1The round table discussions organized by Gwenola Graff and her colleagues within the framework of Prehistories of writing: Iconography, graphic practices and emergence of writing in predynastic Egypt provided a suitable environment in which to recommence the topic of pre-firing applied potmarks and their containers found in proto- and early dynastic, viz., Naqada IIIB-D contexts in Egypt, mainly – though not exclusively – deriving from the mortuary realm.

2The relevance of foresaid potmarks as a non-verbal signification system, developing in tandem with the process of the emergence of writing in early Egypt, has been discussed most recently by van den Brink (van den Brink 2010: 334). This process, evolving from pictographic, through ideographic to eventually phonetic representation, is recorded in different media, e.g. in rock art, in the painted decoration on certain types of pottery vessels, on stamp and cylinder seals and sealings, bone, ivory and wooden labels, ink inscriptions etc. To date this process finds its first condensed, tangible expression in the Schriftzeugnisse from Tomb U/j and neighboring tombs at Umm el-Qa’ab, Abydos, dated to the Naqada IIIA1/2 (Dreyer 1998, Hartung 2001).

3The current paper presents and discusses potmarks applied to one particular class of ceramic vessels only, namely that of ‘wine jars’ (Petrie’s types 75-76 (Petrie 1953); van den Brink’s types III and IV (van den Brink 1996) deriving from the excavations carried out in the past by Z.Y. Saad, resumed in 1997 by E.C. Köhler and her team in the cemetery of Helwan. A group of 148 intact, marked wine jars is discussed here for the first time, against the backdrop of related data culled from four selected, partly contemporary cemeteries, viz. those of Minshat Abu Omar, Kafr Hassan Dawood, Abu Roash and Tarkhan, in the hope of shedding further light on the largely still enigmatic meaning of the potmarks.

The Helwan excavations

4In the early 1950s Z. Saad unearthed two adjacent subterranean tombs within an area of the early dynastic cemetery at Helwan as part of an extensive excavation carried out over numerous seasons, which to date remain largely unpublished. These two tombs were subsequently used by the excavator as repositories for large amounts of intact pottery and stone vessels as well as selected human remains, deriving mainly from his 9th and 11th seasons of excavation in the cemetery.

5During renewed excavations in the Helwan cemetery under the auspices of Macquarie University, Sydney in 1997/1998 and 2002/2003, Köhler located and re-opened those tombs in an area designated Operation 3 (Köhler 2005). Together they held an estimated 1735 pottery vessels of different types (Smythe 2004: 317, 2008: 151), including 106 mostly intact, wine jars with potmarks¹ which form the bulk of the collection under discussion.

The Helwan wine jars bearing potmarks

6This study considers these vessels and an additional 14 (Appendix 1²: tables 1-2), which were also excavated under the supervision of Saad, but which were selected by him for display in the former on-site museum at Helwan/Ezbet el-Walda. Out of safety concerns, this museum’s entire collection was transported in 1967 in 158 wooden crates to the basement of Egyptian Museum in Cairo, where they lay forgotten until their recent re-discovery by Köhler (Köhler 2002, Köhler 2004b).

7Of these 120 vessels deriving from Saad’s work in the Helwan cemetery, 86 still have tomb allocation numbers (nos. 1-86 in Appendix 1: table 1) that link them with 62 different graves. The remaining 34 either lack any identifying numbers or these have become illegible over time (nos. 87-120 in Appendix 1: table 2). Therefore, the vessels cannot be allocated to any specific grave. Together with 28 wine jars bearing potmarks (nos. 121-148 in Appendix 1: table 3) from eight graves excavated more recently by Köhler (Köhler 2004a) in an area designated Operation 4, both groups of vessels are discussed below as a corpus of 148 wine jars and their potmarks deriving from this cemetery. Of this collection, 32 jars with potmarks and six that did not bear potmarks have already been published elsewhere, some only cursorily and others in more detail (cf. Appendix 1: tables 1-3; Saad 1947: figs.11-12, 1951: Pl. 71.3-4, Köhler & van den Brink 2002: fig. 1, 1-4, Köhler 2004a: figs. 3, 4, 7b, Köhler 2004b: fig. 6, Smythe 2004: figs. 11-12, 2008: figs. 1-2, 4, 15-16, 18, 28, 32, 37-38, van den Brink 2010: figs. 1-2).

