Egypt Exploration Society

The Middle Kingdom Offering Formulas: A Challenge Author(s): Detlef Franke Source: The Journal of Egyptian Archaeology, Vol. 89 (2003), pp. 39-57 Published by: Egypt Exploration Society Stable URL: http://www.jstor.org/stable/3822491 . Accessed: 14/10/2013 09:39
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp

.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

Egypt Exploration Society is collaborating with JSTOR to digitize, preserve and extend access to The Journal of Egyptian Archaeology.

http://www.jstor.org

This content downloaded from 134.58.253.30 on Mon, 14 Oct 2013 09:39:32 AM All use subject to JSTOR Terms and Conditions

THE MIDDLE KINGDOM OFFERINGFORMULAS A CHALLENGE*

By DETLEF FRANKE
It is argued that there was no Middle Kingdom re-interpretationof the offering formula. Old and Middle Kingdom offering formulas show no basic differences in spelling, and they should be interpretedas 'an offering which the king has given (and) Osiris has given:...' There are no convincing examples from the Middle Kingdom to the Eighteenth Dynasty for a dative construction with n that would make the god(s) the recipient(s) of the offerings only. Parallel to the king, the gods are always the givers of the offerings, according to the custom of the reversion of offerings. The 'late' spelling of htp dj nswt is an expression of the profane scribal tradition, not predominantly in use before the Sixteenth Dynasty. The method for dating stelae developed by Bennett and refined by Satzinger can lead systematically to wrong results and should be abandoned.

n k3 nj, 'for the spirit of', title and name, etc.). There are several variants of spelling and phrasing occurring from its first appearancein the early Fourth Dynasty down to the latest examples from the later Roman Period. Because offering formulas are so typical and are found on countless inscribedEgyptianobjects, they are generally passed over by Egyptologists rather automaticallyand without hesitation. On the other hand, they are a stumbling block of a certain vexing fascination for Egyptologists, and an Egyptological debate to solve the riddles of their tricky nature in the Old and Middle Kingdoms continues.

the king's formula(htp(r)dj(w) god'sformula(oneor more nswt),the closelyassociated name(s)of divinebeings),the requests (fromthe TwelfthDynasty usuallyintroduced by and the of the of the favours naming recipient (introduced requested by e.g. dj=fls/sn),

THEmost common type of text on Middle Kingdom stelae is a sequence of phrases called by Egyptologiststhe 'offering formula'. It consists regularlyof four parts, namely

The 'king's formula' and the associated'god's formula'

Stela Cairo CG 20318 from the first half of the ThirteenthDynasty has an offering formula with htp dj repeatedthree times and at least four gods of Elkab, Hierakonpolis and Edfu are mentioned. At first sight, it looks strange for the period. The offering formula on the ratherrepellent late Twelfth Dynasty (?) stela British Museum EA 205

(htpdj nswtdj wsjrj...),1with the dj (

beforethe god's name, is another ) repeated

variant rarely found elsewhere on late Middle Kingdom stelae. It is found similarly,
* I thank H. Satzinger and G. Lapp for very critical and helpful remarks on a draft of this article, and I am grateful to the JEA referees and L. Montagno Leahy for suggestions and correcting my English. 1 H. R. Hall, in E. A. W. Budge (ed.), Hieroglyphic Textsfrom Egyptian Stelae, &c., in the British Museum, II (London, 1912), pl. 48; R. Parkinson, Cracking Codes. The Rosetta Stone and Decipherment (London, 1999), 151 (no. 65).

This content downloaded from 134.58.253.30 on Mon, 14 Oct 2013 09:39:32 AM All use subject to JSTOR Terms and Conditions

40

DETLEFFRANKE

JEA 89

writtenwith the single or double 'forearm with hand' (Gardiner sign list D36, 37-8) insteadof the usualA for the seconddj beforethe god's name,on at least five stelae fromthe Sixteenth andearlySeventeenth Dynastyas well.2 From the late Old Kingdomto the early Middle Kingdom,this repetitionof the is a relatively commonfeature.3 Thesespellings element(s) (htp)dj in offeringformulas could mirrorthe interpretation of the commonofferingformulain MiddleKingdom ventured times,whichwas-this is my hypothesis here-virtuallythe sameas in the Old in the tradition of A. H. Gardiner's not, as is usuallytranslated Grammar3, Egyptian p. 171, 'an offering(or: boon)whichthe kinghas given (to) Osiris,...'4 In the Old andMiddleKingdoms the equipment for tombsandfuneral of at buildings least the members of the elite derived(oftenonly theoretically) fromthe favourof the for a 'good king,5and the gods, such as Anubis,for example,were divineguarantors burial'.It is generally that the Old formulas describe boththe accepted Kingdom offering the and as of the and and that the food god(s) givers king 'offerings' favours, offerings to the gods derivefromroyalestatesandmagazines. Thephrase(s) presented htpdj nswt and (htp dj) god's name (abbreviated as GN) could be understood as a formulaic of the traditional customof the reversion of offerings('Umlaufopfer'): the description offers 'food to the and after have 'satisfied' offerings', king htp, gods, they (htp)
themselves, the offeringsare passed on from the (altarsof the) gods to the offering-tables, stelae and/orstatuesof humanbeings to nourishtheir owners.6heir thus describes ownerformula
BM EA 630 (time of Sekhemre-KhutowyPaentjeni), Cairo CG 20335 and Louvre E13057 (C287; reign of Rahotep; D. Franke, 'An ImportantFamily from Abydos of the Seventeenth Dynasty', JEA 71 (1985), 175-6, pl. xix, left), T. E. Peet, The Cemetries of Abydos, II (MEEF 34; London, 1914), 115 fig. 73, pl. xxiv, 2 and 4 (no. 13), and KhartoumMuseum 4448 from Serra East (F. LI. Griffith, 'Oxford Excavations in Nubia', LAAA8 (1921), 98, pl. xxix, 1; PM VII, 128, early Seventeenth Dynasty). All have the 'late' type II of the king's formula, on which see below. See W. Barta, Aufbau und Bedeutung der altagyptischen Opferformel(AgFo 24; Gliickstadt, drawin 1968), 55 n. 1, 74 n. 1 (for BM EA 205, his reading dd derived from a mistake in thedrawin Hieroglyphic Texts II, pl. 48), 82. The dynastic divisions are after K. Ryholt, The Political Situation in Egypt during the Second IntermediatePeriod c. 1800-1550 B.C. (CNI Publications 20; Copenhagen, 1997). 3 G. Lapp, Die Opferformeldes Alten Reiches (Mainz, 1986), 28-9 46; H. 0. Willems, 'Food for the Dead', in W. H. Van Soldt (ed.), Pap Uit Lemen Potten. Papers in Memoriam of Mia Pollock (Leiden, 1991), 98-108; e.g. P. E. Newberry, Beni Hasan, II (ASE 2; London, 1894), pl. vii; A. M. Blackman, The Rock Tombsof Meir, III (ASE 24; London, 1916), pi. ix. See below on types A, B, D and E. In the late Twelfth and ThirteenthDynasty and later, htp dj or dj are but rarely repeated; see the examples cited below as type E. 4 See already N. De Garis Davies and A. H. Gardiner, The Tomb of Amenemhet(Theban Tombs Series 1; London, 1915), 79-93, esp. 89, where Gardineraccepted the interpretationforwarded by G. Maspero, Etudes de mythologie et d'archeologie egyptiennes, VI (BE 28; Paris, 1912), 366-7. Barta, Opferformel, 267, 270, 293, translated 'der Konig gibt ein Opfer (, namlich das Opfer, das) dem Gott NN (gegeben wird)', and maintained 'Die Bedeutungder Ritualformelbleibt ... durch alle Zeiten hindurchunverandertbestehen: der Konig spendet der Gnaden zu erlangen', so that the god's formula then would be merely Gottheit ein Opfer, um von ihr ruickwirkend an apposition of the king's formula (see p. 265). P. Grandetand B. Mathieu, Cours d'egyptien hieroglyphique, II (Paris, 1993), 104-5, explain the offering formula as consisting of three prospective forms (dj, htp, dj=f/s/sn), which is not very convincing. 5Lapp, Opferformel, 103-7, and D. Franke, Das Heiligtum des Heqaib auf Elephantine (SAGA 9; Heidelberg, 1994), 21. For the concept of jm,h, 'provisioning/veneration', see K. Jansen-Winkeln, 'Zur Bedeutung von jmh', BSEG 20 (1996), 29-36. The official's reward with provision and burial equipment is the material effect of his high repute and being in favour with his lord. 6 Comparee.g. Gardiner,EgyptianGrammar3, 170-2; Barta, Opferformel,268-9; H. Satzinger, 'Beobachtungen zur Opferformel:Theorie und Praxis', LingAeg 5 (1997), 182; H. Altenmuiller,'Opferumlauf, LA IV, 596-7.
2

Kingdom: 'an offering which the king has given and which Osiris has given ...', and

This content downloaded from 134.58.253.30 on Mon, 14 Oct 2013 09:39:32 AM All use subject to JSTOR Terms and Conditions

