STEWART, RANDALL, The Oracular "ei" [Greek] , Greek, Roman and Byzantine Studies, 26:1 (1985:Spring) p.67 The Oracular Ei Randall Stewart I N PLUTARCH'S De E apud Delphos, the priest Nicander offers this explanation of the E inscribed on the Delphic temple: fern ,),ap, W() lJ'TToAaf.L{3&IlOVCTt aEAcf>ol ... erxr,~ Kat f..Wpcf>lj Ti/() \ \ () \ "1:. "I:. e ",..." 7TPO() TOil EOV EVTEV",EW(), KaL Ta",LV TJ,),Ef..WVLKTJV EV TOL() EPW- Tr,~CTtV EXH TWV x.pWf.LEIlWIl EKaer'TO'TE Kat Bt(mvll()avo~vwv El IlLKT,erOVCTtV, El ')'af.LT,erovCTtv, Ei crvJUPEPH 7TAEtV, El ')'EWP')'ELIl, Ei a7Tol)TJf.LELV (Mor. 386B -c). For it is, as the Delphians assume ... the figure and form of the consultation of the god, and it holds the first place in every ques­ tion of those who consult the oracle and inquire IF they shall be victorious, IF they shall marry, IF it is to their advantage to sail the sea, IF to take to farming, IF to go abroad. l It is surprising that Nicander would have seen an explanation for the unknown and mysterious E in something so commonplace as the relational particle which, in normal usage, subordinates a question to a verb of inquiry-a particle used not only at Delphi but in everyday speaking and writing as well; his interlocutors dismiss the suggestion immediately. But it is a significant indication of the verbal structure of at least one kind of petition made to the oracle, in which ei was indeed the first word of independent questions. Examples of this kind of petition are uncommon because the emphasis everywhere is on the responses rather than the questions; moreover, when our sources include the question, they almost invariably cast it in indirect form, thereby tending to obscure the original wording. An Eleusinian inscription of 352 B.C. offers an interesting exception (IG IJ2 204 [Syll.3 204]). This lengthy text explains how the Athe­ nians are to ask at Delphi whether they should leave uncultivated a particular plot of land at Eleusis. The portion relevant to our discus­ sion reads as follows (23-30): 1 Tr. F. C. Babbitt, Loeb Moralia V (London/Cambridge [Mass.] 1962) 207-09. 67 STEWART, RANDALL, The Oracular "ei" [Greek] , Greek, Roman and Byzantine Studies, 26:1 (1985:Spring) p.67 68 THE ORACULAR EI [v E] vEtp'Yau~[E] va [rY1~ u.pa~ op'Yaoo~ Til: Evho~ TWV opwv Ei~ ol.­ [K]OOO~V TOV 1TpO[UTWWV Kai E1TI.CTKEvT,V TOW u.POV TO'iV 8EO- " ,~, 'ri []' [ , ~_'":. ''']'' (.II' E(8 oE TOV ETEpOV K a TTtT EpOV' Et IWIWV Kat a~t vov EUTt TWt &r,J.l,.Wt TWt 'A8."vaLwv Ta v[Vv EVTO~ TW] v o[pWV E] VEtp'Y[a]u~v­ a rij~ iEpa~ op'Yaoo~ Eav aVETa (TotV 8]EotV. The inscription then explains that after the two pieces of tin have been wrapped in wool so as to render them indistinguishable from each other, one is to be placed in a silver jar, the other in a gold jar. These jars are to be sealed and left in Athens while a specially se­ lected embassy journeys to Delphi to ask the Pythia which inscription should dictate the Athenians' course of action, that in the silver or that in the gold jar. H. W. Parke correctly points out that "the very fact that the whole procedure was described in such detail in a public decree, instead of merely stating the questions to be asked, certainly implies that it was a highly exceptional