The Sociology of Tomb and Temple Robbers of the Late 20 th Dynasty: Part II,The Bands of Thieves and the Ramesside Society

Abstract

This article is the second part of a study devoted to the sociology of tomb robberies at the end of the 20^th Dynasty and focuses on the organization of the thieves into bands. These structures were relatively stable. Leadership was inconstant and associated with unequal loot sharing and conflict. Households, including enslaved people, played an important role in band organization particularly among tomb robbers, while hierarchy and patronage with corruption and blackmailing characterized temple thefts. Another social pattern relying on equality, solidarity, and merit existed among tomb robbers of an equal social status and probably corresponds to what ancient Egyptians called wnḏw. How these patterns of sociability are representative of the society of the time is discussed.

Introduction

This article is the second of a series on the social aspects of tomb and temple robberies recorded at the end of the 20^th Dynasty in the so-called Tomb Robberies papyri. A first analysis ¹ showed that with the exception of workmen from Deir el-Medina, tomb robbers were of a lower social status than temple robbers, consisting mostly of craftsmen originating from the east bank, whereas temple robbers were mostly priests and staff members of the temple they robbed. Another dichotomy appeared as thefts were investigated before or during the Renaissance. Before the Renaissance, robbers were mostly specialists exercising professional skills for what they looted, whereas post-Renaissance, people without suitable professions frequently committed the thefts. Finally, those who robbed the institution they pertained to, although transgressing taboo, generally respected its most sacred parts, whereas robbers foreign to them did not. Here we analyse the social organization of the bands of thieves with the eventual discussion of whether this may tell us something of the society of the time.

Material and Methods

The material and methods used in this study have been extensively described in the first article of this series. To recapitulate briefly: a database was used containing 298 individuals involved in tomb robberies, 92 in temple thefts, and three in both. The ancient Egyptian terminology calls all of these people ỉṯȝyw. However, we also distinguished between robbers proper and accomplices, fences, blackmailers, suspects, and witnesses. Women represent 9.3% of this population. Statistics used the Chi-2 test and the Fisher exact test for comparison of categorical variables and the ANOVA test for quantitative ones. A p value of < 0.05 indicated significance; marginal significance was p ≥ 0.05 and < 0.10. Social network analysis ² was used to graphically represent inter-individual relationships and provide quantitative indices for the identification of important players in the network including the Eigenvector centrality, which measures influence in a network for being connected to other well-connected actors, and betweenness centrality, which identifies actors bridging players. Here the two values gave similar results. The bands of thieves are detailed in the Appendix. Finally, we used the traditional chronology for the end of the 20^th Dynasty for reasons explained in part I.

Terminology Designating Bands of Robbers and Their Hierarchy

The most interesting term is , wnḏw. In P. Amherst-Leopold II, ³ a thief states that: ỉw rḫ.tw ʿšȝ.t n rmṯ.w n pȝ tȝ ỉṯȝy ỉm.w m mt.t ỉw.w m wnḏw sp-sn, ‘it is known that a multitude of people of the land robbed among them [the tombs] likewise being as wnḏw’, with sp-sn stressing on the word. ⁴ In P. BM EA10052, in a lacuna context, a robber declares: ‘whether the… me alone or a wnḏw’, ⁵ rhetorically opposing robbing collectively and alone. Although the thieves’ declarations are probably not verbatim (see part I), the context suggests that wnḏw was actually used by the robbers and not put in their mouth by the investigators. The use of wnḏw corroborates evidence showing that robbers formed bands, probably because tomb plundering needed a combination of specialized skills and robbing temples attended by many people was impossible for a single man. The same thief of P. Amherst-Leopold II adds that he went about robbing with his accomplices for several years, ⁶ showing that wnḏw were relatively long-lasting structures, although subgroups might form within a band and dissolve according to opportunities.

