboration with De Meur stopp ed. In November 1980, De Meur organized a conference on kinship networks in Brussels. Among the participants were Ira Buchler, Frank Harary, Per Hage, Chris Gregory, Franklin Tjon Sie Fat, Gisèle De Meur and myself. The proceedings of the conferenc e were to be published in 1986 as G. De Meur (ed.) New Trends in Mathematical Anthropology; in addition to 3.5, 3.10 and an introduction to a chapter of T.T. Barnard's 1924 doctoral dissertation on Ambrym, I provided also the foreword to the volume .

From 1981 to 1983 I went on working on algebraic approaches to various aspects of kinship. In particular I lectured on these issues at Cambridge in my own department but also in others such as Statistics and Applied Mathematics, and History a nd Philosophy of Science. I was also invited to speak on these matters at the Institute of Anthropology at Oxford and at the London School of Economics.

When in Paris in the years 1969-1971 I attended the public lectures of the psychoanalyst Jacques Lacan (it is under his inspiration that I decided in 1971 to train as a psychoanalyst). At the time Guilbaud was advising Lacan on all mathematical topics as he was Lévi-Strauss. As opposed to Lévi-Strauss, Lacan was most interested in mathematics proper and in those years in France - apart from Guilbaud's seminar - mathematical modeling was more consistently advocated in the Department of Psy choanalysis of University Paris VIII than anywhere else in the Social Sciences. It is therefore quite logical that I was asked to give a lecture at the Chair of Psychoanalysis in 1983 (by then the Chair was Jacques-Alain Miller; Lacan had died in 1980). T his lecture was published in 1984 (here below 3.8), as « L'inscription dans la structure de parenté » (The embedding within the kinship structure). When I delivered the lecture, I was about to leave for Africa in the following week s and I knew I would not be in a position to write about kinship for many months to come if not years. Consequently I synthesized in this paper what I regarded as my original contributions to the field of algebraic models of kinship. When back in Paris in 1985 Miller offered me to be a visiting professor at the Department of Psychoanalysis of University Paris VIII; I lectured weekly on « Psychoanalysis and Anthropology », concentrating on discussing Freud's comprehensive anthropological reading list in Totem und Tabu .

In 1990, Françoise Héritier, the successor of Lévi-Strauss at the Chair of Social Anthropology of the Collège de France, suggested to Douglas White that he and I meet. This first encounter led to an immediate collaboration w hich first materialized as two papers on kinship networks (here below 3.13 and 3.15), and currently to our joint project in Cognitive Anthropology: « Laws of thought / Design for the brain » (mentioned above under Cognitive Anthropology).

Lévi-Strauss always had an ambiguous attitude towards kinship networks. He has always welcomed suggestions from mathematicians about how to formalise the study of alliance and descent, or later transformations between myths: Weil's permutati on models for the Murngin appear indeed as an appendix to a chapter in The Elementary Structures of Kinship (1949), and Guilbaud's reduced kinship network for the Aranda is shown as a figure in The Savage Mind (1962). However apart from the « canonical formula » of myth introduced in Structural Anthropology (1958) and never used subsequently, Lévi-Strauss has never produced any properly mathematical modelling.

Lévi-Strauss wants his formalisations to literally emerge from the empirical facts with no prejudice whatever about what the formalisation would reveal. The last thing he would wish is to be influenced by any a priori imposition resulting from t he shape of a mathematical object (Jorion 1985a). This feature of his view of theory would be striking in seminars, where he would apply the principle to guests' presentations. At question time he would go to the board and would start somewhat like this, « If I understood you well, we are dealing with a structure of this particular shape'». Never would he say phrases such as « permutation group », even if his modelling would in the end amount to entirely redefine what is actually a permutat ion group. Such exercises would always be brilliant, going way beyond what the speaker him- or herself would have seen about the configurations present in his/her material. It would always however be « configurations » static models, never dynam ics.

This being said, Lévi-Strauss takes the formalisations he has arrived at in real earnest. About these he applies a principle which I would characterise as « No fact can hurt a good model ». A few years later, in 1975, I would move to o ther teachers : Leach, Fortes and Needham who hold the opposite principle: « No model can hurt a good fact » (Jorion 1985b). The three of them shared the same image of being people hostile to the network approach to kinship. Needham has written extensively on his opposition. However all three have been most encouraging of my own efforts in the field. In what way and how come '

From 1977 to 1979, Leach was my tutor while I was writing under his supervision a thesis on Malinowski. In the end he advised me not to submit it as we had uncovered too many things which were sensitive to persons still alive. It is at that time that h e showed me Langham's thesis, which became a book under the title, The Building of British Social Anthropology, W.H.R. Rivers and his Cambridge Disciples in the Development of Kinship Studies, 1898-1931. The thesis was about Rivers and his students and was centered on the work that he, along with Armstrong, Barnard, Gregory Bateson, Deacon, Layard and Brenda Seligman had devoted to setting the bases of a science of kinship networks.

