1. Timarion, ed. Romano, R., Pseudo-Luciano, Timarione (Naples, 1974), ch. 6, 165-ch. 10, pp. 55–9.Google Scholar

2. See especially Genette, G., Narrative Discourse (Oxford, 1980)Google Scholar; Jenny, L., ‘The strategy of form’ in French Literary Theory Today, ed. Todorov, T., (Cambridge, 1982), pp. 34–63 Google Scholar; Foucault, M., ‘What is an author?’ in Textual Strategies: Perspectives in Post-Structuralist Criticism, ed. Harari, J. V., (Cornell, 1979), pp. 141–60 Google Scholar. On the importance of stylistics in the study of Byzantine texts, see Ševčenko, I., ‘Levels of style in Byzantine prose’, JÖB, XXXI (1981), 289–312.Google Scholar

3. On the influence and popularity of the Lucianic dialogues throughout the Byzantine period, see Tozer, H. F., ‘Byzantine satire’, JHS, II (1892), 233–70 Google Scholar. Other Byzantine dialogues in the genre which cannot be dated with the same certainty as Timarion include and see Hunger, H., Die Hochsprachliche profane Literatur der Byzantiner, (Munich, 1978), II, p. 153 Google Scholar. Historical events and persons alluded to in the Timarion are discussed by Tozer, op. cit., 285; Dräseke, J., ‘Michael Psellos im “Timarion”’, BZ, VI (1897), 483 Google Scholar; Romano, op. cit., pp. 25–31; Hunger, op. cit., II, p. 154; M. J. Kyriakis, ‘Satire and slapstick in seventh and twelfth century Byzantium’, V, (1973), 361 (wrongly dates Timarion to the eleventh century).

4. Hunger, H., Der byzantinische Katz-Maüse Krieg: Theodoros Prodromos Katomyomachia, (Graz-Wien-Köln, 1968), pp. 61–3 Google Scholar, suggests Theodore Pródromos as author on the basis of stylistic and other similarities between Timarion and the War of Cats and Mice. R. Romano, ‘In margine al problema della paternità del Timarione: sull’ anonimo dux di Tessalonica’, Vichiana, n.s. II (1973), 187–91, and (1974), 27–31, favours the iambic poet and doctor Nikolaos Kallikles as author, whose five Carmina Sepulchralia lamenting the death of Andronikos, son of Georgios and Anna, he cites in evidence. More recently Hunger, op. cit., II, p. 154, cautiously notes that ‘die Autorfrage konnte bis heute nicht geklärt werden’ (italics mine). In my view the question of authorship cannot be proved on the basis of linguistic and stylistic contingencies, and is in any case irrelevant to my argument, since the possibility of Kallikles composing the work with his patron in mind does not preclude an ironic level of writing — the same court writer could, after all, express different attitudes to his patron in different contexts, see P. Magdalino, ‘Byzantine snobbery’, in The Byzantine Aristocracy: University of Edinburgh Sixteenth Spring Symposium, ed. M. Angold, British Archaeological Reports (1983: forthcoming). Meanwhile the implications of Foucault’s penetrating theoretical analysis of changing concepts of authorship in European literature and thought from antiquity to our own times (1979), pp. 141–60, remain to be worked out in detail with reference to anonymously transmitted texts in both learned and vernacular language during the last four centuries of Byzantine rule.

5. The targets and nature of the satire are summarised and discussed by Tozer, op. cit., (1892), 237; Dräseke, op. cit., (1897), 483; Romano, op. cit., (1974), pp. 19–23; Hunger, op. cit., (1978), II, pp. 153–4, 290. Tozer’s view on the ‘gentleness’ of the satire is now outdated: the difference he notes between Timarion and the thirteenth-century poem Mazaris are to be related rather to a perceptible shift in the thirteenth century from irony to invective, see Magdalino, op. cit.