Wnḏw probably carries more than the neutral meaning of ‘herd/group of people’ etymologically derived from wnḏw ‘cattle’. ⁷ This rare word mostly appears in religious, official, and wisdom texts to designate a collectivity of humans but with differences concerning the fundamental purpose of this collectivity according to the context. In chapter 1 of the Book of the Dead, the deceased declares n.wỉ wnḏw.k Wsỉr, ‘I belong to your wnḏw, Osiris’, and the same with Horus ⁸ to indicate that he is among the gods’ followers. The Abydos stela of Neferhotep I distinguishes wnḏw and mr.wt among the personnel adoring Osiris in his temple, ⁹ and Seti I states that the gold transporters of his Abydean sanctuary are m msw n ḥw.t.ỉ ỉm wnḏw n r-pr.ỉ ‘among the children of my temple, within the wnḏw of my sanctuary’. ¹⁰ Thus, wnḏw includes temple staff members who are not priests but, more importantly, distinguishes them from dependent people liable to a corvée (mr.wt). ¹¹

The use of wnḏw is not limited to an Osirian context. In a letter from the 21^st Dynasty, wnḏw designates a group of people exercising the same occupation. ¹² In a professional context, the scribe of P. Anastasi I describes himself as tp ỉry.w.f ḥȝt ỉm wnḏw.f, ‘first of his comrades, foremost of his wnḏw’, to which a variant interestingly adds m-ẖn mhw.t.f, ‘in his family/tribe’. ¹³ Thus wnḏw may designate a group of professional peers, with a sense of belonging to a tribe or a family. The same connotation melding familial links and the notion of a collectivity of worthy people a good man wishes to belong to is exemplified by another passage of the same text, in which, to demonstrate his probity, the scribe declares he will be buried at Abydos in his father’s tomb and adds: ḳrs.tw.ỉ m wnḏw m ḏw tȝ ḏsr.t (‘I shall be buried among wnḏw in the sacred mountain’). ¹⁴ Wnḏw as designating a community of dead among whom a person of good standing might expect to be buried similarly appears in the Decree of Nesikhons, in which Amon-Ra decides ỉw.ỉ dỉ.t šsp st nȝ nty <m> wnḏw nb m ẖr.t-nṯr m šsp nfr (‘I will make that all those who are <in/as> any wnḏw will receive her in the Necropolis in good reception’). ¹⁵ Finally, the importance of wnḏw as a structure particularly worthy of respect is demonstrated by its second position in the list of negative confessions of Chapter 125 of the Book of the Dead in which the deceased declares n smȝr.ỉ wnḏw (‘I have not penalized wnḏw’). To conclude, the best translation of wnḏw is probably ‘community’, to designate a human collectivity sharing socially meaningful motives and values.

In other instances, the Tomb Robberies papyri use the factual collective tȝ rmṯ or tȝ rmṯ n ỉṯȝy.w (group of men or group of robbers) to designate the bands, ¹⁶ while both investigators ¹⁷ and thieves ¹⁸ refer to partners with the usual ỉry (comrade, colleague, or companion) and use ỉry n. ṯtṯt (opponent) when conflict arises. ¹⁹ In any case, none of the words designating bands of robbers have a specific criminal connotation.

Hierarchy among thieves is seldom mentioned. On one occasion the investigators indicate that a robber and his brother were the leaders, saying they ḥn r-ḥat, ‘went at [their] head’. ²⁰ As we have seen, a similar expression is used in P. Anastasi I to indicate that the scribe is ḥȝt ỉmy wnḏw.f, ‘the foremost of his wnḏw’. ²¹ The situation in P. BM EA10052 6.20–23 ²² can be interpreted in two different ways. A robber tells how he went to the Valley of the Queens with his accomplices and adds: ỉw Pȝ-wr-ỉḫtf r-ḥȝt=n, ‘while Pawerikhtef was at our head’. Here, r-ḥȝt may simply describe a factual situation: Pawerikhtef, a tomb workman, ²³ who knew where the tomb lay, guided the robbers and logically probably walked at their head. But as, in this band, the man who showed the way to the tomb received a double share, ²⁴ r-ḥȝt might also indicate at the same time Pawerikhtef’s precedence over his accomplices. ²⁵ In contrast, when in P. BM EA10053, Vs 7.12 a troop commander and blackmailer is said to be the superior (ḥry) of robbers, it probably means that he was their hierarchical superior in the temple organization according to the usual meaning of the word, rather than the band leader. Thus, what ancient terminology suggests is that thieves formed communities sharing different social commonalities and values, among which hierarchy was not the crucial point, even if leaders or distinguished people might emerge among peers.

The Bands

The analysis was conducted with 279 individuals who could be ascribed to one of the eight bands of ỉṯȝy.w involved in tomb and temple robberies ²⁶ for which we have enough information (see table 1 and the Appendix). These people had 518 identified inter-individual links. A ‘link’ corresponds to a direct relationship between two ỉṯȝy.w identifiable in the texts, whatever its nature. Evidently, detailed declarations as in P. BM EA10052 provide more information than concise ones as in P. Amherst-Leopold II, P. BM EA10053, P. BM EA10054, and P. Mayer A and this should be taken into account when analysing the results.