John Barnes was of course the person who was most knowledgeable about kinship networks in Cambridge at the time. He was the person who drew my attention on Armstrong's pseudo-permutation groups catalogue of section systems published as an appendix to h is book on the Rossel Island gift system (Armstrong 1928). That John Barnes was encouraging was no surprise coming from a pioneer of social network studies. The interest of Leach was triggered by the help that Langham was requesting. Langham had sp ent some time in Britain to gather data, but he was now back in Australia and in the process of rewriting his thesis into a book needed additional material and checks. Writing about Malinowski I was in an ideal position to help Langham fill the holes in h is argument.

Leach was of course an authority on kinship, and more especially on the « circulating connubium » of MBD marriage but because of the reproduction of the economic unit dynamics of both his Kachin and Pul Eliya field material he had never paid much attention to multi-functional kinship networks such as Australia. In addition he had become convinced through his field experience that all kinship was driven by the reproduction of the economic unit dynamics. In the introduction of the paper written in 1981 in collaboration with Leach and eventually published in 1993, I explain how our collaboration worked and why he never wished that I made this collaboration publicly known (Jorion 1993). Leach relished in the attempts at formalising: he had got one of the ve ry early PCs and generated on it the catalogue of all possible two-generator permutation groups up to order 32 which allowed MBD marriage. This we then sorted out in terms of plausibility of male/female generation length. Leach was aware of Warner's comme nt on a 5 x 7 structure: « The two main elements in Murngin kinship are the patrilateral lines and their lateral connections through the intermarriage of the five generations of the seven lines of descent » (Warner 1931: 172). Genealogies collec ted by Shapiro and Kupka allowed then to confirm the 7 / 5 ratio of generation lengths.

Leach's reluctance to have his name associated with this work was founded in his opposition to Lévi-Strauss' « no fact will hurt a good model principle ». To Leach, anthropologists who were dealing with kinship networks could not help taking them too seriously: soon enough they would confuse the model for the actual thing and would defend the validity of the model against the reality of the facts. Leach told me a number of times to read Vaihinger's Philosophy of "as if", which e ncompassed his epistemological views. In Vaihinger, the model is an intellectual construct that lives in the head of a modeller, it is not part of the world outside: it amounts to a scaffolding which can help the thinking process of understanding but shou ld be discarded in the end. Things may look like they materialise the model but they do not: the model has been constructed for heuristic purposes only by a human mind as a schematic representation of the confused and confusing reality. There is here a st rong and consistent anti-Platonist position to which every epistemologically conscious person needs to consider in earnest. The « scaffolding » view is expressed in White & Jorion 1996: « The kinship structures which are mapped in this approach are not intended as any sort of complete representation of kinship "systems", but merely as scaffoldings which help to bring into view kinship as a social field, providing a baseline for other mappings (which may be superimposed) of social proces ses such as communicative fields, exchange processes, transmission of learned behaviours, social rights and inheritance, political and religious succession, and the like » (White & Jorion 1996: 267).

In addition, Leach was very much concerned - like most anthropologists doing mathematical modelling at the time - that there was no critical mass of mathematically-minded anthropologists. In his review of Harrison C. White's An Anatomy of Kinship, Leach wrote, « The book is, in part, explicitly addressed to anthropologists, 99.9% of whom could not read it even if they wanted to » (Leach 1964: 156). Barnes wrote similarly « Hence we have the development of a technical language and a body of literature which, quite appropriately and inevitably, is incomprehensible to other social scientists, and often other anthropologists, as well as to the general public » (Barnes 1980: 297). I myself wrote comments of a similar nature, motiv ated essentially by the difficulty of getting these mathematical papers published. At the end of De Meur & Jorion 1980, I wrote « It should be recalled ['] that in some fields such a demonstration would be regarded as a formidable step but in ant hropology not » (De Meur & Jorion 1980: 20).