2003

FORMULAS MIDDLEKINGDOM OFFERING

41

the performance the theoretical during presented originandsourcesof the food offerings the kingand/orthe gods. of the offeringritual: A dativeconstruction used in the MiddleKingdom? G. Lapp and H. Satzingerhave shown convincinglythat there was no dative times.7Thereis no fundamental construction withn orjn in use in OldKingdom graphic all the variants,to be observedin the king's formula even acknowledging distinction, htp dj nswt and the associatedgod's formula(htp dj) wsjrjbetweenthe Old and the in of bothpartsof the formula the basicmeaning versions.Therefore, MiddleKingdom bothperiodsshouldhave beenthe same.8 the god's name at least for Gardineraccepteda dative construction introducing of the Lateand Ptolemaic certainofferingformulas Periods,andBarta,Lappand Satin Middle Kingdomtimes.9In fact, the zinger were convincedof its existenceeven htp dj n GN by Bartaand Lappderive examplesadducedfor a dative construction versions textsof sundrycoffinsfromAssiut,andvariant mainlyfromsome ornamental were also used as partof the so-calledstarclocks at Thebes, of theseofferingformulas This peculiartype of offeringformulawent with the modelof Gebeleinand Aswan.10 the star clocks to Deir el-Bahari(T3C), Gebelein(GIT and G2T) and Elephantine of the starclock'smodelmust thattheperiodof thecomposition (A1C). Willemsnoted'1 have been before the reign of NebhepetreMontuhotep, perhapsin Tenth Dynasty date fromthe secondhalf of the on coffins Lower but the copies Egypt, Herakleopolitan are versionsonly of These 'Assiut' formulas Twelfth and the Eleventh Dynasty.12 early
7 Lapp, Opferformel, 30-8, and Satzinger, LingAeg 5, 180-2. Compare Gardiner, Egyptian Grammar3,171, 'in the Old Kingdom the king and whatever god is named are mentioned in parallelism with one another as givers of the boon or boons rewarded', and already Gardiner, The Tomb of Amenemhet, 81 2. See also H. G. Fischer, 'Occurrences of jn, Agential and Dative', GM 107 (1989), 69-75. R. J. Leprohon's note on the evolution of the offering formula ('The Offering Formula in the First IntermediatePeriod', JEA 76 (1990), 163-4) is obsolete if there was no dative construction used in the Middle Kingdom. 8 This conclusion was proposed by H. Willems in 1991 in Van Soldt (ed.), Pap Uit Lemen Potten, 104, and later in The Coffin of Heqata (OLA 70; Leuven, 1996), 44-5 n. 15, 332-4. However, he did not deal with the problem of a dative construction. See, too, the translation adopted by M. Maree, 'A Remarkable Group of Egyptian Stelae from the Second IntermediatePeriod', OMRO73 (1993), 8. 9 Gardiner, The Tomb of Amenemhet, 89, and Egyptian Grammar3,171: 'A series of variants shows that the divine name which follows the phrase htp di nsw was now understood as a dative, though it is only at a far later period that the preposition n was inserted'; Barta, Opferformel, 37, 224, 265; Lapp, Opferformel, 32-3 50; W. Barta, 'Zur Bedeutung der Opferformel im Alten Reich', GM 96 (1987), 8, and jn als Pleneschreibung der Praposition n', GM 103 (1988), 10-11; Satzinger, LingAeg 5, 184 n. 34. 10Barta, Opferformel,37, cited for htp dj n GN the coffins of Mezehti, Cairo CG 28118 (star clock horizontal strip text; SIC) and CG 28119 (ornamentaltext; S2C) in P. Lacau, Sarcophages anterieurs au Nouvel Empire, II (CG; Cairo, 1906), 110, 131, and the coffin of Khui (Cairo JE 36445; S4C), published by E. Chassinat and Ch. Palanque, Une campagne des fouilles dans la ne'cropoled'Assiout (MIFAO 24; Cairo, 1911), 157, adding (p. 45 n. 1) the star clock horizontal strip text on the coffin of Aashyt from Deir el-Bahari (C. Fievez, 'Les trois calendries inedits d'Assiout d'apres un article recent d'Alexandre Pogo', CdE 11 (1936), 358-9 (T3C)). See Lapp, Opferformel, 32-3 50; G. Lapp, Typologie der Sdrge und Sargkammernvon der 6. bis 13. Dynastie (SAGA 7; Heidelberg, 1993), 126-9 289; Willems, Coffin of Heqata, 328-34. 11Coffin of Heqata, 331-2. 12Seventeen examples of star clocks on coffins are known (K. Locher, 'Middle Kingdom Astronomical Coffin Lids', in C. J. Eyre (ed.), Proceedings of the Seventh International Congress of Egyptologists, Cambridge, 3-9 September 1995 (OLA 82; Leuven, 1998), 697ff.; C. Leitz, Altdgyptische Sternuhren(OLA 62; Leuven, 1995),

This content downloaded from 134.58.253.30 on Mon, 14 Oct 2013 09:39:32 AM All use subject to JSTOR Terms and Conditions

42

DETLEFFRANKE

JEA 89

god's formulasalone, used withoutthe normalprefix nswt, i.e. withoutthe king's andsometimes no further or requests andno recipient werementioned. formula, offerings to Lapp'sopinion,andthe statement I see no reasonwhy these of Barta,13 Contrary examples, which are graphicallyequal to the Old Kingdom versions of Lapp's andtranslated as perfective 48, shouldnot be analysed Opferformel, sdm.n=f relative forms:'an offeringwhich(the god) Re has given...' (htp(r)dj.n-rCw...). This was also the conclusionsuggestedby Harco Willems in his study of the star clock and its horizontal on the coffin of Heqatafromthe Qubbet el-Hawa(A1C), offeringformulas thatthe reading as a relativeformis impossible to prove. thoughhe rightlyremarked these 'Assiut' and some others have led to an with Nevertheless, examples argument and were used a Middle far-reaching consequences, they by BartaandLappto postulate of the offeringformula.Barta,to sustainhis statements Kingdom re-interpretation (pp. n GN for the Middleand 224, 265), offeredsomeallegedexamplesof the construction New Kingdoms. thereis not much However,if one checkshis examplesscrupulously, for two on a late Twelfth14 and an Thirteenth left, except uniquespellings early Dynasty fromAbydos,whichlook suspicious andfaulty. offering-table15 Bartafurther citedP. Ramesseum E, 11.57-8: 'an offeringwhichthe kinghas given to/for (n) the many westerngods'.16Whatprecedesis lost and no other recipientis butthismightbe an example of theuse of htpdj nswtas a labelfor the ritual mentioned, a on behalf of the label 'making performed by priest gods;compare htpdj nswtfor (the This is all the evidenceBartaofferedfor the Middle gods of) the Northern chapels'.17 considerhis threeexamplesfor the Kingdom,and I know of no other. Let us further Seventeenth andEighteenth Dynasties. The rishi-coffin of KingSeqenenre Tao, CairoCG 61001, was foundin TT 320, the cachette of Deir el-Bahari.18 The coffin was deprivedof its gold inlays and the
58ff.; J. Kahl, 'Textkritische Bemerkungen zu den Diagonalsternuhrendes Mittleren Reiches', SAK 20 (1993), 95-107; J. Kahl, Siut-Theben (Probleme der Agyptologie 13; Leiden, 1999), 197-201). Note that the spelling on the coffin of Khui (S4C) is clearly a mistake or an intrusionfrom the 'star clock formulas' common at Assiut: after Lapp, Typologie, Blatt 19-23, it is the only example with an n before Osiris among all east side/front offering formulas on coffins from Assiut. The offering formula on the opposite west side/back of S4C is spelled quite regularly. 13 Lapp, Opferformel,33 50; Barta, GM 96, 8; also J. A. Wilson, 'A Group of Sixth Dynasty Inscriptions', JNES 13 (1954), 261-2; 0. Neugebauer and R. Parker, Egyptian Astronomical Texts, I (London, 1960), 27 ('A boon which is given to Re...'). 14 Barta, Opferformel, 55, cited Cairo CG 23065: the offering formula with Geb has, contrary to the other formula with Anubis, n (htp dj nswt n gb(b) jmj-r3 tHmhw...), but the n k3 nj phrase is missing. See also D. Franke, Personendaten aus dem Mittleren Reich (AA 41; Wiesbaden, 1984), Dossier nos. 159, 710. 15 The offering-table Cairo CG 23210 was erroneously cited by Barta, Opferformel, 140 n. 4 as from the Nineteenth Dynasty. The photograph in W. K. Simpson, The Terrace of the Great God at Abydos: The Offering Chapels of Dynasties 12 and 13 (PPYEE 5; New Haven and Philadelphia, 1974), pl. 16 shows that the inscription is damaged in the area where A. Kamal, Tables d'offrandes, I (CG; Cairo, 1909), 146, gave an n, and the reconstruction is highly doubtful: note the spelling of Osiris, and the space between dj and the god's name. All other offering formulas on the monuments of ANOC group 7 (adding the 'truncated obelisk' BM EA 177) show the regular spelling. 16 Opferformel,74 n. 3; A. H. Gardiner, 'A Unique Funerary Liturgy', JEA 41 (1955), 13, pl. iv. 17 L. Borchardt, Das Grabdenkmaldes Konigs S'a3hu-Rec,II (WVDOG 26; Berlin, 1913), pl. 19, cited by Gardiner, The Tomb of Amenemhet, 88. 18 Cited by Barta, Opferformel, 82; G. Daressy, Cercueils des cachettes royales (CG; Cairo, 1909), 2, pl. i; PM 12, 658.

This content downloaded from 134.58.253.30 on Mon, 14 Oct 2013 09:39:32 AM All use subject to JSTOR Terms and Conditions

2003

MIDDLEKINGDOM OFFERING FORMULAS

43

was later restored,perhapsin the course of its depositionin the cachette inscription duringthe Twenty-first Dynasty. The gods' names had remainedfrom the original before them but inscription, htp dj nswt was restored,the htp-signunderlined by a horizontal strokethatlooks differentfromthe othern. Perhaps it is a mere space-filler or an examplefor the LatePeriodspellingof the offeringformula withan n. The coffin CairoCG 61005, reusedfor the mummyof KingAmenophis I, comes fromthe same It is a productof the Nineteenth cachette.19 Dynastyand was inscribedon behalf of I in late the withn as Amenophis only Twenty-first Dynasty,with an offeringformula a horizontalstrokebefore the god's name, as was en vogue at that period. A third examplefor n GN citedby Bartaas being fromthe Eighteenth Dynastyis, in fact, the on the baseof the colossalstatueof Amenhotep, son of Hapu,fromKarnak, inscription whichclearlyis not contemporary butdatesto the Twenty-first Dynasty.20 To sumup: thereis not a singleclearexamplefor a dativen beforea god's namefor the whole period from the late Eleventhto the end of the Eighteenth Dynasty.Only and growingevidencefor beginningwith the Nineteenth Dynastyis thereundeniable However,they are found offeringformulas spelledwith an n beforethe god's name.21 with formulas without an n. The late New spelledoffering contemporarily 'regularly' andlaterspellingswithhtpdj nswtn GN couldbe interpreted as an evidence Kingdom for a dativeconstruction and a late re-interpretation of the offeringformulas alongthe and principledo ut des. They can also be explainedas archaizing spellings,imitating the revivingOld Kingdomwritingswith an agentialn or <j>n, 'by', to introduce after a that not to the to the formula: subject participle, belongs king's formula, god's htp (r)djj(j)n nswt GN..., 'an offeringwhichwas given by the king (and)GN...'22K. Jansen-Winkeln explainsthe n simplyas the genitivaladjectiven(j): 'ein Konigsopfer des GottesNN (fir den NN)'.23A specialcase studywouldbe neededto decidethis question. It is puzzlingthat differentversions, with or withoutn, are displayedon contemin roughlyequal distribution accordingto Barta,but in the Old porarymonuments, even moredifferent versionsco-existedat the sametime. The question is why Kingdom
19 Opferformel, 86 n. 2; Daressy, Cercueils, 7, pl. vi; PM I2, 659. See F.-J. Schmitz, Amenophis 1. (HAB 6; Hildesheim, 1978), 14. 20 Opferformel, 108 n. 3. Cairo CG 1199: G. Maspero, Le Musee egyptien, II (Cairo, 1907), 35, pi. xiii; PM II2, 22 (6). All contemporary Eighteenth Dynasty inscriptions of Amenhotep, son of Hapu, display the regular spelling of the offering formula. 21 Barta, Opferformel, 140 n. 4, cited W. C. Hayes, The Scepter of Egypt, II (New York, 1959, reprint 1990), 351 fig. 219 (New York MMA 33.2.1), left vertical naos inscription, which could easily be translated as 'Anoffering-which-the-king-has-givento/for (n) Osiris Khentamenti(made) by (jn) the royal scribe...Yuny', making the god beneficiary of the offering ritual performed by Yuny. Two similar texts, with and without dative n to introduce the god, are found on the right side of the roughly contemporarynaos BM EA 1135 (Hieroglyphic Texts XII, pls. 88-9). Barta's second example is W. M. F. Petrie, Ehnasya 1904 (London, 1905), pl. xxvii, 1: two times htp dj nswt n pth... 22 See Lapp, Opferformel, 30-2 47-8; Satzinger, LingAeg 5, 180-2. Several examples are in K. JansenWinkeln, Agyptische Biographien der 22. und 23. Dynastie (AUAT 8; Wiesbaden, 1985), and see M. Abdelraheim, 'Ein Spatzeitwiirfelhockeraus dem Agyptischen Museum in Kairo (JE 38011)', GM 192 (2003), 15-18, also Barta, Opferformel, 163, 173, 186, 195, 204, 211. 23 K. Jansen-Winkeln, 'Vermerke: Zum Verstandnis kurzer und formelhafter Inschriften auf agyptischen Denkmalern', MDAIK46 (1990), 143-6, and id., Text und Sprache in der 3. Zwischenzeit(AUAT 26; Wiesbaden, ws proposed earlier by G. Farina (see Barta, Opferformel, 258-9). According to his 104ff, 1994). 104ffas interpretation,the gods are also those who give the offerings.