6. The text of the letter was first edited, with commentary, by M. Treu, ‘Ein Kritiker des Timarion’, BZ, I (1892), 361–5, and is included in Romano’s introduction, (1974), pp. 23–31. On the accusations against Psellos and Italos in the text, see Clucas, L., The Trial of John Italos and the Crisis of Intellectual Values in Byzantium in the Eleventh Century (Miscellanea Byzantina Monacensia, XXVI) (Munich, 1981), p. 139 and n. 511.Google Scholar

7. See below, pp. 33–5 and notes 9 and 13.

8. On the significance of Paphlagonians as pig-eating barbarians, see Hendy, M. F., Studies in the Byzantine Monetary Economy, (Cambridge, forthcoming).Google Scholar

9. Genette’s terms of narratological analysis (1980) are not easily reducible to brief summary. The main points are developed from Russian formalism and refined according to the principles of French structuralism and Anglo-American criticism, illustrated throughout by detailed application to Marcel Proust’s Recherche du temps perdu. First, Genette insists upon the proper distinction between the three levels of meaning contained in the word ‘narrative’, as follows (pp. 25–32): (1) story (‘histoire’): signified content of events and actions narrated, (narrative as story, e.g. Timarion’s sufferings and descent to Hades); (2) narrative (‘récit’): signifying statement, or written/oral text in which the story is narrated (narrative as discourse, e.g. the text of Timarion); (3) narrating (‘narration’): instance or event of telling story, or the producing of narrative action (narrative as relating event, e.g. Timarion’s dialogue with Kydion in twelfth-century Constantinople). The inter-relations between these three levels of narrative are defined in terms borrowed from the grammar of verbs: tense = temporal relations between narrative and story (e.g. the ‘narrative’ of Timarion is located in the twelfth century, while the ‘story’ moves from the twelfth century to antiquity and back, in cyclical form); mood = the varying forms and degrees of representation, especially of dramatisation (‘mimesis’) against telling (‘diegesis’) in the story and narrative (e.g. the narrative form of the Timarion, as dialogue, is mimetic, while the story contains both mimesis in live and reported conversations and diegesis in events and actions narrated); voice = varying ways in which the narrating event is implied in the narrative (e.g. our story is told by a first-person hero/narrator, while Kydion’s frequent interjections serve as constant reminders of the narrating instance, which never add to the story as such). Temporal relations between these three levels of narrative are further defined by the terms prolepsis (to denote movement from narrative present to story future); analepsis (to denote movement from narrative present to story past); ellipsis and paralipsis (to denote the deliberate omission or side-stepping of essential elements in the story, e.g. the suppression of information on ‘Who is the dux?’); and paralepsis (to denote the deliberate excess of information, e.g. praise of the dux). These terms belong to the necessary anachrony of narrative: their reach into the past and future is determined by their distance from the present narrating instance; their extent is determined by the duration of the narrated event in the story. Narrative speed, or tempo, is determined by the number of words devoted to the description of events of differing duration. It should emerge from this summary that narratological order is considerably more complex than Kydion supposed!

10. Although not necessarily relevant to twelfth-century actuality, the tradition of Cappadocians as ‘dirty barbarians’ enjoyed considerable force as a literary stereotype, see the tenth-century invective of Liuprand of Cremona against the Emperor Nikephoros, as ‘bristly, wild, rough, harsh, hairy, a rebel — a Cappadocian!’, Antapodosis, VI, 10.

11. Tozer, op. cit., 235, and particularly Dräseke, J., ‘Byzantinische Hadesfahrten’, Neue Jahrbuch für das klassisches Altertumswissenschaft, XXIX (1912), 353 Google Scholar, make this assumption in the most naive terms (‘doch dürfte es sich empfehlen, Timarion für den wirklichen Namen des Verfassers zu halten’). The question is relevant to the study of many Byzantine texts, in particular to the understanding of the four different ego-personas assumed by the author of the so-called Prodromic poems. Genette (op. cit., pp. 213–4) formulates clearly the need to distinguish the narrating instance from the writing instance, noting that the confusion between the author, the narrator and the recipient of a narrative is illegitimate in fiction, since ‘the role of narrator is fictive, even if assumed by the author directly’, hence the narrating situation of a fictional account cannot be reduced to the situation of writing.

