Hostname: page-component-cd9895bd7-gbm5v Total loading time: 0 Render date: 2024-12-23T13:10:20.357Z Has data issue: false hasContentIssue false

Some Aspects of Byzantine Military Technology from the Sixth to the Tenth Centuries

Published online by Cambridge University Press:  22 January 2016

J. F. Haldon*
Affiliation:
University of Birmingham

Extract

Byzantine military technology remains to be examined thoroughly, despite its importance in demonstrating the Byzantines’ ability to absorb ideas and practices from areas outside me Empire as well as developing their own traditions. This article examines arms and armour from the mid-sixth century to the end of the tenth century and puts the military panoply into a wider context, that of the development of offensive and defensive military equipment in Europe and the lands to the north and east of the Empire. Finally, it examines die relationship between military affairs and the economic and political situation of the Empire and attempts to account for the use of particular weapons and methods of equipping die soldiers at particular times.

Type
Research Article
Copyright
Copyright © The Centre for Byzantine, Ottoman and Modern Greek Studies, University of Birmingham 1975

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. See Darkó, E., ‘Influences touraniennes sur l’évolution de l’art militaire des Grecs, des Romains et des Byzantins’, B, x (1935), 44369 Google Scholar; xi (1937), 119-47; idem, ‘Le rôle des peuples nomades cavaliers’, B, xviii (1948), 85-97; and for works which examine Byzantine weapons and armour, see Oman, C. W., The Art of War in the Middle Ages, I (London, 1898-1924), pp. 22ff;Google Scholar Lot, F., L’Art Militaire et les Armées au Moyen Age en Europe et dans le Proche Orient, I (Paris, 1946), pp. 27ff.;Google Scholar Aussaresses, F., L) Armée Byzantine à la fin du 6e siècle d’après le Stratégicon de l’Empereur Maurice (Bibliothèques des Universités du Midi, fasc, xiv, Paris, 1909)Google Scholar. The last four authors did little more than paraphrase the various tactica where arms and armour were concerned. See also Pertusi, A., ‘Ordinamenti militari, guerre in Occidente e teoria di guerra dei Bizantini (secc. VI-X)’, Ordinamenti militari in Occidente nell’ alto medioevo (Spoleto, 1968), 631700 Google Scholar.

2. Eadie, J. W., ‘The Development of Roman Mailed Cavalry’, JRS, lvii (1967), 16173 Google Scholar.

3. See Eadie, , op. cit., 167F.; and the descriptions in Marcellinus, Ammianus, Works, ed. and trans. Rolfe, J. C. (Loeb, London/Cambridge, Mass., 1935), xvi Google Scholar, 10, 8; and Julian, , Letters and Works, ed. and trans. Wright, W. C. (Loeb, London/Cambridge, Mass., 1954), Oratio, i, p. 96 Google Scholar, of Constantius’ heavy cavalry. See also Vegetius, , Epitoma Rei Militaris, ed. Lang, C. (Teubner, Leipzig, 1885), iii, 23; 24 Google Scholar.

4. On this development see Bivar, A. D. H., ‘Cavalry Equipment and Tactics on the Euphrates’, DOP, xxvi (1972), 2736 Google Scholar, 284f.

5. Bivar, loc. cit.

6. For reforms in the 580s and after, see Protector, Menander, in Fragmenta Historicorum Graecorum, IV (ed. C. Muller, Paris, 1851), frg. 58, 9-17 (p. 113)Google Scholar, and Maurice, , Strat. (Mauricius, Arta Militará Google Scholar, ed. H. Milhãescu [Bucharest, 1970]), proem. 2. On the various types of composite bow, see Rausing, E., The Bow (Acta Archaeologica Lundensia, vi [1967])Google Scholar. For its construction, see Brown, F. E., ‘A Recently Discovered Compound Bow’, Seminarium Kondakovianum, ix (Prague, 1937), 110;Google Scholar Payne- Gallwey, R., Projectile-Throwing Weapons of the Ancients: with a Treatise on Turkish and other Oriental Bows (London, 1907)Google Scholar. There is some dispute as to whether the Byzantines really adopted the Hun bow or not; but judging from Procopius’ description and that of the anonymous treatise on archery (see below) it is very probable that they did. See Procopius, , Wars, ed. H. B. Dewing(Loeb, London/Cambridge, Mass., 1940), i, 18. 32 Google Scholar.

7. Wars, i, I.9-15.

8. Scale armour was constructed from a number of small plates of iron, bronze, bone, wood, horn, or leather, depending upon the natural materials most easily obtained, which were pierced at the top and sewn on to a backing garment, usually leather or linen. As a result of the sewing, the backing naturally became puckered and stiff, preventing easy movement; consequently, most of the scale defences we know of are fairly short, having either no sleeves at all or very short ones. For examples, see the second-century B.C. fresco at Pantikapaion, Kertch, in M. Aschik, Sur une Chambre Sépulcrale de Panticapée, ornée de Fresques (Odessa, 1845), pl. iv.