This content downloaded from 134.58.253.30 on Mon, 14 Oct 2013 09:39:32 AM All use subject to JSTOR Terms and Conditions

44

DETLEFFRANKE

JEA 89

as they do usuallyfor the human did not writethe dativen throughout, the Egyptians The rarity if theyreallyhadthought of sucha construction. of the formulas, beneficiary of the god's namewith n could of examplesshows, to my mind, thatthe introduction for adduced not havebeenthe ruleor the norm.The allegedMiddleKingdom examples or otherwise. as slips of the chisel or mistakes a dativen can be explained with an n are statedthatwritingsof the offeringformula FollowingBarta,Satzinger 'zu allen Zeiten belegbar', but he admits 'wenngleichsie [the n] normalerweise for a dativeconstruction haveto admitthatit was wird'.24All the advocates ausgelassen !25Becausethereare no clearexamples of in use 'albeitwithout writingthe preposition' the n in the name with a dative Middle formulas offering introducing god's Kingdom, in his earlystudyin TheTomb shouldnot insertit artificially.26 Gardiner, Egyptologists from 1915, had found no Middle Kingdomevidence for a dative of Amenemhet andhe had to admit(p. 89): 'The strongest thatcan be quoted construction, argument is the above the almost of a before invariable omission against interpretation preposition the divinenamethroughout the MiddleandNew Kingdoms'. so far for a translation of the MiddleKingdom Thereare no convincingarguments versionsof the offeringformula as 'an offeringwhichthe kinghas given (to) Osiris...' This translationcannot have been induced by an alleged existence of a dative constructionfor which there is no evidence at all, but more probably by an (Egyptological) interpretation along the principledo ut des of the frequentMiddle of the requestsby dj=fls/sn, 'thathe/she/theymay give' (on Kingdomintroduction
which see below).27

A few formulasof the early TwelfthDynasty,using the preposition hr beforethe if should be translated as of the usual two consisting god's name, simply parts: 'an Osiris...' (htp-djofferingwhichthe kinghas givenand(whichwas given)fromnearby nswthr-wsjrj...).28The prepositionhr is not equivalentto a dative 'to', and it does not withthe god's name,butbelongsto the god's formula connectthe king'sformula (*htp hr GN). dj
24LingAeg 5, 184 n. 34. R. Leprohon, JEA 76, 164. 26 The statement of Barta, GM 96, 9, that 'die Praposition n vor einem Gotternamenden Dativ angibt, vor njswt dagegen das Agens einfiihrt' does not sound very logical. See the reply of Fischer, GM 107, 70. 27 This is quite clear from Gardiner, The Tombof Amenemhet,88-90, 93 (following a suggestion by Maspero), where his basic argument, beside the introduction of dj=flsn, for a Middle Kingdom re-interpretationis the supposed mercenary mentality of the Egyptians, deduced from Pyr., 1649-51 [599]. The pyramid spell is a kind of decree released by Atum (note the sdm.tj.fj-forms), and the underlying idea is not that of a bargain struck between the king and gods, but that of formulas of promise and implicit threat, of do ut des. 28 For example, BM EA 162 (ANOC 2: Simpson, Terrace of the Great God, pl. 6); L. Habachi, Elephantine, IV. The Sanctuary of Heqaib (AV 33; Mainz, 1985), nos. 4, 5, 19; Lapp, Opferformel, 33 51. It should be stressed that the offering formulas using hr are in no way comparableto the above-mentioned formulas at Assiut (contra Barta, GM 96, 8), simply because these Assiut formulas have no nswt-element, while the hr-formulasbegin with nswt. Accordingly, they are 'complete' versions with the king's and the god's formula. For the meaning of hr, see Franke, Heqaib, 37; it does not denote 'to' (which is but a connotation) or 'raumlich: vor', but 'in the hr, 'veneratednearby' a god, prt-hrw hr, 'an invocation-offering sphere, or aura, of. Comparethe formulasjmWhj (from) nearby' a god (Lapp, Opferformel,95 164), zbj hr k,, 'to go to nearby one's spirit' (H. Roeder, 'Themen und Motive in den Pyramidentexten',LingAeg 3 (1993), 88-9), and hswt njt-hr-nswt, 'a favour from-nearby-theking'. To translatethese offering formulas as 'an offering which the king has given to Osiris...' is accordingly not adequate.
25

This content downloaded from 134.58.253.30 on Mon, 14 Oct 2013 09:39:32 AM All use subject to JSTOR Terms and Conditions

2003

MIDDLEKINGDOM FORMULAS OFFERING

45

are employedonly whenhtp dj nswtaloneis used like a noun, Dativeconstructions the addressee andbeneficiary: as the title or incipitof the offeringritual,to introduce to/for (the spirit (performed) 'an-offering-which-the-king-has-given of)...' NN, or NN (jrt htp-djto/for...' ritual of) An-offering-which-the-king-has-given 'reciting(the nswt n...NN).29 In these cases, the Egyptians obviously did not hesitate to write the dativen. versionof the on Thirteenth Thereareseveralexamples Dynastystelaeof an abridged it is used because where withoutmentioning 'basic'king's andgod's formula requests, with additional of lack of spaceinsteadof the morecompleteversions requests: htp dj k3 nj, 'an of...' NN (BM EA 204, 1. 7),30or htp dj nswt wsjrj...htpdj wpjw3wt...n an offeringwhichWepwawet...has offeringwhichthe king has given and Osiris...and versionis htpdj nswtn k3nj, 'an offering givenfor the spiritof' NN.31An even shorter Likethe OldKingdom whichthe kinghasgivenfor the spiritof' NN.32 version,theking An additional and the god(s) are expectedto give theirfavoursto the stela's owner.33 addedshouldnot changethe meaning, then,butwouldexplainthe dj=flsn withrequests ownerwas simplystatethatthe monument's gifts andfavours.Theseexamples expected of the recipient the offerings. to The offeringformulasof the MiddleKingdomdid not use a dativeconstruction of the offerings.34 transform the role of the god(s)intothatof recipient(s) Accordingly, in the offeringformulas of do ut des was not the basicprinciple the principle expressed do ut des was well at work the principle the Old andMiddleKingdom.35 Nevertheless, in Old Kingdomtimes, as is testified by Spell 599 of the pyramidtexts (Pyr., 1649-51):the kingandGebgive offeringsto othergods, who in turnare expectedto is not the concernof the in favourof the king. Butthis kindof 'extortion' do something aboutthe originof the OldandMiddleKingdom Theyare statements offeringformulas. desired. the favours and offerings
nswt wsjrj n k3 nj..., 'an offering which the king has given and Osiris to/for the spirit

Already Borchardt, S'a3hu-Re' II, pi. 19, cited by Gardiner, The Tomb of Amenemhet, 88, and Lapp, Opferformel,33 52, 153-4 262-3; e.g. stelae Leiden 36 and Berlin 1191 (ANOC 40, pl. 58); Cairo CG 20043 in the 'appeal'; Cairo JE 51979 (G. Jequier, Le Mastabat Faraoun (Cairo, 1928), pl. xii). See below on type D. An early variant is Pyr., 583 [357]: 'an offering which Geb has given to Osiris king NN...' A unique variant was cited by Gardiner, The Tomb of Amenemhet, 91: Cairo CG 20725 (ANOC 48, pl. 66) has jrt htp dj nswt pth-zkr n k? nj..., 'Making An-offering-which-the-king-has-givenand Ptah-Sokar for the spirit of...' as a label to the lector-priest's action (first half of the ThirteenthDynasty). 30 Hieroglyphic Texts III, pl. 16, and e.g. BM EA 215, EA 238, EA 240, EA 252; Cairo CG 20089. It is very unlikely that the god here is recipient of the offerings or part of a (not mentioned) request, as interpretedby G. Lapp, 'Eine spezielle Opferformel des Mittleren Reiches', SAK 14 (1987), 182. 31 BM EA 471: Hieroglyphic Texts V, pl. 6. 32 BM EA 209: Hieroglyphic Texts III, pl. 45. 33 A fine Old Kingdom example is offered by Lapp, Opferformel, 34 fig. 8, which reads: 'an offering which the king has given and an offering which Anubis has given for/to them (the seated couples depicted) daily'. 34 However, nobody doubts that the gods were the recipients of the king's offerings. This is clearly shown by some of the examples adduced by Gardiner, The Tomb of Amenemhet, 89, and Barta, Opferformel,268-9. 35 Contraryto Satzinger, LingAeg 5, 184, for whom the Middle Kingdom offering formula is a 'Manifestation des Prinzipes do ut des: Der (Nekropolen-, Tempel)gott wird durch das staatliche Opfer darauf programmiert,die des Einzelnen zu erfiillen'. Gnadenwuiinsche
29