Needham's support was decisive in my obtaining in 1982 a Nuffield fellowship which allowed me to hire Elaine Lally for six months. Elaine was an Australian graduate student who responded to a note I posted in the Department of Mathematics at Cam bridge. She had no notion of anthropology when she volunteered for the job but she went on to obtain a degree in anthropology as a consequence of our joint work. She wrote FORTRAN and Cobol versions of the algorithm for genealogy analysis which became «&n bsp;P-graph » when Douglas White and I revived it jointly in 1990. « An algorithm for the analysis of genealogies as to prior kin connection between spouses » never found its way into print. The reviewers of Science cla imed the paper had no originality, the reviewers of Man regarded as a matter-of-fact view that the journal would never publish anything as exotic as an « algorithm ». Elaine and I made a presentation of the algorithm at the Department of Statistics and Applied Mathematics at Cambridge in 1983. From there Elaine moved to a fruitful collaboration with Gisèle De Meur.

Needham had shown himself highly critical ' if not vociferous ' about the kinship networks approach in a number of papers, more particularly in his Introduction to Rethinking Kinship and Marriage (1971) and in the paper he had written in collabo ration with Korn (Korn & Needham 1970). The reason for his encouragement was of the kind that only a great mind can afford: encouraging a line of argument for which he has no sympathy for the sake only of the scientific venture. However Needham always made it clear to me that the main reason for his friendly behaviour towards me and my work originated in his regard for the quality of my fieldwork; indeed when my book on the Breton fishermen of Houat was published he encouraged one of his pupils , Malcolm Chapman, to write a glowing review of it for the Times Literary Supplement. The day I spent with Rodney Needham at All Souls College before presenting a paper in the late afternoon at the Institute of Anthropology of Oxford is one of thos e days which remain as a landmark in one's life.

Fortes' attitude finds its rationale in his short book Oedipus and Job in West African Religion (1959). What was common to my approach and Fortes' to kinship was the striving for explanation and the sympathy for psycho-analytical e xplanation of kinship-related behaviour. One of the underlying assumptions of the kinship networks line of approach is the relative indifference towards whether or not an actor is aware that s/he behaves in a way which reveals structure when this b ehaviour is looked upon in a collectiveperspective. This is proper to psycho-analysis which envisages any type of behaviour as having an unconscious motive at heart. Meyer Fortes invited me in December 1983 to share his lunch in a private r oom at King's College and made it clear to me that ' despite any feelings I may have had to the contrary (over eight years of discussions) ' the formal approach to kinship he had seen me developing - when informed by the psycho-analytical concern for the unconscious motive - was according to him the way of the future. A couple of weeks later Meyer fell into a coma, he died in February 1984.

My main publications in the sub-field of kinship studies are the following:

1.     Réflexion sur la formalisation dans les études de parenté (Reflections on formalization in kinship studies - 1980a). In this paper I explored the implications of formalizing kinship studies and establishe d some clear principles for representation. My later papers would systematically build upon this basis.

2.     La question Murngin, un artefact de la littérature anthropologique (with G. De Meur, The Murngin question, an artifact in the anthropological literature - 1980b). Reviewing the literature on the marriage system of the Murngin of N orthern Arnhem Land in Australia, we show that alternative models proposed by various authors (Radcliffe-Brown, Webb, Lévi-Strauss, Weil, Barnes) can be shown to be as many snapshots of a system moving dynamically between various states. Although t he intuition would show to be valid the article failed to « crack » the Murngin puzzle. Solving the puzzle would be achieved in a paper originally written in 1981 and published in 1993 (here below 3.14); a first expression of the solution appear ed in French in 1984 (here below 3.8).

3.     The hordes of discord: Australian Aboriginal social organization reconsidered (with Michel Verdon - 1981a). Reviewing the literature about the Australian Aboriginal we attempt to lift the confusion existing between various types of groupings ha ving territorial implications. We show that the functions of "hordes", "clans", "sections", "ritual lodges", etc. can only be clarified once it is understood that in the « dream » representation, the natural world needs to be maintained and rene wed through rituals attached to « dream » locations. We show also that for such hunting-gathering groups there is no proper ownership of a territory, only degrees of sophistication in the understanding of how specific locations fu nction. Verdon's reflection on corporate ownership as well as my own work on ownership and sharing of fishing grounds (here above 2.3) had opened up the way to a satisfactory synthetic picture which became reference in the literature.

4.     A possible genealogy of Australian marriage systems (with G. De Meur - 1981b). Among all permutation groups with two generators and of order up to 8, only a few have interpretations among the marriage systems of the Australian Aborigines. An at tempt is made in this paper to construct a two generator permutation group of order 16 such that its quotients of order 2, 4 and 8 are valid models for the existing Australian marriage systems and for them only. Models for the Barkindji, Wurundjeri, «&nbs p;Southern Cross », Kariera, Aranda and Waramunga are generated this way.

