Hostname: page-component-cd9895bd7-7cvxr Total loading time: 0 Render date: 2024-12-25T08:07:26.748Z Has data issue: false hasContentIssue false
  • English
  • Français

ΚΥΡΙΕ, ΔΕΣΠΟΤΑ, Domine. Greek Politeness in the Roman Empire*

Published online by Cambridge University Press:  23 February 2012

Eleanor Dickey
Eleanor Dickey
Affiliation:
Columbia University
Get access Rights & Permissions [Opens in a new window]

Abstract

Why did the Greeks of the Roman period make such extensive use of the vocative κύριε, when Greeks of earlier periods had been content with only one vocative meaning ‘master’, δέσποτα? This study, based primarily on a comprehensive search of documentary papyri but also making extensive use of literary evidence (particularly that of the Septuagint and New Testament), traces the development of both terms from the classical period to the seventh century AD. It concludes that κύριε was created to provide a translation for Latin domine, and that domine, which has often been considered a translation of κύριε, had a Roman origin. In addition, both κύριε and domine were from their beginnings much less deferential than is traditionally supposed, so that neither term underwent the process of ‘weakening’ which converted English ‘master’ into ‘Mr’. δέσποτα, which was originally far more deferential than the other two terms, did undergo some weakening, but not (until a very late period) as much as is usually supposed. These findings in turn imply that Imperial politeness has been somewhat misunderstood and suggest that the Greeks of the first few centuries AD were much less servile in their language than is traditionally assumed.

Type
Research Article
Information
The Journal of Hellenic Studies , Volume 121 , November 2001 , pp. 1 - 11
Copyright
Copyright © The Society for the Promotion of Hellenic Studies 2001

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

1 E.g. Svennung, J., Anredeformen: Vergleichende Forschungen zur indirekten Anrede in der Dritten Person und zum Nominativ für den Vokativ (Uppsala 1958) 337Google Scholar; Zilliacus, H., Untersuchungen zu den abstrakten Anredeformen und Höflichkeitstiteln im Griechischen (Helsinki 1949) 20Google Scholar; Wendel, T., Die Gesprächsanrede im griechischen Epos und Drama der Blütezeit (Stuttgart 1929) 88.Google Scholar Sometimes this statement is made in the more accurate from that κύριε is absent from classical Greek except for one occurrence in Pindar (P. 2.58).

2 E.g. Hagedorn, D. and Worp, K.A. ‘Von κύριος zu δεσπότης: Eine Bemerkung zur Kaisertitulatur im 3./4. Jhdt.’ Zeitschrift für Papyrologie und Epigraphik 39 (1980) 177Google Scholar; Svennung (n.1) 336-8; Zilliacus (n.1) 34; Foerster, W. and Quell, G., ‘Kύριος’, in Kittel, G. (ed.), Theologisches Wörterbuch zum Neuen Testament (Stuttgart 1938) 2.1044Google Scholar; Baudissin, W.W. and Eissfeldt, O., Kyrios als Gottesname im Judentum und seine Stelle in der Religionsgeschichte (Giessen 1926-1929) 2.296-7Google Scholar.

3 E.g. Bang, M., ‘Über den Gebrauch der Anrede Domine im gemeinen Leben’, appendix to L. Friedlaender, Darstellungen aus der Sittengeschichte Roms (9th-10th edn, ed. Wissowa, G., Leipzig 1921) 4.82-8Google Scholar; Svennung (n.1) 338-46.

4 Braun, F.Terms of Address: Problems of Patterns and Using in Various Languages and Cultures (Berlin 1988) 57.CrossRefGoogle Scholar