12. Ch. 16, 412–23. ‘Mythic perspective’ is a term used by P. Zumthor to define one of the means by which medieval narrative traditionally refers to events outside the story to provide a different perspective on the narrative present, ‘The great game of rhetoric’, New Literary History, XII, 3 (1981), 506. The major ‘mythic perspective’ of the Timarion is provided by the traditional framework of the descent to Hades. It is fully exploited as a means of satire by the juxtaposition in Hades of figures drawn from contemporary or recent history with figures drawn from a more mythical past. The picture of society in Hades as stratified according to social status cannot therefore be taken as a direct comment on twelfth-century Byzantium, but rather as a significant projection of present upon past.

13. See Tozer, op. cit., 256, and Genette, op. cit., p. 111. The disorder could be interpreted as a parody of Byzantine traditional forms of diegesis recommended in handbooks such as that of the fifth-century Nicholas of Myra, Progymnasmata, ed. J. Felten, Rhetores Graeci, II (Leipzig, 1913), where six elements of diegesis are defined as (p. 13), and its five virtues as (p. 14). This handbook was used by Byzantine writers of the tenth to the fourteenth centuries, including Maximos Planoudes (pp. IX-XIX). I owe this reference to the kindness of Mary Cunningham. The rules are broken by Timarion, who persistently flouts Kydion’s injunction to adhere to the proper of the narrative, and accuses Kydion of creed , ironically pardoning himself in advance for omitting reference to every crow, stone and wayside bramble that may have impeded his ride (ch. 3, 64–9, cf. ch. 7, 206–9, 230–1).

14. See eh. 5–7. S. Vryonis has provided a detailed commentary on this passage, as well as a translation of a substantial part of the text, ‘The panegyris of the Byzantine saint’, in The Byzantine Saint: University of Birmingham Fourteenth Spring Symposium of Byzantine Studies, ed. S. Hackel, Sobornost, V (1981), 202–4.

15. The arguments for and against these historical figures are presented and reviewed by Polemis, D., The Doukai: a Contribution to Byzantine Prosopography (London, 1968), p. 153 Google Scholar, n. 5, who cautiously points out that the text is not explicit; cf. p. 3. Michael Palaiologos was first proposed by Tozer, op. cit., 235, and has received further support from Romano, op. cit., (1973), 187–91, and op. cit., (1974), pp. 27–8. If the author’s patron was in fact a member of the Doukas-Palaiologos family, then the audacity of the ‘panegyric’ would explain the anonymity of the dux.

16. See Polemis, op. cit., pp. 74–5.

17. See Zumthor, op. cit., 495, with reference to court poets and rhétoriqueurs in fifteenth and sixteenth-century France.

18. See Timarion, ch. 22, 550–1 (Romanos IV); ch. 33, 804 (Theophilos); ch. 41–5, (Michael Psellos); ch. 43–4, 1077–1115 (John Italos); ch. 23, 588–600, ch. 39, 976–80, ch. 46, 1158–69, ch. 47, 1192 (Theodore of Smyrna). For discussion of these historical figures, see above, n. 5.

19. See Tozer, op. cit., 245–6.

20. Hunger, op. cit., II, p. 152.

21. For references to stagecraft and spectacle, see Timarion, ch. 5, 121, 126, 127, 129, 138, 144–5, ch. 6, 148, 149, 159, 165–7, ch. 7, 181–3, ch. 9, 255, ch. 10, 280. The question deserves separate and detailed investigation. For modalising locutions, see ch. 5, 114, 133, ch. 6, 171, 175, ch. 7, 184–5, 189, 197, 203, ch. 9, 233, 235, 239–40, 247–8, 249, 258, ch. 10, 284. On their significance in narrative, see Genette, op. cit., p. 102, where his implied analogy between medieval ekphrasis, oral poetry and Proustian description suggests that, absorbed into the narrative through these means, the resulting description is contemplation as an intense activity, not as passive recollection.