This content downloaded from 134.58.253.30 on Mon, 14 Oct 2013 09:39:32 AM All use subject to JSTOR Terms and Conditions

46

DETLEFFRANKE

JEA 89

A MiddleKingdom Theintroduction re-interpretation? of the requests Thereis no needto assumea 're-interpretation'36 anda changeof meaning betweenOld Kingdomand MiddleKingdomofferingformulas.In both periodsthe king's formula mentionin parallelism andthe god's formula the giversof the offerings.37 (r)dj(w)is a form with the the relative a double and The perfective subject: king god(s). graphic of the element(r)dj(w),'whichhas given', before(or, in the Old Kingdom, repetition device.38 after)the god's nameis only an optional Thereis a new elementto be observedin the offeringformulas of the earlyMiddle of dj=flsn. The only structural with the introduction difference betweenOld Kingdom is and Middle formulas the additional insertion of a Kingdom Kingdom prospective the followingrequestfor an invocationdj=flslsn, 'mayhe/she/they give', introducing In OldKingdom mentioned. offering etc., whichrefersbackto thegod or godspreviously weresometimes introduced formulas, offering requests by a verbalformtoo, butthenthe suffixpronoun to therecipient of the favours,andnotbackto the alwaysrefers'forward' him be to tomb dead [the owner]an invocation-offering' god(s): 'may given (dj.wn=f prt-hrw),'mayhe be buried'(qrs.tj=f), 'mayhe travelon the good ways...' (hpj=f.hr w3wt a changeof structure andreference. nfrwt...),etc. Thisis obviously The earliestexamplefor dj=sn datesto the SixthDynasty,andis foundon the false1. 2: 'an offeringwhich all the gods of the door of Neferseshemre, lower, horizontal West have given: may they give you a <very?> good old age...' (htpdj ntr.wnb.w
jmntt dj= sn n = k jw nfr < wr?> t...).39 This means that the Middle Kingdomusage of

in the OldKingdom conceivable also to Lappand when,according dj=flsn was already of the offerings,and obviouslywithoutany Satzinger,the gods never were recipients If there was no dativeconstruction connectionto a supposeddative construction. to introduce the god(s) as recipient(s) of the offerings,thereis also no need for a 'thatform' to introduce the requests.40dj=flsn, then, is no syntactic dependent 'clause of
36

but a 'self sufficientstatement'.42 purpose'41

Gardiner, Egyptian Grammar3, 171, and compare the similar statements of Lapp, SAK 14, 182: 'grundsatzlicherBedeutungswandel', Lapp, Typologie, 212-13 494: 'einschneidende Bedeutungsanderung',and Satzinger, LingAeg 5, 184: 'v6llig neu strukturiert'. 37That there was no Middle Kingdom re-interpretationwas the opinion of Barta, Opferformel,265, 267-70, 293, too. While Lapp's analysis of the offering formula in the Old Kingdom is certainly superior to Barta's interpretation, he argues for a Middle Kingdom re-interpretation(e.g. Opferformel, 46 n. 3, p. 241). The weakness of some of his arguments is demonstrated by Willems, in Van Soldt (ed.), Pap Uit Lemen Potten, 98-108; see, too, Willems, Coffin of Heqata, 333. 38Willems, in Van Soldt (ed.), Pap Uit Lemen Potten, 104. 39 J. Capart, Une rue de tombeaux a Saqqarah (Brussels, 1907), pl. 11; Gardiner, The Tomb of Amenemhet, 83; Barta, Opferformel,27 on Bitte 5c, 233, 267; Lapp, Opferformel,204 345(3). 40 See P. Vernus, Future at Issue (YES 4; New Haven, 1990), 18-19, n. 77. That dj=f prt-hrw... was considered as a ratherseparate(optional) formulaic element might explain its applicationfollowing a god's name on an early Twelfth Dynasty double offering-tablefrom Saqqara:htp dj nswtjnpw h3 t... (var.: dj=fh3 t...) n k, njjm,hw hr wsjrj hntj r3-st3w(var.: ddw) dj=f prt-hrw nfr r'w nb..., 'an offering which the king has given and Anubis... (var.: may he give) thousand of bread...for the spirit of the venerated nearby Osiris, foremost of Rosetau (var.: Busiris): may he give a good invocation-offering every day...' (A. Abdalla, 'The Cenotaph of the Sekwaskhet Family from Saqqara', JEA 78 (1992), 98 (niche no. 2), fig. 3d (lines A-B and B-C), pl. xx, 4). 41 Gardiner, Egyptian Grammar3, 171. See also D. B. Spanel, 'Palaeographic and Epigraphic Distinctions between Texts of the so-called First IntermediatePeriod and the Early Twelfth Dynasty', in P. Der Manuelian (ed.), Studies in Honor of William Kelly Simpson (Boston, 1996), II, 769 n. 13. 42 Vernus, Future at Issue, 19 n. 77. Jansen-Winkeln, MDAIK46, 144, and id., Text und Sprache in der 3.

This content downloaded from 134.58.253.30 on Mon, 14 Oct 2013 09:39:32 AM All use subject to JSTOR Terms and Conditions

2003

MIDDLEKINGDOM OFFERING FORMULAS

47

If one concedesthatthereis no graphical betweenthe Old Kingdom distinction and MiddleKingdom writingsof the offeringformula,exceptfor the elementdj=flsn, why shouldtherebe a fundamental In the MiddleKingdom,too, the changeof meaning? are thoughtof as being presented by the king to the gods and to his food-offerings to be the giversof the invocation-offerings (fromthe subjects.The gods were supposed of offerings)and immaterial reversion favours.It shouldbe keptin mindthatthe most in the Old and afterthe king'sandthe god's formula commonkindof favourrequested thatthere MiddleKingdomwas the same-the pr(j)t-hrw,'invocation-offering'43-and recitedand the the formulas was no changein the (archaic) offeringritualperformed, camefromthe regularly offeringsrequested priestlyactions.In practice,thesematerial in the the of altars temples. gods offering elementsof the offeringritual(the The Old Kingdomformulasunite fundamental of the requests king's formula,the god's formulaand often a simple enumeration for a of the 'chess-board' fromthe offering-list transferred type), givingthe key-words train of no continuous is There the and action recitation thought by priest. 'performative' a gap. butrather, andtherequests, theking'sandtheco-ordinated between god'sformula other on the the The Middle Kingdomversion of hand, is offering formula, of the kingandthe gods andfixed introduction formulaic Afterthe archaic 'smoothed'. as those who give the offeringsthere follows a virtualsentencefor recital, referring backas well as forward: gods]give...for...' NN. 'Mayhe/she/they[theaforementioned or precisionandnot re-formulation an is of The insertion dj=fls/sn merely explanatory and it did not affect the meaningof the king's a MiddleKingdomre-interpretation,44 andtheirco-ordination. andthe god's formula formula Someexamples of the offering formulasrejectinga Middle Kingdomre-interpretation A translation to gives a satisfyingandlogical sense to the variousexamples,whichis often superior In the exampleformulas and does not requireemendations. translations the traditional as hdn + GN. discussedbelow,htp dj(w)nswtplus (bare)god's nameis abbreviated Let us considerfirst some exampleswith only one god mentioned. n=f m tnj 3bdw...(stelaTurincat. no. 1447)45 htp dj nswtwsjrj...pr.t(w)-hrw
clause: 'Ein Konigsopfer no requestbut a circumstantial 104ff., interprets Zwischenzeit, dj=f/sn as introducing
des GN..., indem er gibt...', 'An offering-which-the-king-has-givenof GN, in that he gives/giving...'

have a variantformulawhich never Two Thirteenth Dynastygraffitinearthe Nubianfortressat Kumma which the kinghas given:to consume n k' 'an t wnm nswt on stelae: offering hnqt htp dj appeared nj..., jhw 3pdw aus dem bread and beer, meat and fowl for the spirit of...' (F. Hintze and W. F. Reineke, Felsinschriften
43

sudanesischen Nubien (Berlin, 1989), no. 486), and ...dj=sn wnm t hnqt...n k, nj..., '...may they [the aforementioned gods] give to consume bread and beer... for the spirit of...' (no. 400). 44 It could be a formal assimilation of the specification of the material offerings to the common 'immaterial' requests using prospective forms, likehpj=f m htp or d,j=f ptlbj3, in use since the Fourth Dynasty. My explanation was anticipated already by F. LI. Griffith, as summarizedby Barta, Opferformel,256. 45 Reign of Nebhepetre Montuhotep, year 46, translated by M. Lichtheim, Ancient Egyptian Autobiographies

and Gottingen,1988), 63-4 no. 25. The originfromTheban (OBO84; Freiburg of the MiddleKingdom Chiefly
tomb 240 (PM I2, 331) is questionable; see H. G. Fischer, 'An Eleventh Dynasty Couple Holding the Sign of Life', Ier (Brussels, 1995), ZAS 100 (1973), 20 n. 8. Compare stela Leiden 7 (inv. no. AP.67; C. Obsomer, Sesostris 534 (doc. 30); Senwosret I, year 9).

This content downloaded from 134.58.253.30 on Mon, 14 Oct 2013 09:39:32 AM All use subject to JSTOR Terms and Conditions

48

DETLEFFRANKE

JEA 89

is: 'anofferingwhichthe kinghas givenandOsiris...:maybe brought My translation him in Abydos in the Thinitenome...' pr.t(w) hrw is forth the voice(-offering)...for and merely specifying the aforementioned verbal tw-passive46 htp-offerings.This formula is clearlyin the OldKingdom butcomparable shortversionswithone tradition, (or two) god(s) and 'bare'prt-hrware ratherfrequentin the late Eleventhand early TwelfthDynasty.47 Thisexample couldbe easilytransformed intoa versionusingan additional dj=f with as a noun,andindeed,theseabridged versionswithonly one god mentioned are prt-hrw very commonduringthe MiddleKingdom:htp dj nswt wsjrj...dj=f prt-hrw..., 'an offeringwhich the king has given and Osiris...: may he (Osiris)give an invocationoffering...' (e.g. on BM EA 204). Lapphad translated exampleslike this one: 'Ein das der er ein The god's name Opfer, Koniggibt: Osiris, m6ge Totenopfer geben...'48 thusbecamethe subjectof the requestin anticipatory a 'cleft' emphasis.To introduce betweenthe king's formulaandthe god's namedoes not seem very logical in view of the Old Kingdomand later examplesjust cited, which are structured virtuallyin the
same mannerbut without dj=flsn.49

when more gods are involved.Thereare five other Thingsget more complicated basictypesof theseofferingformulas whichoccurin the MiddleKingdom. * TypeA: hdn + GN dj + GN and 'bare'prt-hrw
htp dj nswt wsjrj...dj jnpw...prt-hrw n jm3hj NN... (Turin cat. no. 1534)50 invocation-offeringfor the venerated' NN...