Wine Jars from Operations 3/1 and 3/3

A. Wine jars with original tomb allocation numbers still preserved on them

8Ninety-nine intact wine jars, including 86 with potmarks and 13 others, unmarked, were retrieved from “storage tombs” Operation 3/1 and 3/3 (Appendix 1: table 1). Based on the preserved original excavation tomb number on each of these jars, they must have come from 62 distinct graves.

1. Types II and III winejars with incised serekh sign
Jarres à vin de types II et III, incisées de serekhs

9The distribution of wine jars with potmarks by grave, based on the corpus under discussion³ is as follows:

• 53 graves each contained a single jar

• 3 graves contained 2 jars

• 3 graves contained 3 jars

• 3 graves contained respectively 4, 6 and 8 jars

B. Wine jars without proveniences

10Thirty-four unprovenanced wine jars with potmarks derive from an unknown, but maximum, number of 34 graves (Appendix 1: table 2). The number of graves for this quantity of wine jars might be lower, however, as it is likely some of the jars were found in groups of two or more.

Operation 4

11With work in this area still ongoing, to date 28 wine jars with potmarks and two similar, unmarked jars, that were analyzed for this study, derive from eight graves (Appendix 1: table 3). In contrast to the 120 jars with potmarks noted above, the full archaeological context of each of the jars from this Operation is known.

12The distribution of wine jars with potmarks by grave is as follows:

• 3 graves contained a single jar

• 1 grave contained 2 jars

• 2 graves contained 3 jars

• 1 grave contained 6 jars

• 1 grave contained 11 jars

Wine jar types represented in the Helwan assemblage

13The earliest wine jar types in the present corpus, characterized by either pushed-up, vestigial, indented ledge handles (n=1) or by impressed, crescentic decoration on the shoulder (n=8) are few in number (Appendix 1: tables 1-2, nos. 7, 9, 45, 61, 73, 85, 86, 116, 117). They are ascribed to van den Brink’s jar Types II and III respectively, and date from Late Dynasty 0 to the very beginning of Dynasty 1 (van den Brink 1996: 144-147). Four of these jars are incised with a serekh.

2. Type IV winejars with [‘g.t] Sm’j sign
Jarres à vin de type IV avec le signe [‘g.t] Sm’j

3. Type IV winejars ḥw.t s3-ḥ3 Adjib estate
Jarres à vin de type IV ḥw.t s3-ḥ3 de l'état d'Adjib

14The vast majority of intact wine jars found at Helwan, however, date from the second half of the First Dynasty to sometime during the Second Dynasty (Smythe 2004: 323). They consist of tall exemplars with three (n=16), two (n=38), one (n=66) or without any (n=19) continuous plain bands applied to their tapering bodies, types classified by van den Brink as IVb-d (van den Brink 1996 : 148). Most jars are made from alluvial Nile silt, with a significantly smaller percentage made of marl clay (cf. Appendix 1: tables 1-2; Smythe op. cit.)⁴. Not a single specimen of the bulkier, Type IVa jar, datable to the end of Dynasty 0 and very beginning of Dynasty 1, has been identified amongst the group at issue. Based on the information deriving from Operations 3 and 4 only, there appears to be a gap in the use of the Helwan cemetery covering the first half of the First Dynasty, although this may well be due to the lack of published materials from other areas of the cemetery.