22. the commonest dependent genitive is ‘army’, ‘people’; but cf. (Philo 1.322) and (Ar. Eq. 852). The word is not listed in lexica of New Testament or Patristic Greek, but its current use with is cited by D. Dimitrakos, (Athens, 1936), s.v. Byzantine writers normally use the word in a military context, see Choniates, ed. I. A. Van Dieten, (CFHB: Berlin, 1975), 31.25, 185.31, 379.92, 391.42, 532.28, 552.77, 567.46, 599.27. The association with is unusual, and possibly ironic. (References to the text of Choniates, here and elsewhere, are owed to the kindness of Professor Kazhdan).

23. for an adulteress, see Vict. Mc. 8: 38, (p. 351, 4): (fifth century), Patristic Greek Lexicon, ed. G. W. H. Lampe (Oxford, 1961), s.v. For homosexual context, see Meleager, AP, XII, 158.2 = The Greek Anthology: Hellenistic Epigrams, ed. A. S. F. Gow and D. L. Page (Cambridge, 1965), 1.4497 and II, p. 657: … (reference owed to kindness of Edward Whittle). For movement of palm grasping for money, see Leontios, AP, IV, 272. The passive sense, ‘be laid upon’ (MPol. 2.4), was later generalised to mean ‘carry’, see Choniates 222.80. The question of authorial intention in the case of ironic levels of meaning is not relevant, since metaphorical uses of words may become buried or obsolete, but they are never deleted from the Greek language, see Derrida, J., ‘Violence and metaphysics’, Writing and Difference (London, 1978), p. 82 Google Scholar, and Benveniste, E., Problèmes de linguistique générale (Paris, 1966), pp. 289–90 Google Scholar, who notes that ‘le “sens” d’une forme linguistique se définit par la totalité de ses emplois, par leur distribution et par les types de liaisons qui en résultent’. This last point is particularly relevant to the decoding of irony in a text where, as in our passage, almost every phrase is ‘laid upon’ allusions to other texts, both classical and religious, see Romano, op. cit., (1974), pp. 19–20.

24. see Liddell-Scott-Jones, s.v. The word is not listed in lexica of New Testament, Patristic or Byzantine Greek, but, for pejorative uses, see Choniates, 99.37, 277.40, 373.69, 392.61.

25. For fanciful interpretations of the name Palaiologos, see Geanakoplos, D. J., Emperor Michael Palaeologus and the West, 1258–82, (Cambridge, Mass., 1959), pp. 17–18 Google Scholar. n. 5. means ‘an old thing’, ‘person of advanced years’, or ‘person knowing and telling past things’; while the verb can be used pejoratively for discussing out-of-date things or for using an out-dated, antiquarian style, see Liddell-Scott-Jones, s.v.v. The verb is used pejoratively in A. PV, 317, S. OT, 290. In Modern Greek the derogatory expression is attested in the Lexica of Proia and the Academy of Athens, s.v.

26. Excessive compliments to the Doukas family on the nobility of its ancestry are found in Psellos, ed. C. Sathas, (London, 1899), p. 234. 8–9: Such reports were probably fabricated, see Polemis, op. cit., p. 13, who also points out that the Akritic epic is, effectively, a twelfth-century glorification of the Doukai of the ninth and tenth centuries, where the names constitute distant echoes of historical realities, but the individual is purely fictitious, and his genealogy ‘quite unfounded’; ibid., pp. 14–15.