'An offeringwhichthe kinghas givenandOsiris....(and)whichAnubis...hasgiven:an

For the dj-signbetweenthe king's formulaand the first mentioned god, Osiris, we can assumea case of haplography, but it is repeated before the secondgod's name, Anubis.It shouldbe stressedthatdj (if written) has to be readbeforethe god's name, backto thehtp-offering. notrepeated,51 butsee belowon type referring htp is frequently
J. J. Clere, 'Le fonctionnement grammatical de l'expression prj hrw en ancien egyptien', Melanges Maspero, 1/4 (MIFAO 67; Cairo, 1961), 778ff.; Barta, Opferformel, 298-300 on Bitte 2; Lapp, Opferformel, 97-102 168-79; Willems, Coffin of Heqata, 46. 47 E.g. BM EA 152 (Hieroglyphic Texts II, pl. 34; late Eleventh Dynasty?); Cairo CG 20263 and 20751 (ANOC 30; Senwosret I, year 10); BM EA 572 (Lichtheim, Autobiographies, 106 no. 45; ANOC 5; Senwosret I, year 39). The first two examples with two gods' names can be subsumedunder a type hdn + GN GN(s) and 'bare' prt-hrw, which is not specially quoted below. 48 SAK 14, 182. A similar translation was put forward already by A. Erman, Die Hieroglyphen (Berlin and Leipzig, 1912), 69 (see Barta, Opferformel,258), and rejected by Gardiner, The Tomb of Amenemhet, 88-9. 49 See Willems, in Van Soldt (ed.), Pap Uit Lemen Potten, 100-5. Lichtheim, Autobiographies, 122 no. 55, translatedthe example on BM EA 559 in the 'traditional'way after Gardiner: 'an offering-that-the-king-gives(to) Osiris..., that he may give a voice-offering...' (ANOC 43, pl. 62; Senwosret III, year 7), and Satzinger, LingAeg 5, 184, translated examples like these: 'Das Opfer, das der Konig Osiris gegeben hat, (dies)er moge veranlassen...'. See below on type C. 50 G. Maspero, 'Rapportsur une mission en Italie', RecTrav 3 (1882), 115-16 (IV); the stela dates to the early Twelfth Dynasty. Compare Newberry, Beni Hasan II, pi. vii; Cairo CG 20518, 20756; the stela of Renu at Chatsworth (ex Salt Collection sales catalogue no. 576; G. B. Deakin, 'Two Egyptian Stelae in the Devonshire Collection', The Transactions of the Hunter Archaeological Society Sheffield 10 (1971), 65-6; C. Eder, in D. Boschung et al. (eds), Die antiken Skulpturen in Chatsworth sowie in Dunham Massey und Withington Hall (MonumentaArtis Romanae 26; Mainz, 1997), 131 no. 167, pl. 115, 3). 51 Satzinger, LingAeg 5, 182: 'gapping'.
46

This content downloaded from 134.58.253.30 on Mon, 14 Oct 2013 09:39:32 AM All use subject to JSTOR Terms and Conditions

2003

MIDDLEKINGDOM OFFERING FORMULAS

49

B. This is the oldest type, in use since the Old Kingdom,when the dj was usually only afterthe god's name.52 repeated nswt htp dj wsjrj...dj jnpw...prt-hrw h3 t hnqt... (Turin cat. no. 1513)53 whichAnubis...hasgiven:an 'An offeringwhichthe kinghas given andOsiris...(and) of breadandbeer,...' of) thousand (consisting invocation-offering translated instead: 'anoffering-that-the-king-gives Lichtheim (to)Osiris...MayAnubis The of 1000 beer...' before Anubis is and bread give...a voice-offering dj repeated to the prospective andshe considered dj=flsn of otherformulas, by her as an equivalent the two it accordingly as a sentence the requests, thusdissociating translated introducing is grammatically to other gods. Of course,this interpretation possible,butin comparison andeven rather not very convincing, improbable.54 examples,it is semantically * TypeB: hdn + GN htp dj + GN and 'bare'prt-hrw djhqt hnf hnmwhtp dj ntr.wnb.w ?bdw htp dj nswtwsjrj....htp prt-hrw...nk' nj...NN EA Texts 573: Senwosret II, pl. 6; II, year 6) (BM Hieroglyphic 'An offering which the king has given and Osiris..., an offering which Heqet and Khnum havegiven, an offeringwhichall the godsof Abydoshavegiven:an invocationoffering...forthe spiritof...' NN. 'an offering-that-the-king-gives translated: Lichtheim (to) Osiris...; an-offering-thatan a voicewith Khnum; offering-that-all-gods-of-Abydos-give: Heket-givestogether andall the gods of Abydosare offering...forthe ka of...' NN.55WhileHeqet,Khnum consideredas givers of offeringshere, shouldonly the first mentioned,Osiris, be a Thisdoes not seemvery logical.Butdoubtless Lichtheim is right of offerings? recipient in her translation the the of the repeated enumeration of divine names. htp dj before of Turincat. no. 1513, she did not translate to her translation the dj as a Contrary with the htp repeated.The prospective impossible sdm=f, becauseit is grammatically of this kind of formulathus should be the same as that for the Old interpretation wherehtp dj is oftenrepeated afterthe god's name.56 examples, Kingdom
.n... .htp dj wpjw,wt...dj ntr.wjmj.w ?bdwh m t hnqt. htp dj nswtjnpw...htp dj wsjrj.. I, year 17)57 NN (LouvreC166; Senwosret

'An offeringwhich the king has given and Anubis..., an offeringwhich Osiris...has given, an offeringwhichWepwawet...has given, and whichthe gods of Abydoshave of beer...for...' thousand NN. bread, given:
52 53

See Lapp, Opferformel,25 41, 27 44. Lichtheim, Autobiographies, 61 no. 23. The same version is attested on the contemporary stela Pittsburgh Z9-497 (D. Craig Patch, Reflections of Greatness. Ancient Egypt at the Carnegie Museum of Natural History (Pittsburgh, 1990), 22-3 no. 14) of the Eleventh Dynasty. See, too, Louvre C3, introducing the pair Heqet and Khnum with dj, and 'all the gods of Abydos' without the dj repeated (ANOC 3, pl. 15; Obsomer, Sesostris ler, 556 (doc. 36); Senwosret I, year 9), and Cairo CG 20024. 54 These examples are analyzed by Willems, in Van Soldt (ed.), Pap Uit Lemen Potten, 102-6. 55 Lichtheim, Autobiographies, 94 no. 40. 56 See Barta, Opferformel, 55 n. 1; Lapp, Opferformel, 28 45, 37; Willems, in Van Soldt (ed.), Pap Uit Lemen Potten, 99. An offering-table of Senwosret I follows the Old Kingdom spelling with honorific transposition h, m t..., 'an offering which the king has given (and) an offering which of the god's name: htp dj nswt htp dj gb(b) of has thousand Geb bread,...' (A. El-Khouly, 'An Offering-table of Sesostris I from el-Lisht', JEA 64 given: (1978), 44,pl. ix). On stela Cairo CG 20544 from the late Twelfth Dynasty, htp dj is repeated even before the first god's name: htp dj nswt htp dj wsjrj dj=f prt-hrw... 57 Obsomer, Sesostris Ier, 560 (doc. 38); W. K. Simpson, Papyrus Reisner IV (Boston, 1986), pl. 31.

This content downloaded from 134.58.253.30 on Mon, 14 Oct 2013 09:39:32 AM All use subject to JSTOR Terms and Conditions

50

DETLEFFRANKE

JEA 89

to the previousexample,a htp before the summary of gods and the key Contrary wordprt-hrware missing.Of course,it is grammatically here: 'an possibleto translate which the has and an which Osiris...has Anubis..., offering given king offering given andan offeringwhichWepwawet...has given:maythe gods of Abydosgive a thousand of breadandbeer...for...' NN. Butthe senseintended is the sameas in the formulas of the htprepeated Z9-497; (stelaeTurincat. nos. 1534, 1513;Pittsburgh typeA, without or with the dj omitted,as on LouvreC3), or on BM EA 573, wherehtp dj is repeated beforethe summary of gods. For BM EA 573 as well as LouvreC166, htpdj couldnot havebeen inserted afterthe god's name,but it is writtenbeforeit. withdj=sn inserted. Let us now considerthe variants * TypeC: hdn + GN GN(s) anddj=flsn prt-hrw
htp dj nswt wpjw3wt ntr.w nb.w t-dsr dj=sn prt-hrw...n...NN (stela 720/12 at Chatsworth)58

'An offeringwhich the king has given and Wepwawet and all the gods of the sacred land:may they give an invocation-offering...for...' NN. The prospective mentioned dj=sn refersbackto the previously gods. If the gods are of the offerings,but those who give the offerings,thereis no need for a not recipients causal'that-form'. * TypeD: hdn + GN dj + GN anddj=f/sn prt-hrw

+ A A J ...A X ... A'l1

...