5.     Le mariage Pende (with G. De Meur & T. Vuyk, Pende marriage - 1982a). The Pende of Kasai claimed to their ethnographer L. de Sousberghe that men married either their Mother's Brother's Daughter or their Father's Sister's Daughter but that these two types of kin were always distinct. Classical genealogy reduction according to identification of siblings of same sex, and generational identification following patricycles and matricycles entail that if MBD and FZD are permitted wives then they should coalesce into bilateral cross-cousins (MBD ºFZD). In de Sousberghe's accounts, various indications about whom is the person a man asks for a wife suggest that the wife is actually a Mother's Father's Sister's Daughter's Daughter. A specific genealogical lattice is shown which implies that at alternate generations, a MFZDD will at the same time be either a MBD or FZD but never both. Data about the Abutia Ewe of Ghana suggest a symmetrical preferential marriage system with the FFZSD. It is shown that this is what obtains when the f and g generators interpreted as respectively representing women and men are inverted from the Pende model.

6.     Relations généalogiques et catégories cosmologiques dans le mariage australien, de Howitt à Radcliffe-Brown(Genealogical relationships and cosmological categories in Australian marriage, from Howitt to Radcliffe - Brown - 1982b). A systematic examination of reports on kinship among the Australian Aborigines until the First World War suggests that the expression of marriage in moieties, sections and subsections is nothing but a « cognitive shortcut » of kinship relationships in terms of cosmological categories (« totemic groups »). Because women and men are allocated within totemic groups along with all animal and vegetal species, locations, meteorological phenomena, etc. it is in most case s possible to express also marriage in this idiom. (This discovery of the relative independence of exogamous units and cosmological categories opened up the way to the solution of the Murngin puzzle in 3.14).

7.     An algorithm for the analysis of genealogies as to prior connection between spouses (with E. Lally ' 1983). This paper is the result of work carried out over a six months period (1982/83) at Cambridge University with my research assistant Elain e Lally. The « Weil - Guilbaud » permutation group representation is taken one step further by generalizing the interpretation of the f and g generators from representation of permutation of marriage type to representation of indiv iduals as links between their parents' marriage and their own. The two generators f and g correspond to the operation of assigning to a marriage [x] respectively that of the parents of the wife [f(x)] and that of the parents of the hu sband [g(x)]. This manuscript remained unpublished: attempts to have it published in Science and Manfailed in the short time left before my departure to Africa. The reason it is mentioned here is that the algorithm is currently used at UCI, Cal Tech, University of Paris X, University of Cologne, etc. due to the industry of Prof. Douglas White (UCI). Indeed the paper was revived when I showed White the unpublished manuscript when we first met in 1990. White then further designed and d eveloped the algorithm and released it under the name "P-graph". The tenor of the 1983 paper can now be found in White & Jorion 1992 (3.13 below). Houseman & White write (Structures réticulaires de la pratique matrimoniale, L'Homm e 139, 1996): « Compared with other software used for the processing of genealogical data' PGRAPH displays the particularity of allowing to represent a marriage network under a unified graphic form » (p. 61).

8.     L'inscription dans la structure de parenté (The embedding within the kinship structure -1984). An attempt is made here to define kinship in a « minimalist » manner so as to avoid getting trapped into any culturally laden perspective. A major effort is made in particular to distance oneself from the Europeo-centric view of kinship. I introduce the phrase reticular kinship to refer to what has been incorrectly called so far « descriptive » or « comple x » kinship. I stress that the apparent complexity of « descriptive » kinship, of which the European is an example, derives only from its amorphous nature. Calling it reticular kinship puts the emphasis on what makes it special: ever y individual is the centre of a network radiating away from him or her. Particular configurations on reduced graphs, identified with two generator permutation groups, are shown to model convincingly well-known cases in the literature: the Miwok, North Pen tecost, Murngin, Omaha, Republican Pawnee, etc. A major part of this 1984 paper can now be found integrated in White & Jorion 1996 (3.15 below).

9.     Foreword - New Trends in Mathematical Anthropology (1986a). Mathematical Anthropology has had so far a difficult relationship with Cultural or Social Anthropology. A short history of algebraic approaches to kinship is sketched and the unfair on slaught by Malinowski emphasized. It is hoped that will emerge in the future an « enriching feedback generated by the confrontation between models and concrete ethnological facts which poses the theoretician new questions and launches new theories&nb sp;».