5 See Bréhier, L., ‘L'origine des titres impériaux à Byzance’, Byzantinische Zeitschrift 15 (1906) 161–78CrossRefGoogle Scholar; Deissmann, A., Licht vom Osten (4th edn, Tübingen 1923) 298–306Google Scholar; Preisigke, F., Wörterbuch der griechischen Papyrussurkunden (Berlin 1924-1927; see also revised edn by Kiessling, E. 1944) s.v.v.Google Scholar; Baudissin and Eissfeldt (n.2); Amantos, K., ‘Γλωσσικά’ Byzantinische Zeitschrift 38 (1928) 18–20Google Scholar; Dineen, L., Titles of Address in Christian Greek Epistolography to 527 AD (Chicago 1929) 56–7 66-8, 76, 78Google Scholar; Wendel (n.1) 88; Foerster and Quell (n.2) 2.1038-94; Rengstorf, K.H., ‘Δεσπότης’, in Kittel, G. (ed.), Theologisches Wörterbuch zum Neuen Testament (Stuttgart 1938) 2.43-8Google Scholar; Zilliacus (n.l) 20, 34; Zilliacus, H., Zur Sprache griechischer Familienbriefe des III. Jahrhunderts n. Chr. (Helsinki 1943) 31–2Google Scholar; Dolger, F. ‘Die Entwicklung der Byzantinischen Kaisertitulatur und die Datierung von Kaiserdarstellungen in der Byzantinischen Kleinkunst’, in Mylonas, G.E. and Raymond, D. (eds.), Studies Presented to David Moore Robinson (St. Louis 1953) 2.985–1005Google Scholar; Svennung (n.l) 336-8; Bureth, P., Les titulatures impériales dans les papyrus, les ostraca, et les inscriptions d'Égypte (30 a. C. -- 284 ap. C.) (Brussels 1964)Google Scholar; Hagedorn and Worp (n.2) 165-77; Pietersma, A., ‘Kyrios or tetragram: a renewed quest for the original LXX’, in Pietersma, A. and Cox, C. (eds.), De Septuaginta: Studies in Honour of John William Wevers on his Sixty-fifth Birthday (Mississauga, ON 1984) 85–101Google Scholar; Bauer, W. and K. and Aland, B., Griechisch-deutsches Wörterbuch zu den Schriften des Neuen Testaments und der frühchristlichen Literatur (6th edn, Berlin 1988) s.vv.CrossRefGoogle Scholar; Dickey, E., Greek Forms of Address: From Herodotus to Lucian (Oxford 1996) 95–101Google Scholar; Grünbart, M., Die Anrede im byzantinischen Brief von Prokopios von Gaza bis Michael Choniates (diss. Vienna 2000) 173–5, 193-4, 206-7Google Scholar; also further bibliography cited in these works.

6 What follows is only an overview omitting many of the details and most of the evidence; for complete information see Dickey, E., Latin Forms of Address: From Plautus to Apuleius (Oxford 2002), ch.2.Google Scholar

7 Except the use of domina and domine to lovers which has been traced to the second century BC (Lucilius fr. 730 Marx) but seems to be a separate issue.

8 The contexts are Cicero's letters (compare the high frequency of domine in the letters preserved at Vindolanda) and Horace's discussions of flattery (S. 2.5.23-38, Ep. 1.6.54-5, 1.7.37-8; compare Juvenal 5.132-9, 8.161 and Martial 2.68, 4.83, 6.88, etc.)

9 Suet. Aug. 53.1: dominumque se posthac appellari ne a liberis quidem aut nepotibus suis vel serio vel ioco passus est atque eius modi blanditias etiam inter ipsos prohibuit.

10 Cf. Kohm, J., Altlateinische Forschungen (Leipzig 1905) 167.Google Scholar

11 Cf. the widespread use of kinship terms such as pater (‘father’) and frater (‘brother’) in address to men unrelated to the speaker, and the frequent combination of such terms with domine (e.g. in the Vindolanda tablets; n.b. also Greek Anthology 10.44 and Horace and Juvenal cited in n.8 above.)

12 The feminine is first attested in Homer (e.g. Od. 7.347, 14.127); the masculine (which poses metrical difficulties in epic) appears in Sappho (95.8 V), Pindar (P. 4.207), Aeschylus (Eu. 60), Herodotus (3.89), Isocrates (4.121), Aristotle (Pol. 1253b), etc. cf. LSJ s.v.