hrw...n...NN (CairoCG 20542: ANOC4, pl. 11; Senwosret I, year 24) If one interprets as an equivalent of the common dj=flsn, djjnpwin the firstexample afterthe god's nameAnubisseems superfluous. It makesno sense to its very addition of type It shouldbe translated makethe god partof the request.60 just like the formulas A without dj=flsn: 'an offering which the king has given and Osiris..., which NN. Anubis...hasgiven:may they give an invocation-offering...for...'
58 After a first example for dj=sn from the Sixth Dynasty on the false-door of Neferseshemre (Capart, Rue de tombeaux, pl. 11), this is the second example, on the stela of Nekhty (H. W. Muiller, 'Die Totendenksteine des Mittleren Reiches', MDAIK4 (1933), 187 fig. 11; Deakin, The Transactionsof the HunterArchaeological Society Sheffield 10, 63-5; Eder, in Boschung et al. (eds), Die antiken Skulpturenin Chatsworth, 128-30 no. 166, pl. 115, 1; Lichtheim, Autobiographies, 67-8 no. 26). Nekhty's stela, Turin cat. no. 1513, PittsburghZ9-497, Louvre C15 and Berlin 1197 are probably all from the first half of the reign of Nebhepetre Montuhotep (E. Brovarski, The Inscribed Material of the First IntermediatePeriod from Naga-ed-Der (Ann Arbor, 2001), 1043ff.; C. Obsomer, 'dj.f prt-hrw et la filiation ms(t).n/jr(t).n comme criteres de datation dans les textes du Moyen Empire', in C. Cannuyerand J.-M. Kruchten(eds), Individu, societe et spiritualite dans I'Egyptepharaonique et copte. Melanges egyptologiques offerts au A. Theodorides (Brussels, 1993), 178-9, 197-8). dj=flsn appears more frequently on stelae only after the end of the first decade of the reign of Senwosret I; see Cairo CG 20515 (ANOC 30, pl. 46) and 20516 (Obsomer, Sesostris ler, 514 (doc. 24), 516 (doc. 25); Senwosret I, year 10). 59Simpson, Papyrus Reisner IV, pl. 30; Franke, Personendaten, Dossier no. 152 (last decades of Senwosret I). 60 This example is of the same type as the offering formulas on BM EA 205; Blackman, The Rock Tombs of Meir III, pl. ix, or on coffin M1C (Cairo JE 42949), discussed by Lapp, Opferformel,28 46; id., SAK 14, 181; Willems, in Van Soldt (ed.), Pap Uit Lemen Potten, 101, 105.

htp dj nswt wsjrj...dj jnpw...dj=sn prt-hrw...n...NN (Vienna AS 90)59 htp dj nswt wsjrj...dj wpjw,wt...dj jnpw...dj hqt hnr hnmw dj=f mw hnqt prt-

This content downloaded from 134.58.253.30 on Mon, 14 Oct 2013 09:39:32 AM All use subject to JSTOR Terms and Conditions

2003

MIDDLEKINGDOM OFFERING FORMULAS

51

In the second example, '(an offering) which Heqet and Khnumhave given' is continued by the request'mayhe (sic!) give waterandbeer,...' This is an examplethat in favourof a dativeconstruction: of Gardiner 'The best proof weakensthe argument the followingclauseof purposehas...di.f "thathe may give"... with a singularsuffixpronoun,whereasif severalgods are namedwe find...di sn "thatthey may give"; had the king and the god (or gods) been still regarded as collateral givers of the funerary benefits,the verbdi sn withpluralsuffix wouldhave been foundin all cases'.61 Now, on the partof the this statement a high degreeof logic andconsequence acknowledges The rather ancientEgyptian scribesandsculptors. frequent phrasedj=f mw..., 'mayhe give water...' (on whichsee the sectionbelow)is clearlyused on CairoCG 20542 with referenceto a single male god (often Anubis,here Khnum?), althoughfive gods are in the offeringformula(s). Therearetoo manyexamplesfor this incongruity mentioned to be explainedas merely simple mistakesmade by individualscribes/sculptors.62 god or godsonly, butthekingis never dj=flsn alwaysrefersbackto theaforementioned thereare no examplesof *htpdj nswtdj=fprt-hrw, *'an offeringwhichthe included: knownto me.63The gods king has given (that)he may give an invocation-offering...', alone are thoughtof as being able to fulfil the materialand immaterial requestsand like *hdn + GN anddj=sn...64 formulas afterlifewishes, precluding * TypeE: hdn + GN htpdj + GN anddj=flsn prt hrw
of this re-interpretation is the fact that, if one god is named after the phrase htp di nsw,

dj hqt hn' hnmwhtp dj ntr.wjmj.w 3bdwdj=sn htp dj nswt wsjrj...htp dj wpjw;wt.....htp prt-hrw...n...NN (DurhamN. 1932; Senwosret I, year 13) htp dj nswt wsjrj.. .htp dj jnpw...dj=f prt-hrw...n. ...NN (Berlin 1192; Senwosret I,

year14)65

in number with structured The two formulas are identically (exceptfor the variation withoutdj=flsn (typeB). It makesno sense dj=flsn) andresembleclosely the variants the last-mentioned to me to interpret gods as partsof the request,writtenin anticipatory was or to assume that placedafterthe god's name. Only readinghtp htp dj emphasis,
Egyptian Grammar3,171; compare Barta, Opferformel,268. See the following example of stela Berlin 1192 and below after type E in the conclusion. Barta, Opferformel, 268, thought of exceptions and mistakes on the part of the scribes/sculptors. 63 Of course, there are short, abridged formulas like htp dj nswt n NN, 'an offering which the king has given for' NN, as on Louvre C7 from the reign of Amenemhet IV, or R. Engelbach and B. Gunn, Harageh (BSAE 28; London, 1923), pl. lxxiv, 2. 64 Barta, Opferformel,267 n. 1 ('Lange-Schafer, Denksteine I, p. 301'), adduced an early ThirteenthDynasty example for a plural suffix-pronoun relating to a single god: Cairo CG 20286. The stela is very crudely made, but belongs to the output of a workshop where correct writings were used on its other stelae (see Cairo CG 20069, 20117, 20716; Franke, Personendaten, Dossier no. 735). The formula htp dj nswt wsjrj...dj=sn... on Cairo CG 20286 is clearly a mistake. A second example is on stela Louvre C111 from Elephantine (unpublished). 65 For both stelae see ANOC 31 (Simpson, Terrace of the Great God, pls. 48-9); Simpson, Papyrus Reisner IV, pls. 32-3; Obsomer, Sesostris ler, 505, 533 (docs. 18, 29). The same type of formula occurs on stelae Leiden 3 (AP.7; Senwosret I, year 33), Cairo CG 20025 (Amenemhet II, year 20) and Glasgow Hunterian Museum (BSAE 37; London, 1925), pl. xxiv) D1922.13, 11.5-6 (W. M. F. Petrie, Tombsof the Courtiersand Oxyrhynkhos from the early Twelfth Dynasty.
61 62

This content downloaded from 134.58.253.30 on Mon, 14 Oct 2013 09:39:32 AM All use subject to JSTOR Terms and Conditions

52

DETLEFFRANKE

JEA89

of the MiddleKingdom dj beforethe god's name, as it was writtenby the Egyptians of the god's name),does justice to the examples: (i.e. withouthonorifictransposition 'an offeringwhichthe kinghas given andOsiris..., an offeringwhichWepwawet...has havegiven, andan offeringwhichthe gods given, an offeringwhichHeqetandKhnum for...' NN. The of Abydos have given: may they give an invocation-offering... translation is virtually the sameas thatgivenabovefor the versionon BM EA 573 with a 'bare' prt-hrw(type B). Type E is attestedalso in the Thirteenth Dynasty, for example,on stela LouvreE20002 (ex Musee GuimetC12; ANOC 55), datingto the reigns of Neferhotep I and Sobekhotep IV, and on Cairo CG 20318 from
Elkab/Hierakonpolis.66

offers neithercompleteness and full references Barta,Opferformel, Unfortunately, nor reliabledatesfor manyof the MiddleKingdom monuments he uses. No diachronic and phraseology of the offeringformulasand the requests, studyof the palaeography restricted one of the main to chronologically periodsof Egyptian history,exists except A case in point for such a more close-meshed for Lapp's Opferformel. study is the formula'mayhe give water'(dj=f mw... is alwayssingular). It hadits heydayearlyin the TwelfthDynasty67 andexperienced a revivalin the lateTwelfthDynasty,around the of Amenemhet The for is attested several times III.68 on reign request liquids offeringtables,whereit seemsvery appropriate.69
66

andGunn,Harageh,29, pl. lxxiv, 3; ChicagoOIM8308 (W. M. F. PetrieandF. LI. See, too, Engelbach

Griffith, Abydos, II (EEF 24; London, 1903), pl. xxix, upper right); Cairo CG 20544 from the second half of the Twelfth Dynasty; Turin cat. no. 1526 (Maspero, RecTrav 3, 119 [VII]), Vienna AS 91, and htp dj

nswt (type II!) pth-zkr wsjrj htp dj rcw m sht-htp dj=sn t hnqt n k3 nj NN on a late Thirteenth or
Sixteenth/Seventeenth Dynasty stela now at Trento, Museo Storico (W. v. Bissing, 'Eine Stele des

Mittleren Reichs mit religi6sem Text', ZAS 40 (1902-3), 118). An example for htp dj repeated without any request is on the early Thirteenth Dynasty naos BM EA 471 (Hieroglyphic Texts V, pl. 6), cited above.
67 See Barta, Opferformel,on Bitte 25; Lapp, Opferformel,95-6 166; Lapp, Typologie, 216 503; stelae BM EA 1660, EA 241; Cairo CG 20295, 20456, 20542, 20548, 20756, Leiden 6 (AP.64; Amenemhet II); Munich AS

33; LouvreC2, C19; stela Petrie,Tombs of the Courtiers, pl. xxvi, left; Zagreb2 (ANOC71, pl. 40); statueof Tetiemzaf at Saqqara 1907-08 (Cairo,1909), 113, pl. lvii, left side, (CairoJE 40032:J. E. Quibell,Excavations 1. 2 [Senwosret dell'Universita di Pisa in I]); statueof Wadjetat Khelua(E. Bresciani,'L'attivita archeologica
Egitto (1981): Fayum, Gurna, Saqqara', EVO 4 (1981), 8 fig. 4). 68 On stelae Cairo CG 1481 (J. De Morgan, Fouilles a Dahchour [I] Mars-Juin 1894 (Vienna, 1895), 39 fig.