Table 1.

Band characteristics. Date: date of investigation. Band organization is according to the texts. Band leader designated or not. Hierarchy: influence of hierarchy of the structure embedding the band. Empty boxes: no information. None: not mentioned. R: Ramesses. Ren: Renaissance. Y: year. RP: robbers proper. A: accomplices. F: fences. %: % of links. * Result limited to a subgroup.

Band name	Date	Source	Band organization	Leader	Band composition	Structure/ hierarchy	Familial links	Institutional links	Booty sharing	Physical violence	Blackmailing	Comment
Sebekemsaf’s pyramid & non-royal necropolis	RIX y 16	P. Amherst- Leopold II, P. BM EA10054	wnḏw no leader		14 RP specialized craftsmen	None	3%	15%	Equality and solidarity	None	2% prison scribe	Robbing in subgroups. Destroyed royal and non-royal mummy and coffins
Deir el-Medina Valley of the Queens (Queen Isis)	RIX y 17	Necropolis Journal, P. BM EA10068, P. BM EA10053			8 RP + 136 F Specialized craftsmen F: mostly traders & craftsmen	Position in the gang of workmen	100%	100%	Inequality			Band composed of two families linked by marriage. Broke the sarcophagus, stole metal, linen, and pieces of the inner coffin
Tomb of RVI	?	P. Mayer B	No leader		7 RP Craftsmen, ʿʿȝ.w, child of the tomb				Equality and disputed	2 killed	14% to enter the band	Robbed the content of boxes and chests. Text incomplete
Valley of the Queens/Bukhaaf’s band (Queen Isis and 3 other queens)	Ren y 1	P. BM EA10052, P. Mayer A, Abbott Docket	tȝ rmṯ wnḏw with named leader	Herdsman	13 RP + 12 A + 15 F	Possible patronage	17%	20%	Inequality and dispute	Rivalry and violence (62%)	10% scribe	Complicity with a tomb workman. A subgroup with equality and merit sharing robbed without the leader’s assent, leading to violence. Robbed the inner coffins and mummy shrouds
					People involved in cult and temple economy. Women are A
Royal & Theban and non-Theban non-royal necropolis Iufnamun’s band	Ren y 1	P. BM EA10052, P. Mayer A, Abbott Docket	2 brothers named as leaders	Troop commander & prophet, herdsman	20 RP and A Heterogeneous professions. No military	Probably the leader’s position	2%	10%	Dispute on the quality of the booty	Rivalry violence (5.6%) killed people		Complicity with a tomb workman and a chief Medjay. An enslaved person had to steal from the leader’s brother because the latter murdered the brothers of the enslaved person’s patron
Ramesseum I	RIX y 18	P. BM EA10054			7 RP + 1 A Priests and priests-craftsmen	Temple hierarchy (patronage)	30%		Inequality and accepted	None	18% Sem-priest scribe of royal records	Robbed metal on temple statues and furniture. The thieves spontaneously gave a large share to the blackmailers
Ramesseum II	(RXI) y 9	P. BM EA10053	Leader identified from the context	Temple scribe	9 RP + 5 F Priests and priests-craftsmen	Patronage	0%	70%	Inequality and disputed	None	64% troop commander, scribe of royal records	Several robbers are sons of the preceding band. The booty of a subgroup which robbed without the leader’s consent was confiscated. Robbed metal on doors and temple furniture
Medinet Habu	Ren y 1	P. BM EA10403, P Mayer A, Abbott Docket	No leader		18 RP + 10 A Heterogeneous professions. Only one priest	None	10%	33%	Inequality and accepted	None	80%* ʿʿȝ.w	Several subgroups. Robbed portable chests

Two categories of consistently recorded links could be analysed with confidence: familial relationship and booty sharing. An analysis of the affiliated institutions is less compelling since these are not always mentioned. The following categories of sharing were considered: egalitarian and accepted (shares are equal and the thieves agree), unequal and accepted (shares are unequal with the agreement of all), conflictual (the amount of share or its quality is contested or a share is confiscated by others, sometimes violently), and shares obtained by extortion or blackmailing. The results are summarized in table 1.