Composition of the Helwan potmarks

15The Helwan corpus includes 27 potmarks on wine jars that are legible (cf. below, table 3). These comprise the four jars incised with a serekh already noted above (fig. 1; Appendix 1.1-.2: 7, 84, 85, 116) and 15 jars bearing the floral mark [‘g.t] Šm’j (U.E. barley) (fig. 2; Appendix 1: tables 1-2: 20, 22, 26, 34, 35, 45, 60, 70, 72, 73, 81, 93, 96, 104, 120). These have been discussed elsewhere at length (Köhler & van den Brink 2002, van den Brink 2010). Four additional jars are inscribed with the ḥw.t s3-ḥ3 estate-sign of A[ne]djib, all of which have accompanying marks, two of which (nos. 5 and 118) are identical (fig. 3; Appendix 1: tables 1-2: nos. 5, 6, 64, 118). Five jars bear the domain-sign of Semerkhet (fig. 4; Appendix 1: tables 1-2: nos. 1, 3, 4, 119, 120)⁵, one (no. 4) without the usual crenellated oval frame, and another (no. 120) notably associated with the Šm’j-sign.

4. Type IV winejars with Semerkhet domain
Jarres à vin de type IV avec la marque de Semerkhet

16Excluding these 27 legible inscriptions, the remaining 121 potmarks (figs. 5-6) in the corpus under discussion can be subdivided into marks consisting of a single sign (n=31), or different combinations and quantities of signs including two (n=63), three (n=22), four (n=4), and, in one case, eight distinct, usually horizontally aligned signs, thus forming composite or compound potmarks. Of these, seventy-three simple and compound potmarks⁶ are already attested elsewhere as documented in van den Brink’s 1992 corpus, and annotated in Appendix 1: tables 1-3. In the absence of a paleography, 32 additional simple and compound potmarks could not be attributed with 100 % confidence to either a specific known potmark consisting of a single, basic sign or a combination of two or more basic signs as categorized in the 1992 potmark corpus, because they can be interpreted as minor paleographic variations of attested signs or their combinations, or else must be considered to be new, previously unrecorded signs and their combinations. Amongst these are several compound potmarks that are known on basic sign level, but not on compound sign-group level; several others consist of one recorded and one previously unrecorded sign; nearly a dozen are new (that is, previously unattested), compound potmarks, each comprised of two signs, both of which are known as distinct, basic signs but thus far had not been recorded in combination with one another; several others are completely new signs, one in combination with Šm’j, a sign consciously excluded from the 1992 potmark corpus, but incorporated in the 2007 data base of the www.potmark-egypt.com as basic sign group group CI (see van den Brink 2010: 331). More definitive assessment of these observations is not possible without the aid of a paleography of all published early dynastic potmarks, a resource not yet realized (see also van den Brink 2011: 1010).

17Several simple and compound potmarks occur more than once in the Helwan corpus: there are three compound potmarks each consisting of two distinct signs (van den Brink 1992 Groups I.9, I.30-.33 and VII.11), each attested twice or trice (Appendix 1: tables 1-2: nos. 78, 94; 24,30, 54; 58, 76 respectively); one triple combination (Group I.58) is attested twice (Appendix 1: tables1-2: nos.5 and 118)⁷; two single signs (Appendix 1: tables 1-2: nos. 31, 50, 82 and 91, 136) are attested three and two times, respectively. Most frequently appearing basic signs are Šm’j (n=15; Group CI), a square (n=15; Group I), a fish (n=7; Group XXVI), a pair of arms (n=6; Group III), a tree-like sign (n=4; Group IV), a moon or star (n=3; Group XIX), a pair of snakes (n=2; Group XXV) and the city sign (?) niw.t (n=2; Group XXXI).