HarerCollectionno. 43 (G. D. Scott III, Temple,Tomb,and Dwelling:Egyptian 80), 20346; San Bernardino, Museum of Art, California StateUniversity; (R. V. Fullerton Antiquities from the HarerFamilyTrustCollection SanBernardino, 1992), 87-8; Sotheby'sNew York,Antiquities (May29th 1987), lot no. 29; c. Amenemhet III); BM EA 200; Berkeley BtlO(H. F. Lutz,Egyptian Tomb Stelesand Offering Stonesof theMuseum of Anthropology and Ethnology (Leipzig, 1927), no. 77); Leicester4 (K. Kitchen,'FourStelaein of the University of California Leicester CityMuseum',Or29 (1960), 92, pl. xxii); Leiden41 (AP.69), 48 (AP.36);on statueCairoCG 456 (c. Senwosret III/Amenemhet III). Also on late TwelfthDynastycoffins, whereit belongsto the requestfor a good
burial on the back (west side), e.g. Cairo CG 28033 (G3: Lacau, Sarcophages I, 87), Cairo CG 28114 (G4) from

Gebelein,Berlin9 and 10 (TlBe and T2Be)fromThebes;see Lapp,Typologie,190 434, 192 440, 170 394, somemoreexamples on 216 503. 69 Offering-tables fromthe earlyTwelfthDynasty:Abdalla,JEA78, 100 fig. 4, pl. xxi, and 104, fig. 6, pl. xxiii from Saqqara; CairoCG 23056 and 23064 (Lisht;Senwosret of Snofru(A. Fakhry,The I); offering-table Monuments at Dahshur,II. TheValleyTemple,PartII: TheFinds(Cairo,1961), 91 no. 17, fig. 427); of Sneferu CairoCG 23006, 23019 (ANOC13, pl. 23; Amenemhet Inscribed Material II); CairoJE 91220 (W. K. Simpson, Excavations at Abydos(PPYEE6; New Havenand Philadelphia, from the Pennsylvania-Yale 1995), 43-4 no. C15, fig. 72; also ownerof stelaLouvreC169);Peet, TheCemeteries II, 121 no. 26, pl. xxiv, 1; Leiden of Abydos

This content downloaded from 134.58.253.30 on Mon, 14 Oct 2013 09:39:32 AM All use subject to JSTOR Terms and Conditions

2003 Conclusion

MIDDLEKINGDOM OFFERING FORMULAS

53

becausethereis no convincing Thegods' namesneverbecamepartof the king'sformula

evidence for a dative constructionat work that would make the god(s) the recipient(s)

of the king's offerings.The gods' namesare also neverpartof the request-alreadyin

1991 that was the conclusion of Harco Willems in his study 'Food for the Dead', in Van Soldt (ed.), Pap Uit Lemen Potten. Accordingly, there was no change of meaning of the offering formula between the Old and Middle Kingdom, with or without the Qitp) dj repeated,with or withoutthe insertionof dj =flslsn. But there is a difference in spelling:

the honorifictransposition of the god's name(written beforehtpdj) is abandoned. The is certainly of dj=flslsn thatinitiates the requests morethana variant.It is introduction an innovation,a variationbecominga norm, but not necessarilythe result of, and of the formula. combined with, a new interpretation
as argued above. Even the numberof the dj-form is no proof for a re-interpretation, Quite often, a singular suffix-pronoun(dj=j) was used, when there was more than one An example is from the ensemble called god mentionedearlier in the god's formulas.70

CairoCG 23210, 'Ptah-Sokar-Osiris, ANOC 7: on offering-table may he give...' was invoked,but on stela Leiden50 (L.XI.2) 'PtahSokarOsiris, may they give...'71 On
stela BM EA 197, pth-zkrwsjrj nb cnht-wj is followed by dj=sn, 'may they give', and we are dealing with two gods, not the synthetictrinityPtah-Sokar-Osiris. The same gods and the same epithet appearon stela Cairo CG 20434, but with dj=f, 'may he give', in In the middle register of stela BM EA 101 from the reign of AmenemhetIII, above the offerings and the offering-table,are two significantshort labels addressingexplicitly Abydos: may he (sic!) give every good thing to/for the spirit of.. .Nebipu', and the one

The Egyptians were ratherindecisiveon this this case as if a single god is meant.72 particular question. Osiris as giver of the offerings. The first says 'Osiris, Wepwawetand the gods of

inv. no. AP.82 (ANOC 9; Franke, Personendaten, Dossier no. 535). Offering-tables from the late Twelfth Dynasty, c. reign of Amenemhet III: Cairo CG 23035 (ANOC 3, pl. 9); Hannover 1926.191 (M. Cramer, 'Agyptische Denkmaler im Kestner-Museum zu Hannover', ZAS 72 (1936), 89, pl. v, 2; New York MMA 22.1.107A + B (R. Hdlzl, Agyptische Opfertafeln und Kultbecken (HAB 45; Hildesheim, 2002), pl. 9, from Lisht); Moscow 33 (I.I.a.5339); W. M. F. Petrie, Antaeopolis, The Tombs of Qau (BSAE 51; London, 1930), pis. v, 4, xvii upper right (Qaw el-Kebir). 70 See the examples in Willems, n Van Soldt (ed.), Pap Uit Lemen Potten, 105-7; coffin MIC; Cairo CG 20157, 20242, 20273, 20454, 20460, 20516, 20545, 20604, 20638, 20655, 20717, 20758; Los Angeles County Museum of Art 50.33.31 (R. 0. Faulkner, The Stela of the Master-SculptorShen', JEA 38 (1952), 3-5, pl. i); miniature stela of Dedu (Simpson, Inscribed Material, 42-3 no. C13, fig. 70; Senwosret I); stela Glasgow D1922.13, 11. 5-6 (Petrie, Tombs of the Courtiers, pl. xxiv) from the early Twelfth Dynasty; offering-table Philadelphia UM E15413 (W. M. F. Petrie, G. Brunton and M. A. Murray, Lahun, II. The Pyramid (BSAE 33; London, 1920), pl. xxxvi, from tomb 906); stela Rio de Janeiro 2 (645 [2435]; Amenemhet IV); Louvre E20002 (ex Mus6e Guimet C12); Leiden F1939/2.48, Cairo CG 20642, and Florence 61 (see Maree, OMRO73, 18-21). M. Maree, OMRO73, 9 n. (e) and n. 15, thought of simple mistakes by careless sculptors, because otherwise the mention of more than one god would be senseless. I think the 'mistake' is systematic, and the favours requested are expected in these cases specifically from the last-mentioned (male) god, the last in the row of the gods who took part in the reversion of offerings. 71 Simpson, Terrace of the Great God, pl. 16. 72 See further on BM EA 903, or BM EA 242; J. Spiegel, Die Gctter von Abydos (GOF IV/I; Wiesbaden, 1973), 20, 60; G. Lapp, 'Die Stelenkapelle des Kmz aus der 13. Dynastie', MDAIK 50 (1994), 235; Franke, Heqaib, 129 n. 386, and the contrary opinion of Barta, Opferformel,267-8.

This content downloaded from 134.58.253.30 on Mon, 14 Oct 2013 09:39:32 AM All use subject to JSTOR Terms and Conditions

54

DETLEFFRANKE

JEA 89

to its left: 'An invocation-offering of Osiris'giving (prt-hrw m dd wsjrj)for the spirit

of the veneratedone, Nebipusenwosret'.73

The introductionof the recipientof the favours The typical Old Kingdomoffering formulaintroducesthe recipientof the wishes n orjm3 n hj, 'for the venerated, or blessedone' (by requested by a simplepreposition the king/agod) NN. Fromthe SixthDynasty,the phrasen 'for the ka-spirit',was k3 , used in tombrelief, accompanying the presentation of offeringsto the tomb'sowneras of the EleventhDynastyas a kindof abbreviated recitalof the giver of the offeringin in while the 'standard' used in the main scenes, comparable offering offeringformulas the is still n introduced These two different inscription, recipient by jm3hj only.75 contextsshouldnot be confused. Thephrase'forthe spiritof' k3 the recipient of the invocation(n nj) NN to introduce and the did not find its in the milieu of the offeringformula offering piousrequests place of the maintext on stelaebeforethe late firstdecadeof the reignof Senwosret I. Only about 37% of the stelae with offering formulasfrom Senwosret'sreign show the of the recipient introduction of the favoursby 'for the spiritof the venerated' NN (e.g. CairoCG20024, 20518, 20516;LouvreCl, C3), andonly one singlestela(Louvre C2; i.e. 4%) has the bare 'for the spiritof' (n kgnj) NN.76 'Old and Middle Kingdom'writing and 'New Kingdom'writing of htp dj nswt The graphicsequenceof the hieroglyphicsigns of the king's formulain horizontal fromthe late TwelfthDynastyto the early Seventeenth changedgradually inscriptions writtenin a sequence nswt + htp + dj ( A Dynasty.Whilemost commonly timesuntilthe end of the Thirteenth (typeI) fromOld Kingdom Dynasty,the majority of the offering formulasin the New Kingdomshow a sequencenswt + dj + htp has shownthatthis was not the resultof a r) _( 4^ A 1 (type II). PascalVernus77
73

a kind of label.74'For your spirit' (n k3=k or n k3njjm3hj)is found frequentlyon stelae

A. M. Blackman, 'The Stela of Nebipusenwosret: British Museum No. 101', JEA 21 (1935), 3, pl. i; R.

74Tombs of Ankhmahorand Neferseshemptah at Saqqara (Capart, Rue de tombeaux, pis. 61, 98, 101; H. G. Fischer, Dendera in the Third Millenium B.C. (Locust Valley, 1968), 193), more examples: Lapp, Opferformel,

Voices Parkinson, from Ancient Egypt(London,1991), 140-1, e andf (no. 54).

208 355. See, too, in the TenthDynasty tombsat Assiut(SiutIII, 1.59 andSiutIV, 11. 68, 83), andon the roughly coffins from Assiut 215 500). contemporary (Lapp,Typologie,
E.g. stelae Moscow 25 (I.l.a.1137), 26 (I.l.a.5603); Cairo CG 20543; Cairo T. 27.4.22.5; BM EA 614, EA 152; Turin cat. no. 1447. The presence of n k3 nj in these 'labels' should not be confused with its presence in the regular offering formula, as H. Selim, 'An Eleventh Dynasty Stela in the Cairo Museum (Cairo Temp. 27.4.22.5)', MDAIK57 (2001), 265 n. (o), obviously did. Contra Selim, there is no room for his restoration *n k3 nj jm3hw... on Cairo T. 27.4.22.5, 1. 7, but only for n jm;hw... 76 A referee pointed out to me a further example, probably from the same reign: Cairo CG 20339. On stela Cairo CG 20026, n kg nj is used in a formulato be recited by the priests who give offerings (Senwosret I, year 10). 77 'Sur les graphies de la formule "L'offrande que donne le roi" au Moyen Empire et a la Deuxieme Periode Intermediaire', in S. Quirke (ed.), Middle Kingdom Studies (New Malden, 1991), 141-52. See, too, Barta, Opferformel,261-2; Lapp, Typologie, 194 444.
75

This content downloaded from 134.58.253.30 on Mon, 14 Oct 2013 09:39:32 AM All use subject to JSTOR Terms and Conditions