10. Alternative approaches to the Ambrymese kinship terminology. A critique of Scheffler (1986b). This article aims at offering a formally acceptable definition of the homomorphisms between reduction of genealogies due to marriage patterns and kins hip terminologies. As a consequence it is also shown that the permutation group approach provides more elegant and parsimonious models than the structural semanticsapproach advocated by Scheffler. The tenor of this article has been extensively use d in the « The logic of kin relations » chapter of Ascher's Ethnomathematics(1991).

11. Le sujet dans la parenté africaine (The subject in African kinship - 1987). Illustrating a point made in 4.10 aboutstructure and sentiment, I analyze the phenomenon of village fission in African communities of slash -and-burn horticulturalits as a consequence of excessive demographic pressure. Drawing the attention on the fact that fission is commonly preceded by witchcraft accusations I show how the inner logic of witchcraft entails the existence of two antagonistic personalities, favoring therefore a type of representation that will ease the splitting of a community which was seeing itself so far as a single individuality. My earlier experience in Brittany with the splitting of all-brother crews under mounting econ omic pressure (as analyzed in 2.5) had offered me the template for understanding the quasi-physical nature of fission in African villages.

12. Le frère de ma mère sera toujours mon oncle (My mother's brother will be my uncle forever - 1991). Lucien Scubla revives a question first raised by Françoise Héritier: « Why is the Father's Brother often identified with the Father but never the Mother's Brother ' ». It is first observed that question as such only makes sense within the context of our European reticular kinship where the terminology suggests a possible identity of two type s of « uncles ». It is then shown in a step by step demonstration that the principles of genealogy reduction would only allow a Mother's Brother to be identified with a Father if all men of the father's generation were collapsed into a single ca tegory, disposing therefore of any notion of « kinship ». Thus is shown one of the possible benefits of a formalized approach to kinship: to reveal that some empirical cases will never exist because they would correspond to either formal impossi bilities or self-defeating trivial cases.

13. Representing and computing kinship (with D. White - 1992). In this paper a method for representing and processing genealogical data is introduced. The method consists in inverting the traditional convention of representing genealogies as graphs with one color for edges standing for descent and two colors for vertices representing men on the one side and women on the other: in the P-graph there is only one type of vertex used to stand for a "marriage" and two colors used to connect respectively a marriage to that of the husband's parents and to that of the wife's parents. To the marriages contracted within one generation put under vector form correspond then two other vectors, one containing the marriages of husbands' parents (g function) and a second one containing the marriages of wives' parents (f function). In this manner is suggested an easy manner for exploring systematically genealogies where spouses are kin-related (implementation of the algorithm described in 3.7). The fec undity of the method is shown on material borrowed from the Book of Genesis.

14. Matrilateral cross-cousin marriage in Australia(1993). This article was originally written in 1981, jointly with Professor Sir Edmund Leach (King's College, Cambridge). It shows that Lloyd Warner's statement of 1931 that the Murngin social sys tem displays « the intermarriage of the five generations of the seven lines of descent », regarded until now as mistaken is actually an accurate description of the Murngin marriage system. It is shown in the paper that both kinship terminology a nd demographic data support a reduction of the genealogy into a 5 by 7 two generator permutation group. Processing of demographic data gathered by W. Shapiro and K. Kupka reveals through relative generation distance between men and women the distribution of the 35 exogamous units over the 8 subsection cosmology.

15. Kinship networks and discrete structure theory(with D. White - 1996). The kinship network as derived from the P-graph as defined in 3.13 provides a universal idiom for the representation of kinship - whichever way this is defined. Kinship as c ulturally shaped can be represented with the toolbox of the P-graph - and the biological acting as underlying template - with the emphasis put on any particular feature. Cultural particularities often materialize as determining equivalence classes over re lationship, allowing reductions of the genealogical graph, following sibling identification, cyclical generational collapse, etc. Marriage between kin-related spouses show on the graph as cycles. Illustrations are provided by the genealogy of American Pre sidents, the genealogy of Canaan in the Book of Genesis and genealogical data of Australian Aborigines of Groote Eylandt.

 

 

4. Critique of anthropology

 

The critique of anthropology, in particular the hesitations of the field between Humanity and Natural Science has been a constant theme in my written work over the years. From 1975 to 1977 I was one of the editors of Cambridge Anthropology, a jo urnal run by graduate students from the Department of Social Anthropology of Cambridge University. I contributed a number of critical articles to this journal (4.2, 4.3. 4.7).