Among the eight identified bands, the pattern of organization was far from unique and appears independent of whether thieves robbed temples or tombs, or were investigated before or during the Renaissance. Only three bands had an identified leader. ²⁷ Of the two wnḏw mentioned in the corpus, only one had a leader in keeping with what was seen previously. However, two of the eight bands stand apart. The first one, which robbed Sebekemsaf’s pyramid, consisted of craftsmen gathered for their complementary skills who manoeuvred in smaller formations according to the circumstances. Due to the importance of professionalism in the formation of this band, familial links played a secondary role. Pertaining to the same institution was not a major factor, although a core subgroup of the band came from the temple of Amun. This band probably had other points of sociability to facilitate member meeting and contact. ²⁸ An important characteristic is booty sharing, which was totally equitable. Moreover, solidarity was a strong bond: when a partner was deprived of his share, his colleagues compensated his loss. ²⁹ This explains the longevity of the band, which stayed together for several years. ³⁰ An equal division of booty apparently occurred with the thieves who robbed Ramesses VI’s tomb, ³¹ although one member used menace to enter the band. ³² This band which, as far as we know, had no leader was similarly composed of low-status people.

The band of Deir el-Medina also shows specific features. It is the only one totally composed of family members originating from a single institution. The particular organization of the tomb is the probable reason. The men in charge of the construction of the royal sepulchres where probably among the rare group of people who may have had access to the royal wadis. They lived in a settlement specifically assigned to them and their family and they followed a strong tradition of intra-familial transmission of the function. ³³ Thus, in a community where everybody could know what his neighbour was doing, family offered the best warranty for discretion and security. Another interesting feature is that loot sharing probably took into account the traditional precedence of fathers over sons, but also hierarchical position in the tomb workforce (see Appendix).

Unequal loot sharing was the most frequent situation, whether there was an identified leader or not. The reasons vary. It was a case of merit with the man who showed the way to the tomb receiving a double share with the agreement of his comrades (Bukhaaf’s band). ³⁴ In two other bands, although without a designated leader, it was because thieves who considered themselves prominent players requested a larger share (Ramesseum I ³⁵ and Medinet Habu ³⁶ ).

How and why band leaders appeared is unclear. It did not depend on whether a tomb or a temple was robbed. In two cases the prominent positions of scribe and troop commander probably explain leadership ³⁷ but the third one was a simple herdsman. ³⁸ Possibly, his skill in networking with people of higher status gave him precedence (see Appendix). Being the leader of a band was not straightforward: in all known examples, the leader’s authority was contested. In two cases he was compelled to sanction his subordinates for robbing without referring to him, thereby impeding his control of the loot (Ramesseum II ³⁹ and Bukhaaf’s band ⁴⁰ ). The result was a division between those who supported the leader and the others. In the last case a plot was hatched against the leader’s brother, probably in retaliation for the leader’s violent behaviour (Iufnamun’s band). ⁴¹ In the three cases, booty sharing was contested, leading sometimes to violence, which is not mentioned in bands without leaders.

Other than for the band from Deir el-Medina, familial links were decisive for band formation, probably because kinship offered trust, confidentiality, and security. Familial ties are more frequent among robbers and accomplices than among the other ỉṯȝy.w (62/321, 19% versus 11/197, 6%: p <0.01%). The most frequent relationship is that between fathers and sons (41% of cases), followed by siblings (27%), and spouses (20%). The authorities were fully aware of this, since they regularly questioned children or wives to get information. ⁴² Women are involved in 30% of familial links. They played a crucial role in specific stages, in particular by concealing in the house their husband’s share and the stones or pieces of wood used for the sharing and kept in case of litigation, as evidence of what fell to each accomplice. ⁴³ They might also interfere with the sharing when they felt deprived of their right, opening the way to violence in the band (Bukhaaf’s band ⁴⁴ ). In contrast, belonging to the same institution does not distinguish robbers and accomplices from other ỉṯȝy.w, indicating that this was probably not a factor in band cohesion (40/314, 13% versus 24/204, 12%, non-significant).

Another important group of players is the blackmailers. Numbering thirteen men, they affected six out of eight bands and sometimes impacted inter-individual relationships. A variety of situations occurred. The most frequent (7/13 blackmailers and 3/8 bands) happened when the crooks saw or heard that a theft was ongoing, which allowed them to extort a part of the booty, supposedly in exchange for their silence (see Appendix). In other cases, the situation lay at the border of corruption and blackmailing. It is sometimes difficult to decide which prevailed without precise context, such as when an arrested thief purchased his freedom by abandoning his share to the prison scribe. ⁴⁵ More interesting is the situation of the Ramesseum bands. Thieves were repeatedly compelled to transfer a substantial part of the booty to temple authorities, including a sem-priest, a troop commander, and a scribe of the divine records, who clearly threatened to report them to the high priest of Amun. ⁴⁶ Blackmailing was in all cases a position of power exerted over robbers by individuals external to the band. ⁴⁷ It resulted from the impossibility of keeping robberies secret, particularly in temples, and implied that a low proportion of people (6% of ỉṯȝy.w) aware of what was occurring took advantage of their position.