2003

MIDDLEKINGDOM FORMULAS OFFERING

55

of scribalcustomco-existed of the offeringformula.Two traditions changein meaning monuments in Egypt,one for the hieroglyphic (on stone),andone for dailyaffairsand on papyrus).Therewere constant administration (in the hieraticscript,predominantly from the more living from one sphereinto the other,but predominantly interferences sacraltradition. morerepetitive Slowly, some features stagnant, profaneinto the rather of the profane tradition,regularly suppressedby the decorum of court culture, influencedthe standardsof the sacral hieroglyphicscript, at first only in rock like stelae.Sometimes monuments thenon non-royal typesI andII areeven inscriptions, and foundside by side on one stela (e.g. CairoCG 20043, 20160 20313).78 becamea fashion of scribes,or a resultof theirinadvertence, A specialpredilection or Sixteenth in the late Thirteenth of growingpopularity Dynasty,whileotherstandards of court style in art and calligraphywere loosened as well, and a new standard eventually gained dominance. Obviously, it was only a variation, not a new betweenthe 'early' and the 'later' of the formula,a shift of preference interpretation of writingthe archaicformula. variants Thechangebetween relevance. of somehistorical Thisis a conclusion typeI andtype two if sometime II in horizontal was ever, (perhaps generations) inscriptions completed, IV (c. 1730 BC),and probably afterthe reign of King Khaneferre shortly Sobekhotep VII (c. 1665 BC).This dateroughlycoincides afterthe reignof Merkawre Sobekhotep of Northern withthe conquest Dynastyandthebeginning Egyptby the HyksosFifteenth 1650-1570 or FirstTheban BC),andit wouldfit of KimRyholt'sSixteenth (c. Dynasty and of cultural centres to the access had lost which of this dynasty, well for monuments in the Memphite artistic region.79 production of with an offeringformula to dateall monuments Of course,it wouldbe dangerous date to or or Seventeenth Sixteenth the to II and Dynasty straightforwardly type simply features. otherand additional all examplesof type I beforethem, withoutconsidering co-existed both where fromthe late Thirteenth Thereareclearexamples types Dynasty, stelae.80 side by side or on roughlycontemporary Generallyspeaking,palaeographic coincidewithdynastic change.Thismeansthatmost, but, changesmustnot necessarily of course, not necessarilyall, of the stelae with the 'late' offeringformulaof type II and/orto the Sixteenth shoulddateto the finalyearsof the Thirteenth Dynastyor later. Culturalchange, dating criteria and the 'methodof Bennett-Satzinger' Finally, a generalnote on culturalchangeand datingstelae seems to be appropriate. andchangein to development are subject Evenscribaltraditions andformulaic patterns tastesand a livingculture,andwereusedby human predilections, beingswithindividual
78 The 'late' spelling is clearly graphically more attractive when written vertically, and merely graphical reasons might have further encouraged a change of spelling. 79 After Ryholt, The Political Situation, 159-60, 302-9. It is apparent that I try to avoid the label 'Second IntermediatePeriod' because of its relative vagueness. It does not coincide with Upper Egyptian dynastic divisions and is certainly not apt for most of the Thirteenth Dynasty; see the remarks of Vernus, in Quirke (ed.), Middle Kingdom Studies, 151-2. 80 See Vernus, in Quirke (ed.), Middle Kingdom Studies, 145-8, e.g. on BM EA 220 and EA 226, but cancel EA 248 from Vernus' examples. This means that a rigid 'Smither's rule' (P. C. Smither, 'The Writing of htp-dinsw in the Middle and New Kingdoms', JEA 25 (1939), 34-7) does not exist, but grosso modo a trend which is now chronologically newly defined.

This content downloaded from 134.58.253.30 on Mon, 14 Oct 2013 09:39:32 AM All use subject to JSTOR Terms and Conditions

56

DETLEF FRANKE

JEA 89

that and innovation. abilities,capableof variation Accordingly,it seems only natural in different features severaltrendsanddifferent spheresof culture degreesof changing and for differentclasses of objectsare to be observed.For example,tombdecoration in the seemedto have been a ratherprogressive,more inventivefield for innovations like n as faras certainformulas, thancoffinor stelaedecoration earlyMiddleKingdom like the male doublekilt, are concerned.Severalformulaic k3nj, or types of garment, andpictorial motifs,like the sceneof the sem-priest actingor of individuals expressions on stelae.These before in met tomb decoration are a long theyappear holding flywhisk, into stelae of elite tombs featuresfound their way from the decoration programme of decorated elite tombsbeganto cease in the Onlywhenthe long tradition production. on stelae,andseveralnew motifs effortsconcentrated lateTwelfthDynastywereartistic andformulas theirproducts. Certain features were introduced by the artiststo decorate of ritual, such as the basic were ratherresistantto changesdue to the requirements for Others moreeasily, influenced, of the offeringformula. elements mayhavechanged scribeswho wrote hieraticscriptof the daily the example,by the practiceof business and speed,and sometimes differentiated, affairs,with different regionally consequence or subterranean for differenttypes of objects and texts, destinedfor above-ground like tombs,coffins, canopicboxes, offering-tables, stelae,statues,mirrors, disposition, shabtis, sealings, etc.81Featureshavinglong since becomethe norm on a furniture, class of objectsmight gradually,from exceptionand variant,become the norm on
another class of objects. It is importantfor any diachronic study of typology, style or palaeographyand phraseology to be aware of the fact that things could have evolved differently on different types of objects in different regions of Egypt. Everything is

subjectto change,but thereis no automatism. H. G. Fischeronce wrote: 'Since a so-calledfirst occurrence is, in any case, only later than the point at which it was the first we know of, and may be considerably A is of some importance'.82 the relativeincidenceof the criterion actuallyintroduced, in the lunetteof roundof the two wedjat-eyes good examplemightbe the appearance topped stelae. The earliest more exactly datableexamples are from the reign of from evenearlier,probably III.83 Senwosret However,stelaCairoCG20606 is certainly for datingstelaeto II. Thus,a pretended securecriterion aboutthe reignof Amenemhet the late TwelfthDynastyis weakened.We have to reckonwith the possibilitythat existedearlierthantheirfirstdatedoccurrence, andthat'early' seemingly'late'features or revival,but for datingpurposes, typesexistedalso lateras survivals,as an archaism of a late featureshouldnormally the occurrence outweighthe earlierones.84 for one-sidedcriteriaare dangerous This is meantas a warning.One-dimensional, for datingpurposes.The seeminglyobjective,because statistical,system or method
81 See also on the topic of a different speed of change for graphic patterns between stelae and sealings: W. Grajetzki, 'Der SchatzmeisterAmenhotep und eine weitere Datierungshilfe fuirDenkmaler des Mittleren Reiches', BSEG 19 (1995), 10-11, and for the regional differences in contemporarycoffin production the study of Lapp, Typologie. 82 In R. Caminos and H. G. Fischer, Ancient Egyptian Epigraphy and Palaeography (New York, 1976), 47. 83 Louvre E20350 (ex Musee Guimet C6, dated by Durham N. 1936) and Cairo CG 20296 (dated by Rio de Janeiro 1 (627 [2419])). The two eyes appearalso in the upper part of stela Cairo CG 20686 of the naos-front type, dated to the reign of Senwosret III, year 15. 84 Compare Fischer, in Caminos and Fischer, Ancient Egyptian Epigraphy, 39; id., 'Some Early Monuments from Busiris in the Egyptian Delta', in Ancient Egypt in the Metropolitan Museum Journal, Volumes 1-11 [1968-1976] (New York, 1977), 161 (reprint from MMJ 11 (1976), 5-24).

This content downloaded from 134.58.253.30 on Mon, 14 Oct 2013 09:39:32 AM All use subject to JSTOR Terms and Conditions

2003

MIDDLEKINGDOM OFFERING FORMULAS

57

stelaedeveloped andrefinedlaterby Helmut datingMiddleKingdom by JohnBennett85 suffers from this drawback. It works of a single Satzinger86 clearly only with variables element, the offering formula,of a whole set of variable thoughnearly ubiquitous elementswhichconstitute a stela, andits reliability andresultsare extremely dependent on the database at hand.Bennett's database was a conglomerate of datedstelae, several a few coffins and a single offering-table, but contained no tombs, rock inscriptions, statuesand none of the many other objectsinscribedwith the offeringformula.For datingstelae, only stelae shouldbe assembledin the database.But even a 'cleared', improved and more up-to-date, enhanced database would not eliminate the of the system.87 Theseare three,in the main. (1) The divisioninto three disadvantages Twelfth-later Twelfth Dynasty) is ratherwide-meshedand periods (Eleventh-early differences. andTwelfthDynastiesare neglectspossibleregional (2) Onlythe Eleventh covered, so any applicationof the method to the hundredsof stelae which were producedafter the TwelfthDynastywould lead to no or wrong results. (3) Wrong such as imitation or results,i.e. false dates, could also be due to othercircumstances, archaism or mistakes as a determinative (such copy, simple forgotten) on the errorscan only be detectedby correlation with scribe/sculptor's part. Such systematic othercriteria,features andmotifsat hand.Theostensible of themethod leads objectivity to objectively wrong results because the subjective element, the conscious and Thereis no deliverance deliberately actinghuman being, was left out of consideration. for the statistical method.Satzingerrightlyhas labelledit as a (ratherinconvenient) and it shouldnot be appliedwith confidenceany longerby anyonestudying 'crib',88 MiddleKingdom stelae. Formulas andphrasesdo not necessarily mirror changesin otherspheresof society, notto speakof historyor sociology.Onlyfromthe viewpoint of culture andits artefacts as complexentitiesmay changesbe observedandinterpreted adequately by us at all.

85 'Growth of the htp-di-nsw Formula in the Middle Kingdom', JEA 27 (1941), 77-82; id., 'Motifs and Phrases on Funerary Stelae of the Later Middle Kingdom', JEA 44 (1958), 120-1. 86 See Satzinger, LingAeg 5, 184-8. 87 I have tried to work along the method of Bennett and Satzinger with new percentages from a database consisting exclusively of stelae, about 155 from the Twelfth Dynasty alone, 104 of them with offering formulas. Bennett's database were 121 offering formulas: 14 from the Eleventh, 102 from the Twelfth and only 5 from the ThirteenthDynasty. 88 LingAeg 5, 187, and he adduces stela Vienna AS 164 as an example for a wrong dating that resulted from the application of the method. Bennett's criteria are certainly indicative of formulaic and graphical changes of the offering formula. However, it is no exact method for dating but 'might help to date' stelae, as put forward by Bennett (JEA 27, 77). See also the statement by Spanel, in Der Manuelian (ed.), Studies in Honor of W. K. Simpson II, 767 n. 9.

This content downloaded from 134.58.253.30 on Mon, 14 Oct 2013 09:39:32 AM All use subject to JSTOR Terms and Conditions