13 E.g. Hdt, 3.85.2; Eur. Cyc. 250; Ar. Ran. 1, Vesp. 142; Men. Dysc. 589.

14 E.g. Hdt. 1.8.3, 1.90.2; Xen Cyr. 6.6.2; Aesch. Pers. 1049; Eur. Heracl. 785.

15 E.g. Eur. IT 271, Bacch. 582; Ar. Vesp. 389, Nub. 264, Ach. 247.

16 E.g. Chariton 2.1.3, 4.2.9; Luc. Dial. Meret. 2.3; Achilles Tatius 3.20.1, 5.17.3; for New Testament usage, see Rengstorf (n.5) esp. 47-8.

17 Statistics concerning papyrus documents are based on electronic searches of the Duke Database of Documentary Papyri, using the search facilities on the web version of the Perseus program (perseus.csad.ox.ac.uk) in June 2000. This software allows very sophisticated searching, locating even misspelled versions of the words concerned, and the database contains virtually all papyri published before 1996. Whenever possible (i.e. in about 80% of the cases) I checked the data thus obtained against a printed edition to verify dates and contexts. In addition, I have deleted from my statistics all occurrences which are purely supplements, retaining only those of which some trace remains on the papyrus.

18 P.Berl.Frisk 3.3; P.Cair.Isid. 66.3; P.Flor. i.58.14; P.Kell. 19a.3; P.Leit. 7.14, 9.6; P.Oxy. x.1252.2.14, xvii.2131.7, 2133.4, xxxiv.2713.8, xlv.3243.1. fr. 1.5; PSI ix.1076.5; PSI Congr. xxi.13.1.14; P.Tebt. ii.326.3; P.Vind.Bosw. 4.3; SB iii.7205.3; perhaps P.Flor. i.58.3.

19 CPR vii.15.2, xiia.15.8; P.Amh. ii.82.4; P.Cair.Isid. 74.5, 76.9; P.Col. vii.169.3, 170.9, 173.4; P.Kell. 20.17; P.Mert. ii.91.6, 7; P.NYU i.1a.6; P.Oxy. i.71.3; P.Ryl. iv.706.11; PSI vii.769.1; P.Sakaon 38.3, 40.5, 41.3, 44.4; P.Turner 44.4; SB xiv.11929.16; Chr. Mitt. 63.16, 77.11, 78.10.

20 P.Amh. ii. 143.24; P.Abinn. 26.4, 26.26, 31.7; P.Haun. ii.25.7; P.Herm. 7.15, 8.18, 9.7; P.Kell. 5.26, 46.29, 69.16; P.Neph. 10.7, 19.4; P.Ross.Georg. iii.9.23; SB i.2266.15, viii.9683.15, 21, 25, xiv. 11882.3, 4, 10; Stud. Pal. xx.111.5, 4th or 5th c.; perhaps P.Ross.Georg. iii.8.23.

21 P.Kell. 75.31; P.Select. 18.40; P.Stras. iv.286.5, 16; PSI vii.838.6, 4th or 5th c. Kinship terms (both in reference and in address) are commonly used in papyri to close friends and in-laws as well as to relatives, and in some situations to more distant associates as well. Whenever such terms appear I have examined the letters for other evidence about the nature of the relationship concerned; if such evidence is present, the letters are classified as communication to family members or non-relatives accordingly. Sometimes, however, I can only conclude that the addressee is either a relative or a friend; in such situations I indicate the problem by putting inverted commas around the kinship term in question, as ‘brother’.

22 These centuries are treated together because many of the papyri involved cannot be dated to a specific century.

23 E.g. P.Cair.Mas. i.67020.5; P.Flor. iii.296.19; P.Lond. v.1674.21; P.Oxy. i.130.16, xvi.1944.12; SB vi.9239.19.

24 E.g. BGU ii.546.2, 547.6, 9; P.Cair.Mas. i.67068.14,67069.16,67076.12; P.Grenf. i.66.3; P.Herm. 16.6; P.land. ii.22.3; P.Köln vii.317.28; P.Lond. v.1786.29; P.Oxy. xvi. 1834.6,1858.4,1866.6, lix.4008.3; P.Prag. ii. 197.7; SB vi.9400.25,9616.37.