27. Tozer, op. cit., p. 246, cf. Romano, op. cit., (1974), pp. 27, 130–1, Hunger, op. cit., II, p. 188.

28. See Jenny, op. cit., pp. 55–61.

29. See Frisk, H., Griechisches etymologisches Wörterbuch, (Heidelberg, 1954–70)Google Scholar, s.v.

30. Od. IV. 230, cf. Eustathios of Thessaloniki, Commentarli ad Homeri Od., ed. van der Valk, (Amsterdam, 1960), I, p. 262. For a penetrating discussion of the significance of the Homeric passage and its interpretations in later writers, see Bergren, A. T. L., ‘Helen’s “good drug”’: Odyssey IV.1–305, Contemporary Literary Hermeneutics and the Interpretation of Classical Texts, (Ottowa, 1981), pp. 201–14.Google Scholar

31. By the twelfth century, peithṓ seems to have become an acceptable quality in generalship and aristocracy, see Psellos, ed. Sathas, 90.8, 168.32, 178.6, 208.13, 16. In this passage, however, the Platonic opposition between peithṓ and alḗtheia is evoked (Phd. 260a), as is Plato’s denunciation of the Gorgianic equation of peithṓ with the power of language to poison and enchant the soul (Smp. 203c-e, cf. Gorgias Enc. Hel., ed. A. Diels and F. V. Kranz, Fragmente der Vorsokratiker, (Berlin, 1955), no. 82, pp. 131–4. Further, the context of peithṓ here is effeminacy of social intercourse, not generalship.

32. On Sapphic allusions and citations in Byzantine texts, see: Cataudella, Q., ‘Saffo e i Bizantini’, REG, LXXVIII (1965), 66–9 CrossRefGoogle Scholar, who points out that they became fashionable in the twelfth-century erotic romances of Eustathios Makrembolites and Niketas Eugenianos; and Garzya, A., ‘Per la fortuna di Saffo a Bisanzio’, JÖB, XX (1971), 1–5 Google Scholar, who observes that the context of Sapphic allusion in Byzantine writers is always nuptial or erotic. In our passage, there is triple literary allusion: first, to Aristotle (Eth. Nic. 1145a, 29), where the discussion is about the relationship between Intellect and Desire, and the oppositions vice/virtue, self-indulgence/self-restraint; second, to Plato (Men. 99d), where the practice of women and Lakonians in calling politicians and other fallible mortals is mentioned; and third, to Sappho (fr. 31, ed. D. L. Page (Cambridge, 1965), no. 1), where the epithet is used of the fortunate mortal who sits opposite the object of the speaker’s desire, rendering her senseless with envy and ‘not… far short of death’. The Doric form closely linked in the text with suggests an ironic hyper-emphasis on the ‘Lakonian tongue’. The verb implied in text and context, means both ‘speak in a Lakonian manner’ and ‘be a pederast’ (Liddell-Scott-Jones). On homosexuality and pederasty among the Spartan aristocracy, as seen from the perspective of Attic comedy, see Henderson, J., The Maculate Muse: obscene language in Attic comedy (New Haven, 1975), pp. 204–5 Google Scholar, 211, 218 ().

33. See, for example, the seals and inscriptions cited by Polemis, op. cit., pp. 152–3.

34. See Magdalino, op. cit.

35. The standard rules on which most Byzantine enkomia are directly or indirectly modelled can be found in Menander Rhetor, ed. D. A. Russell and N. G. Wilson (Oxford, 1981). It is remarkable how few are observed by the author of the Timarion, and how many are inverted. The order of priorities recommended by Menander is: country and city of origin (if famous); family (if good); birth, nature, nurture; accomplishments and actions in war and peace, including mention of all military campaigns; cardinal virtues see II, 368–77, pp. 76–94, and II, 388–94, pp. 114–26. Strictly speaking, all the ‘good’ qualities of our dux fall outside the range of qualities which convey prestige; wealth and good birth are a matter for congratulation rather than praise, while physical beauty is appropriately praised in non-human subjects, like plants (see pp. xxi, xxvii-xxviii).

36. See Zumthor, op. cit., 488–9.

37. Ch. 46, 1170–8.

38. Tozer, op. cit., 243; cf. 250.