In the years 1985 to 1988 I held the Anthropology column in the literary magazine L'Âne, edited by Judith Miller, the daughter of psychoanalyst Jacques Lacan. The subjects I covered in these columns are exoticism, the rich looking down on the poor, the classification of mankind, the feral: wild and savage, cultural uses of anthropology, shifting social interests within anthropology, « filling the historical record », colonialism, fact and fiction: ethnography vs. novel, the ps ychic unity of mankind and « primitive mentality », the writing of ethnography, « friendship » in the field, Colin Turnbull and the Ik, « contemporary cavemen », etc.

My main contributions to a critique of anthropology are the following:

1.     Quelques réflexions sur les conditions de l'enquête en anthropologie (Some reflections on anthropological fieldwork - 1974). The deep immersion within a small community implied by the traditional type of fieldwork through pa rticipant observation is a landmark in the life of the anthropologist. It is also a one-time event in the life of the host community forced to « stage » itself. An anthropologist doing ethnography within her/his own culture is more expected to c onform than someone seen as an outright alien as s/he is seen as « one of us ». However much s/he tries s/he will end up being a major catalyst of modernization for the host community. The snapshot s/he is taking is that of a world that s/he is contributing to making a disappearing world.

2.     Anthropological fieldwork : forerunners and inventors (1976). The charter of modern « participant observation » fieldwork is shown to be a little known foundation report written by W.H.R. Rivers. After Rivers' death in 1922, Malinowsk i who regarded himself as Rivers' most accomplished pupil would consistently erase the memory of his idealized mentor.

3.     What are anthropologist talking about ' (1977a). An attempt is made to render explicit a number of axes underpinning the anthropologist's theorizing: the objective vs. the subjective, the quantitative vs. the qualitative, the universal vs. the particular and the public vs. the private.

4.     Ethnologie et archéologie de l'anthropologie(Ethnology and archeology of anthropology - 1977b). This was my inaugural public lecture as a Professor at the Free University, Brussels, draws a historical portrait of anthropology fro m its late eighteenth century beginnings to the current scene. The hesitations between natural science of man or subdivision of the humanities are underlined.

5.     La notion spontanée de magie dans le discours anthropologique (with G. Delbos, The spontaneous notion of magic in the anthropological discourse - 1980). It is shown in this paper that the word « magic » as used in the an thropological literature has not got the positive content usually assumed. What it refers too is a remnant: whatever is left over when what is religion and science has been withdrawn. There is however a hard core to magic: a theory of singular events. Sci ence as Aristotle aptly defined it is a discourse about things as far as they are universal. About things as far as they are singular, science has got nothing to say. Hence a place remains for « magic ». This article is often quoted by French ph ilosophers of science.

6.     The downfall of the skull (1982). The first attempt at making anthropology a natural science (from the late eighteenth century to the 1920s) centered on craniology, or the attempt to classify human groups on the evidence of the shape of the skull. The reason the attempt failed is essentially due to the mathematical difficulty of modeling a high-dimensionality problem. A great deal of ingenuity was mobilized in the process: some innovative statistical tools such as r, Pearson's «&n bsp;correlation coefficient », were invented to serve craniometry. The path was opened however for a classification of human populations on other grounds: the variety of their social institutions and cultural productions '

7.     Emic and Etic (1983). Anthropologists have adopted with enthusiasm Kenneth Pyke's distinction between « emic » and « etic » molded on the linguistic distinction between « phonemic » and « phonetic ». Whil e anthropologists often suggest that « emic » corresponds to a « seen from the native's point of view » and « etic » to the analytical point of view of the anthropologist, analysis of texts where the two labels are effectivel y used reveal that in practice, etic corresponds to the perspective of a particular sub-field of anthropology such as economic anthropology or religious anthropology, while emic is used when the anthropologist talks to colleagu es across the divisions of sub-fields.

8.     Reprendre à zéro (Starting from scratch - 1986). This article had had repercussions outside the anthropological community. Observing that French anthropology had ossified since the 1960s, the text suggests that most of the issues central to the field are improperly addressed and that the question arises therefore if it might not be wise to start from scratch. Totemism is proposed as an example of an issue which has been inadequately handled; in this instance the post ulate of the « psychic unity of mankind » has acted as a self-imposed obstacle to the discovery of an viable alternative to our Western and modern ways of apprehending the world conceptually.

