The first part (the first four chapters) consist of a systematic study of what is commonly termed in archaeology “the accompanying dead” (funerary retainers), i.e. men and women that have been (intentionally) executed after a person’s death (the latter generally possessing a relatively high social status). This research uses data and hypotheses drawn from all available documents, be they archaeological, ethnological or even sociological. The first conclusion derived from this in-depth study is that executed accompanying dead. are an extremely widespread social custom: they are frequent in Asia, amongst the Scythians, the Turkic and Mongol peoples, in ancient China, in sub-Saharan Africa in the years immediately preceding colonization, and finally among South and North American societies (this tradition is attested in populations studied by both pre-Columbian archaeology and ethnological research).

This custom has been considered by all social sciences as a form of “sacrifice”. This is a complete misnomer: sacrifice consists of an offering to the gods or to spirits and is performed by the sacrificer, who disposes of the sacrified object or being and offers it to the deities. Quite the contrary, the individual who has ordered the execution of his followers, his slaves or concubines intends to keep them at his service after his death. The notion of offering is in this case entirely absent. Moreover, none of the documents relating to people who carry out this practice (written material from China or, less frequently, data from Mesopotamia, as well as ethnographic descriptions) appear to describe the executed accompanying dead. as primarily a religious custom. Discarding of the sacrificial interpretation paves the way for a totally different understanding of this practice and of its social implications; this new interpretation is paramount in clarifying the main issue of this book.

The initial interpretation of the existence of this custom is wrong because far too often it has been seen a kind of cruel but, in the end, unimportant ritual (without any social implications), to be explained only as the product of strange and barbaric beliefs. Additionally this explanation is mistaken because this custom was often considered to be a court ritual and a sign of royalty. Yet the data arguing against this is abundant: the accompanying dead are also frequently attested in many societies devoid of any form of state organization, and whose social structure has absolutely nothing to do with the institution of kingship: this is particularly well illustrated in the case of African lineage-based societies or communities of the north-western American coast (the latter without any contacts with “great” Pre-Columbian civilizations) [the data has been discussed in such a way so as to present the main points behind ideas and notions like “lineage-based societies”, which are part of a terminology commonly used in social anthropology but little-understood by the general public].

On the other hand, the custom of the executing followers upon the death of a powerful person is notoriously lacking in great states that one would like to describe as “consolidated” (at least those that are bureaucratically structured, like China after imperial unification by Qin Shi Huang Ti, Mesopotamia or Egypt from very the early time of unification). The overall aim of this first part is to establish facts that are not always obvious, to record the motivations behind this funerary custom and its social context. It is also tempting to define both its geographic and chronological outlines.

This first part ends with a very technical discussion of some still unresolved and enigmatic archaeological facts, particularly from the Upper Palaeolithic period, during which the idea of “sacrifice” or of the accompanying dead has been called to mind. The study discusses certain differences in the archaeological materials themselves and some little-known ethnological or historical observations, and concludes with the absence of any executed accompanying dead during this prehistoric period.

The second part (chapters 5 and 6) is entirely devoted to the investigation of the social implications of the accompanying dead. It relies naturally on all the material mentioned in the first part described above. It names the social categories from which the accompanying dead originate, these being quite varied: slaves, followers of all kinds, royal servants, wives, friends or lovers, etc. It is possible however to define an archetypal portrait of the accompanying deceased, i.e. in other terms to name the four major characteristics of the accompanying dead.

Firstly, he is a dependent of sorts, because nothing indicates dependence better than the fact that one cannot survive his master. Secondly, the escort can be a faithful follower, an individual who kills himself over the grave of his master in order to show his extreme loyalty. Data revealing the number of suicides amongst accompanying dead. is far from lacking: these are very widespread, not only amongst Indian widows subjected to the satî tradition or among samurais performing junshi in order to follow their master into the netherworld.

At first sight these characteristics (loyalty and dependence) appear to be contradictory. This is absolutely not the case. The author bases himself on his knowledge of pre-modern slavery (a theme that constitutes the topic of one of his books) in order to demonstrate that in populations as different as the Amerindians of the north-western pacific coast or Arabia at the beginning of the twentieth century the most reliable servants are supposedly to be found among slaves.

The third characteristic of the accompanying dead lies in his personal link to the person he is following into the grave. In order to explain this idea of “personal link” meant by the author, the author contrasts loyalty/faithfulness to a principle and loyalty to a person, functional hierarchy and personal hierarchy, etc… The use that is made of this concept is not very different from Marc Bloch’s in his book Société féodale (Feudal Society): this historian imagined the “personal link” as being filled with personal relationships of dependence.

This third characteristic is no doubt the most important one. The main argument of the book emphasizes that the tradition of accompanying dead signals the existence of relationships of personal loyalty within a society. The importance of this custom reflects the importance of these relations. The second part includes two main points. The first consists in a critique of traditional social anthropology, a field that has focused excessive interest on kinship relationships and has neglected all other social bonds whose importance, demonstrated in several works, is obvious in many aspects of social life related to exchange and power: what are being specifically referred to here are personal links outside kinship. The second point relates to archaeology: there is a systematic archaeological bias that leads to an underestimation of accompanying dead’ importance. This can be explained by the fact that not all the accompanying dead (as specified in the first part) are obligatorily deposited in their master’s grave: the slaves of the north-western Pacific coast are thrown into the sea, and only a fraction of those executed in Africa are buried in the tomb (the others’ remains are discarded in the forest), not to speak of those incinerated together with their master (this can be archaeologically traced only in very specific situations).

The fourth characteristic of the funerary retinue is that it agrees with despotism and autocracy: the fact that a person has at his disposal other individuals whose absolute loyalty can cause them to die for him means that he obviously wields a lot of power. This point is transitional and logically leads to the third part of the book.

The third part elaborates on the idea of the state as the product of a man’s creation, this person resting on his followers and personal dependents’ support in order to ensure his power.

Chapter 7 defines the state in very orthodox terms, in accordance with Max Weber and anthropologists who are in agreement on this notion. In this chapter explanatory theories on the origins of the state are examined critically, and some opinions and positions on the part played by religion and the economy are discussed.

Chapter 8 deals with political sociology and focuses on traditional states (where power is strongly integrated and bureaucracy poorly developed). By quoting a few well-recorded ethnographic and historical examples, it shows how a prince gains support from special categories of loyal subjects while exercising power. This chapter only broadens and generalizes conclusions gleaned from ethnographic research and oriental studies in history.

Chapter 9 presents the main argument: if state power derives exclusively from influence created by personal links of loyalty towards a prince, why should these loyalties not be the prime movers behind the origins of the state ?

Other geographical and chronological arguments are also examined, one of the principal points being that the custom of funerary retinues (and consequently the existence of personal loyalties) is already present in stateless societies: they precede the state and are not, as previously thought, one of its products or concrete expressions.

The epilogue places the above-mentioned phenomena within an evolutionist social framework.