Two occupations were significantly overrepresented among blackmailers, ⁴⁸ namely scribes (3/13, 23%, compared to the fact that scribes made up only 4% of the bands, 7/162, 4%: p<0.05) and foreigners (4/13, 31% versus 14/162, 9%: p<0.05), and there was a marginally significant overrepresentation of subaltern administrators (3/13, 23% versus 11/162, 18% p=0.07). ⁴⁹ The presence of scribes is certainly explained by their powerful position, which gave them easy access to the highest authorities, their most efficacious weapon for blackmailing. The same probably explains administrators. In contrast, foreigners (ȝʿʿ.w) are unexpected. ⁵⁰ They appear under Ramesses XI with the arrival at Thebes of the King’s Son of Kush Panehsy and probably formed his Nubian troops. Some were settled on fields at the south of Thebes. ⁵¹ In the dossier, we find them as cultivators, soldiers, bakers, ⁵² and even a prophet of a temple of Sobek, ⁵³ but more frequently without specification. As occupation troops, perhaps they took advantage of their power to racketeer the thieves. ⁵⁴

Discussion

Here we will focus on points not yet discussed. The first issue concerns the use of ‘band’ and not ‘gang’ to designate groups of robbers. In our modern terminology, gangs are well-structured collectivities following regularized rules, with differentiated roles among members and placing primary importance on group survival. ⁵⁵ Clearly, ancient Egyptian robbers were less structured, prompt to denounce colleagues, and their leaders faced difficulties in affirming their authority, showing that regularized rules were seldom accepted, hence the use of the term ‘band’.

Another point of discussion is the criminal nature of the activities undertaken by the groups concerned here. Post-internment interventions were commonplace in ancient Egyptian tombs, resulting in a fragmentation and a rearrangement of the initial burials to accommodate newcomers, whether they were descendants of the initial owner or of a different origin. ⁵⁶ Accidental tomb collision was another common cause of these rearrangements. Despite the frequent reappropriation of a part of the initial grave goods, including coffins, ⁵⁷ these rearrangements did not exclude respect toward the dead, ⁵⁸ though they were a source of conflict between claimants to ownership of the tomb and thereby to the right to rearrange the burial. ⁵⁹ In the known cases, the court which arbitrated the conflicts did not sanction the rearrangement itself, but the right to do it, ⁶⁰ showing that this behaviour was not considered intrinsically deviant.

The activities concerned here are of a different nature. The context of the Great Tomb Robberies is clearly that of such serious and reprehensible acts that they needed to be investigated by the highest authorities and exposed the culprits to a death sentence. The Abbott papyrus uses the verb thi to qualify what the accused did against royal and non-royal tombs, ⁶¹ a term which basically signifies ‘going against’ someone or something with the connotation of wrongdoing, so that its primary meaning is to transgress or violate. ⁶² The investigated people were well aware of the transgressive nature of their acts and that they were exposing themselves to the risk of a death sentence. ⁶³ Thus, their behaviour was not the result of a more or less conscious shift from a tolerated and even recognized behaviour toward tombs, but a conscious transgression of a social taboo. ⁶⁴

This brings us to the question of what bands tell us about Ramesside society as a whole. One could argue that the disarray of the times marked by an economic crisis would have deeply altered social structures by pushing people to transgressive behaviours, such that the bands of thieves cannot be considered representative of the ‘normal’ society. To address the issue, several counterarguments can be put forth. First, an important part of the western Theban population (perhaps 15 to 20%, and possibly more) ⁶⁵ was implicated in the robberies, so that it is unlikely that all of the thieves easily abandoned the principles of their society to join bands governed by a completely different set of rules. Second, the absence of a universal definition of modern gangs because of their different sociocultural background indirectly demonstrates the weight of the society in which gangs develop. ⁶⁶ In addition, a modern gang feature is their ability to evolve with changes affecting their society, confirming that gangs wholly reflect this society. ⁶⁷ Things were probably similar in ancient Egypt.