18It has been noted in the past (van den Brink 1992: 271) that in 35.2% (or 25 records) of all (n=71) attested compound potmarks that consist of a combination of three distinct signs (*ABC) within the 1992 corpus, at least one of three possible double combinations (*AB [or *BA], *BC [or *CB] and *AC [or *CA]) is observed, in 31 % (22 instances) of the cases two out of three possible double combinations is observed, and only in 14.1 % (10 instances) of the cases all three possible double combinations contained in any individual triple combination are known, while in 19.7 % (14 records) only the triple combinations are recorded, while none of the three possible double combinations within each triple combination are attested. Helwan wine jars nos. 18 and 23 supplement the 1992 record in this respect: in the already previously recorded triple-compound potmark Group IV.41 (*ABC), only two of the three theoretically possible double combinations (IV.2 [*AB] and IV.28 [*AC]) had been attested prior to the rediscovery of Helwan material. Helwan wine jar potmark no. 18 constitutes the unattested third and possible combination of *BC within this triple compounded potmark. Helwan wine jar potmark no. 23 presents the reverse side of the same coin: it presents the otherwise unattested triple combination (*ABC), that consists of three previously attested double combinations (*AB, *AC, *BC, in this case sign Groups IV.2, IV.49 and VI.16).

19The illustrations accompanying this paper begin with the wine jars and their potmarks from Operations 3/1 and 3/3, starting with the 27 legible marks distributed across four groups (figs. 1-4), listed in the order of appearance in table 3. They are followed by the remaining jars of the aforementioned Operations, organized in the order of their appearance in Appendix 1: tables 1-2, (figs. 5.1-5.6). These are then followed by the 28 jars of Operation 4 (listed in Appendix 1: table 3; figs. 5.6-5.7), which chronologically are also the latest exemplars, for the most part, of wine jars attested in the corpus. Finally, all 148 potmark drawings of these are presented together in fig. 6, in order of their appearance in Appendix 1: tables 1-3.

Comparison of the distribution of jars with potmarks by grave between the cemetery at Helwan and selected contemporary cemeteries in Lower Egypt

20Although crucial, contextual information concerning actual tomb locations and possible additional funerary gifts associated with the wine jars excavated by Saad are mostly lacking, we suggest that a comparison of available data on the numeric distribution of wine jars with potmarks per grave in Helwan with that of other cemeteries may, nevertheless, be informative. Comparison may aid the discernment of quantitative patterns in the deposition of wine jars in graves that, in turn, could perhaps shed light on the personal wealth or social status or persona of the deceased, if not members of the living associated with the deceased. Accordingly, tables 1-2 present pertinent, comparative data from the cemeteries of Minshat Abu Omar (based on Kroeper 2000; cf. Appendix 1: table 6), Kafr Hassan Dawood (based on Tassie et al. 2008; in prep.; cf. Appendix 1: table 7), Abu Roash (based on van den Brink 1992: 267-269, fig. 3), and Tarkhan (based on Mawdsley 2006, 2009; cf. Appendix 1: tables 4-5). The cemetery sample thus includes provincial cemeteries somewhat distant from Memphis, considered to be the core area where a unified Egyptian polity was developing, and cemeteries closer to that center. Although the cemeteries were all in use during the height of production and use of wine jars bearing potmarks (Naqada IIIB-C), some were established earlier as well as continued in use later than others.

21Table 1 indicates that graves containing marked pottery vessels form but a small percentage (3.7% – 23.8%) of the total number of graves within the selected cemeteries and that within this small group, graves containing wine jars with and without potmarks are a smaller percentage still (ranging from 30% – 42% of all graves containing marked pottery vessels) of the total number of graves⁸.

22The information from table 2 shows that the majority of the early dynastic graves (44.8 % – 80 %) that include wine jars with potmarks, contain only one such jar. The data further suggest that a significantly lower percentage of graves (38.4 % – 14.2 %) contain 2-4 such jars, while only a small percentage (3.9% – 17.4%) contain between 5 and a maximum of 23. It is tempting to interpret such differences in groups of grave goods as evidence of social distinctions between the interred. In many instances there is a correlation between tomb size and number of funerary gifts included; e.g., the two graves at Minshat Abu Omar that yielded the largest number of wine jars, Graves 1590 and 2275 (with respectively 25 and 16 wine jars), are relatively large, mudbrick structures, referred to by their excavators as elite-tombs (Kroeper 1992: table 2). A similar case also occurs at Kafr Hassan Dawood, where the two largest mud-lined graves 913 and 970 produced the greatest number of wine jars, and in this case also the largest number of potmarks. This would suggest that the greater the number of marked jars, the higher the social standing of the owner of the tomb. This interpretation seems logical given that the greatest number of relatively small graves, representing the mass of the population of the cemetery contained only one or no wine jars accompanying the deceased to the after-life (cf. table 1).