9.     La nature ou le réel forclos (with G. Delbos, Nature, or the foreclosure of reality - 1988a). In this reflection on the common usage of the word « nature » we observe that the word is most often used about a resource, i. e. an economic value. Nature is defined as a capitalin need to be managed with a clear distinction being made between the returnswhich can be reaped and the asset which needs to be protected. In Brittany, when oyster spat was first collected on tiles at the end of the nineteenth century a clear distinction was made between this artificialtechnique and the dredging of "wild" oysters out at sea. However, when in the 1970s, oyster spat got induced from living shells with in hatcheries, the new technique was branded « artificial » and the spat collected on tiles was from now on referred to as being the « natural » technique. Similarly naturalists see in the field the world of man and in the hedge that s urrounds it the realm of nature, although they were both created by man in a single process. Colonization and taming of the natural world by man is so much a thing of the past that we use the word « nature » to refer to the penultimate tech nological stage of exploitation of our planet.

10. La logique du chaos ou une physique sociale de Durkheim à Lacan (The logic of chaos, a social physics from Durkheim to Lacan - 1988b). The popularity of anthropology as a general subject is rooted at any one time in a cultural dis position to see or not the past as shaping the way to the future. Anthropology flourished as a subject in the 1920s and its popular appeal cannot be dissociated from the debate around "free love" when books such as Sex and Repression in Savage Society (1927) and The Sexual Life of Savages in North-Western Melanesia(1929) by Malinowski or Coming of Age in Samoa (1928) by Margaret Mead made it to the best-seller lists. The 1960s were again one such period when in a climate of environme ntal and quality of life awareness, societies traditionally studied by anthropologists seemed like having managed to avoid the pitfalls of modernity. The romantic opposition to anthropology as a "natural science of man" is founded on a misapprehension of what such a natural science would entail. Because of the complexity of human interaction, some cultural expressions such as ritual are bound to show near infinite variation, resisting therefore simple classification. Also, the particularities of non-linea r dynamics mean that even deterministic behavior may manifest itself under a pseudo-random form indistinguishable from true chaos. An authentic « natural science » anthropology will still meet the demands of the romantic current within it as it achieves the same aim of liberation. Indeed, in the terms of Needham's contrast between « structure » and « sentiment », structure is nothing but the regularities which cultural and social behavior reveals when actors act in uni son, while sentiment is what emerges to the individual's consciousness as s/he resists the imposition of the collective behavior of all others.

11. Le relativisme en anthropologie (Relativism in anthropology -debate with Dan Sperber - 1988c). I was invited by the MAUSS (Anti-Utilitarian Movement in the Social Sciences) to represent the relativist position in anthropology ver sus Dan Sperber, holding a hard-line scientific position for the field. It appeared rapidly that the critique of science position I often adopt, deriving from my observation that science most often does not meet its own standards of rigor, had been confus ed in the public with an anti-science stance. As opposed to Sperber who refuses any identification with the relativist view, I defined my « relativist » position as an acknowledgment of viable alternative solutions to the problems of knowledge c onstruction and collective organization. This position does not entail however any compromise as to the goals of progress and liberation.

12. La vraisemblance discrète du préjugé (The discreet plausibility of prejudice - 1989). The historical succession of two discourses, by travelers and ethnologists, about primitive societies leads us to believe in a con tinuity between them. As a consequence, the ethnology that for a century was written in a monographic style borrowed from the sciences is placed in parenthesis, but it should not be: the partly illusory aim of expounding a scientific discourse leads to co ntrolled « explicitation » whereas the novelistic form deliberately combines the true and the plausible and thus lets prejudice, or at least accessible intuition, evolve freely. This form, by resorting to psychological processes of spontane ous identification with the hero of the narration, omits the founding question of ethnology: whether or not the Other is irreducibly different.

13. La communication dans l'oeuvre de Claude Lévi-Strauss(with C. Assaba, Communication in the work of C. Lévi-Strauss - 1993). In this contribution to a dictionary devoted to communication we concentrate on issues related to this theme in the work of French anthropologist Claude Lévi-Strauss. The topic of epoch-making The elementary structures of kinship(1949) was the exploration of the possible shapes societies take when women are exchanged between them (this is LS' view) in various ways. Central to Lévi-Strauss' perspective is that mankind is universally aware of the impossibility of keeping one's women for oneself. Once this view gets implemented the path is open that leads from Nature to Culture. In Race and History (1952), Lévi-Strauss introduced the hypothesis that a people's technological progress is determined by the number of its neighboring cultures, conceptual confrontation of alternative approaches constituting the very source o f invention. Finally, Lévi-Strauss' work on mythologies recognizes in inversion of themes at the occasion of borrowing from neighbors the main motor of mythological creativity.