25 E.g. BGU xii.2135.10, xiii.2296.7; P.Oxy i.66.1, xii. 1470.1, xiv.1627.1, xvii.2113.27. See Hagedorn and Worp (n.2).

26 E.g. BGU i.315.1; CPR iv.16.2; P.Oxy. xxxiv.2729.3, xvi.1868.11.

27 E.g. BGU iv. 1033.19, v.1210.7.164; PSI v.447.22; P.Oxy. lix.3997.18.

28 There are 129 examples of δέσποτα in papyri of the 5th-7th c. and 90 in those of the lst-4th c., so the difference is not due to a decline in the amount of available data.

29 Cf. LSJ s.v. and Foerster and Quell (n.2) 1040-3.

30 46 examples in the Pentateuch, including Gen. 18:3, 19:18, 20:4, 23:6, 24:12, 24:18, 24:42, 31:35.

31 Cf. Wackernagel, J., ‘Lateinisch-Griechisches’, Indogermanische Forschungen 31 (1912-1913) 262–5.CrossRefGoogle Scholar

32 See Foerster and Quell (n.2) 1056-7.

33 In Paul only at Romans 10:16,11:3, and Heb. 1:10, all in direct quotations from the Septuagint; in Mark only at 7:28, to Jesus.

34 Matthew 13:37, 21:30, 25:11, 25:20, 25:22, 25:24, 27:63;Luke 13:8,14:22,19:16,19:18,19:20,19:25; John 12:21, 20:15.

35 An additional difficulty with the New Testament material is that much of the conversation reported in it would actually have been spoken in Aramaic, so that memories of the words originally used could in theory have influenced the Greek writers’ language: on some occasions κύριε seems to translate מרי (‘my lord’) and διδάσκαλε to translate רבי (‘my teacher’). Even if one genaccepts the likelihood of such influence, however, it would not explain the differences among the different Gospels, for it is thought that the later writers deliberately changed to κύριε addresses in their Greek sources which had used the term διδάσκαλε (Foerster and Quell (n.2) 1092-3).

36 Leg. 356; Philo also uses κύριε on a number of other occasions (Post. 132.6, Plant. 47.2,47.3, Ebr. 54.4, Conf. 173.5, Heres 20.9, Abr. 131.5, 6), but these are all quotations from the Septuagint.

37 1.29.48, 2.7.9, 2.7.13, 2.15.15, 2.16.13, 2.20.30, 3.10.15, 3.22.38, 3.23.11, 3.23.19, 4.1.57.

38 One might think that as the majority of papyri come from the Imperial period most words are likely to be unattested earlier, but there is in fact a considerable amount of papyrus material earlier than the first century AD; according to my estimates of the quantity of material various centuries in the Duke database (for which see below), the material from the first two centuries BC is equivalent in volume to that from the first century AD. However, one very fragmentary but allegedly early papyrus (BGU 1187.13) contains the words ]τελευτην κυριε[, and the editors suggest that this may be the vocative κύριε rather than a form of κυριεύειν, on the grounds that the latter would be expected to take the genitive rather than the accusative which seems to precede it. It is, however, unwise to assume in such a fragmentary document that the word preceding κυριε[ is necessarily the object of the verb, quite apart from the fact that κύριεύω sometimes takes an accusative object in papyri (cf. Mayser, E., Grammatik der griechischen Papyri aus der Ptolemäerzeit (Berlin 1934) 2.2.217Google Scholar). In addition, the papyrus is undated and has been assigned to the first century BC only on the evidence of the handwriting, which is not always reliable. It therefore seems unwise to count this fragment as the sole example of an address otherwise unattested at this period.