method. "2 But in view of Nicander's inter­ pretation of the symbolic E, and the fact that the phrase AWWV Kat al-'EtvOv is so common in oracular petitions as to be formulaic,3 it seems that the inscriptions on the tin plates preserve the verbal structure of one of the standard forms of oracular inquiry, however exceptional the use of sealed jars and alternative questions may have been. Apparently this standard form was simply subsumed into the system the Athenians had devised to obviate the possibility of fraud.4 Were this not the case-were, that is, the entire system a nov­ elty-one would expect statements rather than questions on the tin plates, since one of them was to serve, in effect, as the answer of the Pythia. Many of the small lead tablets excavated at Dodona preserve orac­ ular petitions in their original form, with the structure El + interroga­ tive: , , f , ae. TVxae.xa /-We. a E1Tc.- , , " TP011'Ec.a Tav EXW ya,.u;JV AVKKiOO~;5 2 H. W. Parke, The Oracles of Zeus (Cambridge 1967) 104. 3 In addition to Joseph Fontenrose's discussion (with references) of the phrase in The Delphic Oracle (Berkeley 1978) 22lf, see Parke (supra n.2) 261f no. 5, 263 no. 1, 268 nos. 15-16, 269 nos. 18 and 21,270 no. 22, 271 no. 25. 4 Parke (supra n.2) 104 suggests that strong feelings on the question in Athens and/or uncertainty about the reliability of the Phocian occupiers of the sanctuary may have given rise to the precautions. Alfred Korte expressed virtually the same hypothe­ sis in "Zum Orakel tiber die iEpa op'Ya~," Klio 5 (1905-06) 280-82. 5 Parke (supra n.2) 266 no. 10. D. Evangelides, 'H7TEc.pwnKa )(polltKa 10 (935) 252 no. 36, dates the tablet to the fifth or fourth century B. C. STEWART, RANDALL, The Oracular "ei" [Greek] , Greek, Roman and Byzantine Studies, 26:1 (1985:Spring) p.67 RANDALL STEWART 69 roxa ayalla. 7, 7'lJ')'xaVOL~ Ka Ef.L71'0PEVOf.J,EVOr; 071'Vr; Ka BOKTlL croJJ4>opov Ef.J,ELV, Kat aywv, TTlL Ka BOKTlL • I I I 6 aJ.LaTaL TEXvaL x.pEOf.J,Evor;; This construction seems to be an abbreviated form of the structure exhibited on other tablets where the inquiry is introduced by the name of the petitioner and a verb of asking in the third person: (J ' I , ...... ''A I .... EOr;. roX7l. E1TLKOLV7ITaL AKLvoor; TWL A' .... N ' ' .... A' .['] U,LL TWL aLWL KaL rat u,LWVat Et A wwv Kat] af.LEtvOV NtKEal. KaraO'KEOOdEtV] , • , 7 TO EpyaUT'T1pwv. EPWrT1L A vua­ vtar; 4ta Nnwv ''''' " KaL U,71wvav 71 ov- " '1:'''''' K Eun E~ aVTOV , s:,..!. TO 1TaLuupw v <> 'AVVVAa KVEL.8 • .... K\ I 'A ' " EpOVTaL I\.EOVTaL TOV u,ta KaL Tall A' ........ f3a I u,LWvav, at EUTL aVTOL 71'pO TEvovn .. , • A,. '\ • 9 ovawv Kat WEPEL [scil. r, OEm,,], 'TT'A1]V on , t , " " " " 'TT'PO(TW'TT'WV WpL(TJ-LEVWV KaL TO'TT'OV KaL )(pOVOV KaL TPO'TT'OV KaL aL- Tia" e(TTiv a'TT'apEJ.L¢aTov, OLOV OE(TL" J-LEv €i 'TT'PO(TT,KEL 'TT'OAWP­ KOVJ-LE VOL" (TTPCXrWj.LU 'TT'EJ-L'TT'ELV El" T1] V V'TT'EpOpia V, tmoOE(TL" BE El 'AOT/VaioL" 'TT'OAWPKOVJ-LEVOL" tmo n€AO'TT'OVV71(TLwV €l" LLK€Aiav (TTpaTWj.LU 'TT'EJ-L'TT'ELV (Spengel, Rhet. II 61.6ft). According to this scheme, the questions given as examples in Mor. 408c would be theses, since they are not person-relative, while most of the questions in the Sortes, as well as the oracular petitions cited above, would