14. Les trois moments historiques du sacrifice (The three historical moments of sacrifice - 1995). As a contribution to a special issue of La Revue du MAUSS, I propose a three stage history of sacrifice. Drawing from my field experien ce in Dahomey of the « vaudoun » religion (Dahomey is the birthplace of Haitian voodoo), I take this "mystery" religion as typical of the religions that must have preceded Christianity in the Mediterranean. Sacrifice is shown to be an aggressive mode of defense against deities essentially hostile to man: sacrifice is here used as an alternative to vengeance - which would be appropriate in similar circumstances would the deities be visible and mortal. In the founding moment of Christianity, Chris t sacrifices himself to appease a God who has marked man with an original sin: a nature marked by evil. Through self-sacrifice Christ signals the coming of a new type of God: a benevolent God which can be addressed through prayer. Sin has become a persona lized feature and can be redeemed through speech also under the form of confession. A third stage in the history of sacrifice is attained with Jean-Jacques Rousseau: sin is now removed altogether from the picture, as man is seen good by nature. In the fir st stage of sacrifice, evil as displayed by others is blamed for whatever unfortunate becomes to us; in the second stage we ourselves are the source of any misfortune becoming to us; in the third stage, nobody is responsible anymore for misfortune: evil o ccurs through ignorance and one may say only accidentally. In Rousseau's Confessions, neither his enemies nor himself are to blame, evil is attributable to adverse circumstances: blaming anyone in particular would be as silly and as irresponsible a s Rousseau's father blaming his son for his wife's death while giving birth.

15. Also 32 editorial columns in magazines and daily newspapers L'Homme, L'Âne, Libération,Le magazine littéraire, The New Statesman, Synapse. La Revue du MAUSS.

 

5. Maritime anthropology

 

Items 2.1 to 2.15 listed above as contributions to Economic Anthropology are at the same time contributions to Maritime Anthropology. Also

5.1     L'ordre moral dans une petite île de Bretagne (The moral order on a small island in Brittany - 1977). This paper provides powerful illustrations of the Durkheimian theme of ethics as internalized social order. The reproductio n of the economic unit which is a crew is shown as a crucial factor of the social organization. The division of space with the sea seen longing to the men and the land as belonging to the women is shown to leave but little space to men whenever not at sea : the decks of the boats and the jetty are the locations where men gather spontaneously when not at sea.

2.     The Priest and the Fishermen (1982). Representations of death among Breton fishermen confront the anthropologist with the uncomfortable representation of his own death. At the same time spontaneous and intuitive identification with the plight o f the fisherman faced with death by drowning may lead to misinterpretation as the anthropologist may not partake any more in the highly religious dialectics of vows and pilgrimages.

References

Armstrong, W.E., Rossel Island, Cambridge: Cambridge University Press, 1928

De Meur, Gisèle & Paul Jorion, A possible genealogy of Australian marriage systems, Morphisms in "matrimonial class" systems, Mathematical Social Sciences, 2 (1), 1981: 9-21

Fortes, Meyer, Oedipus and Job in West African Religion, Cambridge: Cambridge University Press, 1959

Jorion, Paul, Affaires de famille. Le Magazine Littéraire, 223, 1985a: 62

Jorion, Paul, La planète Lévi-Strauss: Grande-Bretagne. Le Magazine Littéraire, 223, 1985b: 63

Jorion, Paul, Matrilateral cross-cousin marriage in Australia, Social Science Information, 32, 1, 1993: 133-146

Korn, F. & Needham, R., Permutation models and prescriptive systems: the Tarau case, Man5, 1970: 393-420

Kupka, K. & Testart, A., A propos du problème Murngin: le système des sous-sections, L'Homme 22, 2, 1980: 71-90

Langham, Ian, The Building of British Social Anthropology, W.H.R. Rivers and his Cambridge Disciples in the Development of Kinship Studies, 1898-1931, Dordrecht (Holland) : D. Reidel Publishing, 1981

Leach, Edmund, Review of H.C. White, An Anatomy of Kinship, Man, 1964, 156

Needham, Rodney, Introduction to Rethinking Kinship and Marriage, London: Tavistock, 1971

Shapiro, W., Miwuyt marriage: Social structural aspects of the bestowal of females in Northeast Arnhem Land, Ph.D. thesis, Australian National University, 1969

Warner, W. Lloyd, Morphology and functions of the Australian Murngin type of kinship, Part II, American Anthropologist, 3 (1), 1931, 172-198