39 E.g. 2.7.9, 2.7.13, 2.15.15, 3.10.15, 3.23.11.

40 The amount of papyrus material in existence was calculated by searching the Duke database (see above) for the word καί, on the grounds that this word is common at all periods and in all types of document. For these purposes I used the search engine's date-specific search capacity and searched century by century; the result is that papyri dated to more than one century (e.g. second or third AD) have been counted more than once, and that in the (very small number of) cases in which the dating on the disk is inaccurate, those inaccuracies have not been corrected; moreover, instances of καί occurring only as supplements have been included. To correct for the fact that these parameters are different from those that I used for my own statistics on κύριε and δέσποτα, I thought it best to use, in this table alone, statistics for those words derived in the same way as the statistics for καί (in other words, the raw numbers generated by a century-by-century electronic search for the masculine vocative only). The numbers in the table are therefore based on the following figures: 2nd c. BC: καί = 10,166, κύριε = 0, δέσποτα = 0; 1st c. BC: καί = 4,573, lcupie = 0, δέσποτα = 1; 1st c. AD: καί = 14,021, κύριε = 21, δέσποτα = 3; 2ndc. AD: καί = 37,272, κύριε = 277, δέσποτα = 4; 3rd c. AD: καί = 26,393, κύριε = 209, δέσποτα = 25; 4th c. AD: καί = 17,493, κύριε = 276, δέσποτα = 78; 5th c. AD: καί = 5,208, κύριε = 47, δέσποτα = 44; 6th c. AD: καί = 19,382, κύριε = 46, δέσποτα = 111; 7th c. AD: καί = 7,165, κύριε = 6, δέσποτα = 44; 8th c. AD: καί = 2,782, κύριε = 0, δέσποτα = 2.

41 E.g. BGU iii.821.1 (father); P.Brem. 54.16 (‘brother’), 61.59 (nephew), 63.20 (daughter), 65.11 (friend); P.Flor. iii.332.20 (son); P.Giss. 11.20 (equal or subordinate), 15.6 (unrelated subordinate), 85.16 (brother); P.Haun. ii.16.19 (father); P.Mert. ii.82.7 (friend); P.Mich. iii.212.9 (‘son’), viii.477.23 (father), viii.480.5 (father), xv.751.9 (mother), xv.752.9 (mother); P.Mil. ii.87.17 (‘brother’); P.Oxy. xviii.2192.25 (‘brother’), xxxiv.2726.10 (business partner); P.Princ. ii.69.6 (‘brother’); PSI xiii.1359.5 (mother); P.Warr. 13 A 3 (‘father’); P.Wisc. ii.lxxi.25 (‘brother’); P.Würzb. 21 A 12 (father); SB iii.6263.8 (mother), viii.9903.17 (‘sister’), x.10277.12 (father), xiv.11900.14 (father).

42 E.g. BGU iii.816.28 (father), iii.949.7 (‘brother’), iv.1080.25 (son); CPR vii.57.21 (‘sister’, 3rd or 4th c); P.Berl.Zill. 12.4 (‘mother’, 3rd or 4th c ); P.Flor. ii. 154.7 (‘brother’), iii.338.16 (‘brother’); P.Harr. i.109.20 (unrelated subordinate, third or fourth century); P.Iand. vi.115.10 (‘brother’); P.Oslo iii.161.11 (mother); P.Oxy. i. 122.13 (unrelated subordinate, third or fourth century), i. 123.24 (son, third or fourth century), vi.937.9 (sister), xiv. 1678.4 (‘mother’), xiv. 1679.3 (‘mother’), xvii.2151.10 (‘mother’), xlii.3065.14 (mother); P.Rein. ii.116.5 (‘mother’); P.Ross.Georg. iii.2.2 (mother); P.Ryl. ii.441.3 (‘father’), iv.695.10 (‘brother’); PSI vii.833.8 (‘father’); P.Tebt. ii.420.16 (‘brother’); P.Find.Sijp. 26.22 (‘brother’); SB iii.6222.41 (‘sister’), iii.6262.24 (‘father’).