23Continuing the cross-cemetery comparison, table 3 lists the totals of wine jars with legible inscriptions similar to those found at Helwan (n=27; see above), deriving from four additional cemeteries in Lower Egypt. The distribution of wine jars in Egypt with either A[ne]djib’s estate sign or Semerkhet’s domain sign is very restricted, with the majority of both former (55) and latter (116) deriving from their respective tombs situated in the royal necropolis in Umm el-Qa’ab, Abydos⁹. Inscriptions with the estate name of A[ne]djib and the Semerkhet domain are unattested at the cemeteries at Kafr Hassan Dawood and Abu Roash. The A[ne]djib estate sign is also absent from the Tarkhan cemeteries. In almost all cases legible signs are accompanied by additional single or compound potmarks that are less intelligible.

24Excluding the 27 inscriptions listed in table 3, the remaining 121 potmarks in the corpus under discussion can be subdivided, as already mentioned above, into potmark groups consisting of one, or combinations of 2 or more distinct signs (up to a maximum of 8). The way in which this distribution compares to those recorded for the additional selected cemeteries is shown in table 4.

25Table 4 indicates that the vast majority of potmarks are those that either consist of a single sign or a composite of two distinct signs. Compound potmarks consisting of two signs are slightly more numerous than those consisting of a single sign within the group of selected cemeteries, although some cemeteries have more vessels bearing single signs than others. At Minshat Abu Omar and Helwan, for instance, compound potmarks consisting of two signs are more numerous than those consisting of a single sign. The importance of this patterning, however, does not concern so much a specific cemetery or groups of cemeteries; it rather reflects something about the internal organization and composition of potmarks and perhaps possible associations with administrative institutions.

26Compound marks of three individual signs make up a minor, though still significant portion of all recorded potmarks in the cemeteries surveyed above, while compounds of four and up to eight distinct signs are rare.

Additional Recently proposed legible potmark signs

27Readings or interpretations have recently been suggested for several other incised potmark groups based on their apparent similarity to some of the slightly later hieroglyphs (Anselin 2007, 2008, 2010a-b, 2011a-b, Bréand 2008, Kroeper 2000, Mawdsley 2006, 2009, Tassie et al. 2008: 215-217), to the extent that they have been called “preformal hieroglyphs” by some (Tassie et al., op.cit.: 217-218). Several of these also occur in the Helwan corpus. They include the square or rectangle (n=33; van den Brink’s [1992] Group sign I.1), suggested reading ḥw.t [cf. Kroeper 2000:216] or pr [Mawdsley 2006; see also Anselin 2011b). In the Helwan assemblage this sign either appears alone (n=2) or in combination with one or two additional signs (n=31), e.g., in combination with a fish (n=6), a frequent combination (van den Brink’s [1992] Sign sub-groups I.30-I.32 (n=87) which Kroeper – in analogy to certain Old Kingdom onomastica – suggests represent an estate name in the Delta (ibid.; Anselin 2011a); another frequently occurring mark concerns the generic, non-perpendicular, crois simple (n=12) interpreted as a mnemonic device in the production of pottery vessels by Bréand (Bréand 2008).