be hypotheses, because they deal with the case of a particular person, namely the individual consulting the book or the oracle. In the rhetorical treatises, theses and hypotheses are often depen­ dent on a verb such as ~"f1'TEW, e.g. EG"'TW 8' o~v r,JUi~ ~"f1TEtV, ei 1rPO- gued that the form originated as a translation of the Hebrew interrogatives ha- and 'irn in books translated from or dependent on Hebrew or Aramaic texts (e.g. Gen. 17.17, Joel 1.2, Matt. 12.10) and then, having become familiar, was employed in the 'free' biblical Greek of such works as 2 Mace. (7.7), Clem. Hom. (15.9, 16.16, 17.19), Ev. Thorn. (B8.3 =Tischendorf p.153), T.Abr. (recension A 8.3, 15.13, 18.3), and the sections of Acts that are independent of Semitic sources 09.2, 21.37, 22.25). Though not completely beyond criticism, this theory seems to be the easiest explanation of the facts, in view of the absence of the form in secular Greek apart from the constructions discussed in this article. One must therefore view as products of a separate tradition the oracular petitions in the Septuagint that would otherwise seem to be syntactical parallels to the petitions from Greek oracles (e.g. 1 Ki. 30.8 Kai ET1"T/PWTrWEV aavLD DW TOV Kvpiov Af:YWV Ei KamDw;/;w o'Tfiu-w TOV )'EDDOVP TOVTOV; Ei KamA,""f.L"'O/_UXL aVTovs-;). STEWART, RANDALL, The Oracular "ei" [Greek] , Greek, Roman and Byzantine Studies, 26:1 (1985:Spring) p.67 72 THE ORACULAR EI VOOU(TI. (JEOI. TOU KOU#-WV (Theon, in Spengel, Rhel. II 1126.2ff); but they are also commonly encountered in constructions where no such verb is present, as the following examples illustrate: ~ 0;:,' 8' .., \ e ,... <'~, "'" , TWV uE EUEWV at J..LEV a7TAat, at uE KaTa TO 7TpO'O TL Aa/-L{3itvoVTat, c'I ~,~ ,.... •• :r "\,,, " , ," f ,..., at oE ot7TAat vo,.....,..,OVTac.. Eav yap AE'YWJ..LEV EL yaW'lTEOV, a7TA-r( "£:'"'RIV.\''''' I", ,'~" ""8 Eav uE EL ,.....,.UW\.EL yaWT/TEOV, 7TpO'O TL' Eav uE AEYWJ..LEV EL a - ATITEOV ~AAOV T1 YEWPYTITEOV," 8L7TA7, (Hermogenes: Rabe, Rh.Gr. VI 25.16ft). '0;:,' , 0;:, ""..% " ., ,,' , , .. " OVuEV J..LEVTOL uta~PEL, Eav TE OVTW I\.EY{/ TL'O, EL ya/-LTlTEOV Tl OV, Kat 7TCiALV, Ei aipETEO'O 0 YO:/-LO'O T1 ~VKTEO~' EV yap Kat TaVTOV Eun 8w 7TClVTWV TOVTWV TO 8TlAOVJ..LEVOV (Theon: Spengel, Rhet. II 121. 14ft). sed tamen, ne me totum aegritudini dedam, sumpsi mihi quasdam tamquam 8EueL~, quae et 1TOMTLKai sunt et temporum horum, ut et abducam animum a querelis et in eo ipso de quo agitur exercear. eae sunt huius modi: Ei J..LEVETEOV EV rfi 7TaTpi8t TVpaVVOV~vTl~ avTi,~' Ei. 1TavTt TPO'TTCp TVpavvioo~ KaTO:AvuLV 1TpaY~TevTEov, Kav ~AATI 8w TOVTO 1TEpi TWV clAwv r, 1TOAL~ KLV8vVEVUELV. Ei. eVAa{3TlTEOV TOV KaTaAVOVTa /-LT, aVTo~ aiPTlTaL (Cicero: Alt. 9.4). This independent use is also evident in the titles of several rhetorical works of Plutarch: ei 8tOOKTOV ..;, apE'T'l-]; (439A), Ei aVTapK'Yl~ ..;, KaKia 'TT'pO~ KaK08at#-Wviav; (498A), Ei 'TT'PEU{3VTE{JC{J 'TT'OAtTEVTEOV; (783A), Ei KaAW~ ELP'YlTat TO Aa(JE {3tWua~; (1128A).20 There is a striking similarity between the rhetorical form as it appears in these passages and the questions of the Sorles and peti­ tions with Ei made to Greek oracles. As the passage from De Pythiae oraculis demonstrates, Plutarch was aware of