43 E.g. BGU iii.984.28 (‘brother’); CPR vi.82.12 (father), viii.28.4 (friend), viii.52.19 (father, 4th or 5th c ), xviia.39.5 (subordinate); P.Congr. XV 22.6.17 (mother); P.Harr. i. 110.4 (‘father’); P.Haun. ii.40.3 (‘brother’); P.Kell. 74.33 (father); P.Köln iv.199.12 (‘son’); P.Lond. ii.480.15 (‘brother’), v.1655 (orders to a tradesman), v.1659.15 (‘father’); P.Oxy. xii. 1424.21 (brother), xii. 1589.19 (subordinate), xiv. 1682.16 (wife), xiv. 1776.23 (subordinate), xx.2275.4 (‘brother’), xxxi.2602.3 (‘brother’), xlvi.3314.5 (wife), xlviii.3398.23 (brother), xlviii. 3399.10 (brother), xlviii.3430.29 (‘son’), lv.3818.8 (‘brother’), lvi.3858.7 (unrelated equal), lvi.3860.33 (husband), lvi.3861.25 (subordinate, 4th or 5th c.), lix.3998.15 (daughter), lix.4000.5 (‘father’); PSI ix.1082.3 (husband), x.1161.15 (‘mother’), xiii.1366.10 (‘son’, 4th or 5th c.), SB xiv. 11437 (daughter, 4th or 5th c.), xiv. 11588.1 (son), xiv.11881.8 (‘mother’), xvi.1260.10 (‘brother’), xviii.13589.3 (‘son’).

44 E.g. P.Amst. i.56 (unrelated equal, 6th c.); P.Batav. 21.13 (sister, 6th c.), 21.17 (‘daughter’, 6th c.); P.Bour. 25.7 (aunt, 5th c.); P.Mil. ii.87.17 (‘brother’, 6th c.); P.Oxy. x.1300.5 (‘mother’, 5th c.), lix.4004.19 (‘brother’, 5th c.); PSI v.478.18 (‘brother’, 5th c.); SB vi.9158.30 (‘mother’, 5th c.), xvi.12572.17 (son, 5th or 6th c.).

45 E.g. in contracts, to invoke divine protection: CPR 9.1.4, etc.

46 E.g. P.Amh. ii.143.15, 24 (4th c.); P.Berl.Frisk 3.3, 11 (3rd. c.); P.Cair.Isid. 66.3, 19 (3rd. c.); P.Giss. 17.3, 5 (2nd. c.); P.Haun. ii.25.7 (4th c.); P.Kell. 20.6, 17 (4th c.); P.Kell. 74.33 (4th c.); P.Herm. 6.4, 33 (4th c.); P.Lond. v. 1675.7, 8 (6th c.); P.Oxy. i. 123.7, 24 (3rd or 4th c.), xxvii.2479. 28 (6th c.).

47 Not, as used to be thought, because of Christian scruples about κύριος; see Hagedorn and Worp (n.2) 177; also Bréhier (n.5) 164.

48 Dineen (n.5) 56, 66, 76, 78.

49 Zilliacus, Familienbriefe (n.5) 32; Amantos (n.5) 20; Shipp, G.P., Modern Greek Evidence for the Ancient Greek Vocabulary (Sydney 1979) 347.Google Scholar

50 Sylloge Vocum Atticarum, s.v. δεσπότης (cited in Foerster and Quell (n.2) 1042).

51 Foerster and Quell (n.2) 1043-4.

52 Foerster and Quell (n.2) 1085.

53 Foerster and Quell (n.2) 1052; Baudissin and Eissfeldt (n.2) 2.298; also Svennung (n.1) 336.

54 See Baudissin and Eissfeldt (n.2) 2.298.

55 That is, Egypt (the papyri), Judaea (Philo and the New Testament), and Greece/Rome (Epictetus, who was born in Phrygia, taught in both places).

56 P.Oxy. ix.1202.15, xii. 1466.4; probably originally PSI x.1101.6.

57 Wackernagel, J., ‘Genetiv und Adjektiv’, in Mélanges de linguistique offerts à F. de Saussure (Paris 1908) 151Google Scholar; Svennung (n.1) 245.

58 The precise figures are: 1st century, with μου: κύριε / κυρία = 5, πάτερ = 0, άδέλφε = 2; without μου: κύριε / κυρία = 11, πάτερ = 7, άδελφέ = 35; 2nd century, with μου: κύριε / κυρία = 74, πάτερ = 1, άδελφέ = 2; without μου: κύριε / κυρία = 218, πάτερ = 14 άδελφέ = 159.

59 See Svennung (n.1) 336.

60 Cf. Biville, F., Les emprunts du latin au grec: approche phonétique (Louvain 1990-1995)Google Scholar and further bibliography cited therein, esp. 2.521-3.

61 E.g. Cassius Dio 55.12.2, 67.4.7.