28Other signs appearing in the Helwan corpus that perhaps can be read or are similar to the slightly later hieroglyphs, include van den Brink’s (1992) Sign groups III.1 (n=6; k3 or sḫn), IV.1 (n=5), X (n=2; 3bd “lunar festival”; see Anselin 2010b), XXXI.1 (njw.t), appearing twice in the Helwan corpus in combination with another sign and XXXVII (n=1) in combination with a serekh. Notably absent from the Helwan corpus are the frequently recorded signs for e.g. nṯr (Group XI) and mr (Group XVII), and the ḥs-vase like sign Group X and Group XXI.

Conclusions

29It is difficult, perhaps too early to comment at this stage on the possible wider ramifications of the analyses and resultant trends emerging from this small and incompletely preserved aspect of early dynastic mortuary behavior. The bulk of data used for examining the wine jars and their potmarks, and information concerning their distribution among graves in the Helwan cemetery as presented here derive from Z.Y. Saad’s excavations. In the absence of relevant field records or publications, these data are hopelessly inadequate to work with. These circumstances preclude a deeper investigation into how, for example, wine jars bearing potmarks and their distribution over parts of the Helwan cemetery relate to e.g., other vessel types and their distribution, to other tomb contents, tomb size, location within the cemetery, skeletal remains if preserved, sex, gender, age, etc. Thus, attempts to cast additional light on the social status of the owners of those tombs or others affiliated with them (cf. e.g., Parker-Pearson 1999) are severely hindered. Kroeper’s (Kroeper 2000: 215) observation that “the highest amounts of marks was not found in the grave with the highest amount of offerings” stands as a caveat to guard against assuming simple correlations between quantities, size and social status. Past cultural frameworks of significance might be more complex than one might initially expect.

30Although the contextual analysis of a class of grave goods, even within a single cemetery—in this case of wine jars bearing potmarks— goes well beyond the scope of the present paper, we could begin to overcome the problem of incomplete archaeological records for Helwan by investigating comparable data from the four selected contemporary cemeteries for which field records and published excavation reports do exist. It is therefore hoped that the selected data presented here might encourage others to continue to pursue potmark research in new ways and build on the groundwork laid by this study.

31As discussed above, the various groupings of both legible and intelligible potmarks are slowly, but gradually expanding, and the application of the term preformal hieroglyphs to several of these seems fitting. With the renewed excavations at the Helwan cemetery still ongoing, the present assemblage of wine jars with potmarks will undoubtedly continue to expand and contribute not only to the theme of early dynastic potmarks per se, but also to contextual and social analyses as well as script development more generally.

5-1. Winejars in the order of appearance as in Appendix 1: tables 1-3
Jarres à vin dans l’ordre d’apparition de l’Annexe 1 : tableaux 1-3

5-2. Winejars in the order of appearance as in Appendix 1: tables 1-3
Jarres à vin dans l’ordre d’apparition de l’Annexe 1 : tableaux 1-3

5-3. Winejars in the order of appearance as in Appendix 1: tables 1-3
Jarres à vin dans l’ordre d’apparition de l’Annexe 1 : tableaux 1-3

5-4. Winejars in the order of appearance as in Appendix 1: tables 1-3
Jarres à vin dans l’ordre d’apparition de l’Annexe 1 : tableaux 1-3

5-5. Winejars in the order of appearance as in Appendix 1: tables 1-3
Jarres à vin dans l’ordre d’apparition de l’Annexe 1 : tableaux 1-3

5-6. Winejars in the order of appearance as in Appendix 1: tables 1-3
Jarres à vin dans l’ordre d’apparition de l’Annexe 1 : tableaux 1-3

5-7. Winejars in the order of appearance as in Appendix 1: tables 1-3
Jarres à vin dans l’ordre d’apparition de l’Annexe 1 : tableaux 1-3

6-1. Corpus of Helwan potmarks
Corpus des marques incisées d’Hélouan

6-2. Corpus of Helwan potmarks
Corpus des marques incisées d’Hélouan

6-3. Corpus of Helwan potmarks
Corpus des marques de potiers d’Hélouan

6-4. Corpus of Helwan potmarks
Corpus des marques de potiers d’Hélouan