this similarity; and it appears that the compiler (or a redactor) of the Sorles identified the questions in the work as hypotheses. In manuscripts ELM, we find as an incipit to the list of questions the heading apxT, TWV ''YlT'Yl,."aTWV (KAr,pWV in all other witnesses), and in the introductory epistle (miss­ ing in M) an explanation of the use of the book begins KEW(JW ''rITEtV nva Ei 7TPOK07TTEt EV nIJ-ii. In the rhetorical treatises, ,r,T'YlJLa often serves as a generic label for theses and hypotheses and, as noted above, one is said to investigate (''YlTEtV) these questions. Further­ more, some petitions in the Sorles- e.g. Ei EiJpW oovEwau(Jat aPTt; (25), and Ei EL7TW TT,V 8iK'YlV; (51)-are clearly of a deliberative na- 20 In addition to these surviving works, the so-called Lamprias catalogue lists the following titles: Ei apErT, T, PT/TOPUOJ; (86), Ei AOYOV EXEL Ttl ~4Ja; (135), Ei 1Tam CTVv­ T/YOPT/TEOV; (156), Ei 8Wa-EL YVW/-LT/v 0 1TOALTTj<; 1TPOELBW<; on ov 1TOVT,UEL, OV 1TELUEL; (64), 1TEpt TOll YVWfJL uaVTOV Kat Ei afJavaTO<; T, t/lvXT,; (177), 1TEpt apETTj<; Ei 8LOOKTOV T, apErT,; (180), Ei a1TpaKTo<; 0 7TEpi 1TavTwv E1TEXWV; (210). STEWART, RANDALL, The Oracular "ei" [Greek] , Greek, Roman and Byzantine Studies, 26:1 (1985:Spring) p.67 RANDALL STEWART 73 ture, suggesting that the author made no distinction between hypoth­ eses and oracular questions. The structural likeness and the witness of Astrampsychus or his redactor notwithstanding, the oracular petitions and most of the ques­ tions in the Sorles are not, in the technical sense, hypotheses: their intent is to obtain from a supernatural force information about what fate ordains in regard to some event or person, while the hypothesis is intended to elicit discussion about what is or what should be the case. Nonetheless, the thesis/hypothesis form is instructive as a syn­ tactical parallel to the questions of the Sorles and the Greek oracles, for it shows that a question with eL can stand independently of a leading verb. In these rhetorical forms, the deep structure is a dependent inter­ rogative, since the eL is generated by ~7JTEW or some verbal equiva­ lent, the force of which is felt even when the verb is omitted. How­ ever, in its surface structure the question is direct, because of the suppression of this governing verb. The evidence presented by these forms suggests that in the questions of the Sortes and in the direct questions with eL at Delphi and Dodona, we also have to do with an elliptical construction resulting from suppression of an element so closely associated with the form that it did not need to be expressed. The fact that some of the petitions on the Dodona tablets are in the form of indirect questions introduced by a verb of asking, with the name of the enquirer as subject, argues that the omitted element is a verb such as epwTw. Thus, in a sense, both the views of Hoogendijk and Bjorck are partially correct: the questions in the Sortes, as well as their parallels from Delphi and Dodona, Qre indirect from a dia­ chronic perspective, because in their deep structure they depend on a verb such as epwTw; but synchronically they are direct, in that this verb has been omitted. THE UNIVERSITY OF ILLINOIS AT URBANA-CHAMPAIGN February, 1985