Households and patronage are currently viewed as the main axes structuring ancient Egyptian society. At the ‘horizontal’ level, ⁶⁸ households collect people forming the headman’s nuclear family, his kinship, enslaved people, and dependants. ⁶⁹ Hence, a household is an economic unit contributing to the wealth of all and hierarchically structured by the function of its members. It may be crossed by dissensions, since land and house were frequently collectively inherited between siblings. ⁷⁰ Patronage is the ‘vertical’ axis, as any household depends, for its survival, on a man of higher status, or supports the household of a poorer one. ⁷¹ Asymmetrical reciprocity is the principle of patronage. ⁷² The patron protects his dependants in exchange for loyalty. In this way patronage crosses society from top to bottom and adds a stratum of power inextricably interwoven with that of the state, whose highest servants were themselves patrons. ⁷³ However, patronage may deviate from the ideal when the patron behaves abusively to keep his dependant under control. This is what Eyre calls corruption. ⁷⁴

That bands followed at least one of the two axes is demonstrated by a number of examples, particularly in the case of households. The prominent role of familial links, including siblings and spouses over two and sometime three generations, ⁷⁵ has already been stressed. Household is also present in the participation of enslaved people (sometimes bought with the booty) in supporting or witnessing the crimes of their patron. The detailed declarations of Bukhaaf’s band amply show how the house was a sheltering place for preparing thefts, ⁷⁶ receiving accomplices or fences, ⁷⁷ and sharing and concealing the booty. ⁷⁸ Households were the place to find support when conflicts arose among partners. ⁷⁹ They were also the place where familial dissent broke out, such as, for example, with the father, probably the head of the household, who expelled his son-in-law because he disagreed with the theft. ⁸⁰

Patronage is less apparent because there is no specific word to designate the patron. However, we may suspect its presence with the priests who robbed the Ramesseum, who were compelled to transfer a part of the booty to the scribe of royal records, the sem-priest, and the troop commander of that temple, their administrative superiors whose role, tarnished by corruption, is clearly a diversion of that of patron. With tomb robbers, patronage is particularly difficult to identify because bands collected people from different backgrounds without hierarchical links. However, a probable case is the overseer of fields of the House of Amun, Akhmin, who was likely the patron of the herdsman Bukhaaf of that House. Bukhaaf and one of his partners bought a jar of wine with honey before going to his dwelling, the kind of behaviour we expect from a dependant eager to obtain his patron’s goodwill, ⁸¹ here with the intention to purchase land. ⁸² Once more, the context is that of hierarchy among people from the same institution.

When Akhmin was interrogated by the court, he was only requested to give the money back. ⁸³ Similarly, the Ramesseum patrons were apparently not summoned before the court. The scribe of royal records and the sem-priest stayed in place several years to reappear in later documents. ⁸⁴ Patrons, if these men should be seen as such, were not considered by the authorities as being directly implicated. Whether this was a conscious decision to spare them owing to their status or because, from the ancient Egyptian point of view, they did not commit a crime is open to discussion.

What robbers also tell us is that inter-individual relationships frequently escaped the frame of household and patronage. This is best exemplified by fences and especially those of the band of Deir el-Medina, who show the extent of the relationship robbers could develop far beyond their family, patron, colleagues, or institution. Unfortunately, how fences and robbers came into contact remains mostly unknown to us. ⁸⁵

Finally, there is another and unexpected mode of sociability that bands illustrate, that of equality, solidarity, and recognition of merit. This mode develops outside the protection of households since the persons concerned with it seldom have familial links. Interestingly, the known cases never concerned temple robberies, where patronage and hierarchy predominated. It is the appanage of people of the same social rank, mostly craftsmen, sometimes of the same institution. Although this sociability is not without conflict, out of the five bands on which we have information, it appears with two of them and with a subgroup of a third one. This shows that it is not rare. It goes hand in hand with the absence of a leader and, interestingly, is qualified in one case as wnḏw, which probably designates a community of peers sharing the same social values. Whether this sociability was informal and developed according to circumstances or was a regular and structured mode of collective relationship is unknown. It would be worth analysing the documentation to find traces of that sociability of peers.

In conclusion, the corpus of the Great Tomb Robberies sheds light on a part of the ancient Egyptian society which otherwise left few traces. The context in which these people evolved is certainly particular, but any socio-economic crisis inevitably tells us something about the society in which it occurs.