9. For a general note on scale armour and its history, see Russell-Robinson, H., Oriental Armour (London, 1967), pp. 27 Google Scholar. For finds of Scythian scale, see Minns, E. H., Scythians and Greeks (Cambridge, 1913), pp. 1837 Google Scholar, 225, 231. For representations of Scythian and Sarmatian soldiers wearing such armour, see Compte Rendu de la Commission Impériale d’Archéologie (1871), Supplement by L. Stephani, 295-312 and pls. vi, ix, x (frescoes of the first century A.D. or later); Compte Rendu … (1874), 115, fig. 2 (a second- or third-century B.C. fresco); stylized Sarmatian soldiers are also shown in Trajan’s column: see Cichorius, C., Die Reliefs der Traiansãule (Berlin, 1896 and 1900), pls. xxiii and xxviii Google Scholar; and for Roman booty, pls. ii and iii. For Roman scale, see pl. lxxxvi and also Caprino, C. et al., La Colonna di Marco Aurelio (Roma, 1955)Google Scholar, figs. 29, 30, 42, 43, 54. 90, etc.; Guiliano, A., Arco di Constantino (Milano, 1955), figs. 6, 18, 20, 32 Google Scholar; and Florescu, F. B., Monumentul de la Adamklissi Tropaeum Troiani (Bucharest, 1909), figs. 186b, 196 Google Scholar. Cf. Webster, G., The Roman Imperial Army (London, 1969), pls. iv, v Google Scholar. For scales from Roman sites, see Groller, M. von, Der Rõmische Limes in Österreich II (Vienna, 1901), cols. 85-95, pls.xv, xvi Google Scholar; IV (Vienna, 1903), cols. 103-4, fig.46. For scale in Persia and the Middle East, see Bivar, op. cit., pp. 273-4. Note also the two scale horse-trappers, one of bronze, the other of iron, excavated at Dura Europos, dating to the third century A.D., in Rostovtzeff, M.I., ed., The Excavations at Dura Europos: Preliminary Report of the Second Season of Excavation (New Haven, 1931), pp. 194200 Google Scholar. Cf. the graffito of a heavily armoured horseman at Dura, showing the horse armed with a similar trapper, in Ghirshman, R., Iran: Parthians and Sassanians (London, 1962), fig. 63c Google Scholar; and the paintings at Khalchayen in central Asia, in Belenitsky, A., Central Asia (Geneva, 1968), p. 101 Google Scholar, where again the horse is armed with a scale trapper. Cf. the second- or third-century A.D. rock relief at Tang-i-Sarwak in north-east Iran, in Girshman, op. cit., fig. 69. For scale cuirasses worn in conjunction with laminated arm- and leg-guards, see the Sassanid reliefs at Naqsh-i-Rustam, in Schmidt, E. F., ed., Persepolis III (Chicago, 1970), pls. 89, 91, 93, 95 Google Scholar.

10. Cf.Russell-Robinson, , Oriental Armour, p. 4 Google Scholar, fig. 2; Persepolis II (Chicago, 1957), pl. 77, nos. 9, 10; Thordemann, B., ‘The Oriental Splint Armour in Europe’, Acta Archaeologica Copenhagen, IV (1933)Google Scholar, fig. 20. Thordemann gives a good summary of the history and use of lamellar, pp. 125ff.

11. Westholm, A., ‘Cypro-Archaic Splint Armour’, Acta Arch. Copenhagen, IX (1938), 16373 Google Scholar; Laufer, B., Chinese Clay Figures, part one: Prolegomena on the History of Defensive Armor (Chicago, 1914), pp. 25891 Google Scholar. For a general note on the origins of this type, see especially pp. 273-4.

12. For lamellar such as was used among the nomads, see the Pantikapaion fresco in Aschik, op. cit., pl. iv, where several of the soldiers wear ankle-length coats of lamellar; also the Khalchayen fresco in Belenitsky, op. cit., p. 101. For finds of lamellar in Eurasia and Korea, see Laufer, op. cit., pp. 35, 201-36; Bivar, op. cit., 274, and n. 8; Thordemann, op. cit., 127ff., 146, n. 40; W. Arendt, ‘Ein alttürkischer Waffenfund aus Kertsch’ Zeitschrift für Historische Waffenkunde, XIII (July 1932), figs. 3, 3a, 3b; F. Altheim, Attila et les Huns (Paris, 1952), pp. 21-4. For lamellar in Persia and Europe, see the Dura preliminary report, second season, loc. cit., and the lamellar represented on a mid-mird-century Palmyrene relief, in Ghirshman, Iran: Parthians and Sassanians, fig. 10. For finds of Roman lamellar of a similar period, see von Groller, op. cit., II (1901), cols. 85-95, pls- xv, xvi; IV (1903), fig. 47; VI (1905), fig. 56/3; VII (1907), fig. 11/12; IX (1908), fig. 10. Both the Romans and the Parthians were fully acquainted with lamellar by this time.

13. For the Roman material, see below. For the Persian evidence, note the lamellar horse trapper of Chosroes II depicted on the relief at Taq-i-Bostan; the warriors fully equipped with such armour shown on a silver plate of the sixth century; and the fragments of lamellar excavated from the site at Kasr-i-Abu, near Shiraz. See Pope, A. U., ed., A Survey of Persian Art (London/New York, 1938-9), IV, pl. 161a Google Scholar; Gardner, E. A., ‘Preliminary Report of the Excavations at Kasr-i-Abu’, Bulletin of the Metropolitan Museum of Art (December 1934)Google Scholar, sect. 2, fig. 11; for the plates, see Coq, A. von Le, Bilderatlas zur Kunst- und Kulturgeschichte Mittel Asiens (Berlin, 1925)Google Scholar, fig. 74.

14. See Thordemann, op. cit., p. 129; and for its use at a later date, Hewitt, J., Ancient Armour and Weapons in Europe (London, 1855), pp. 132ff Google Scholar.

15. Russell-Robinson, op. cit., p. 9.

16. European commentators noted its use among the Mongols. See The Texts and Versions of John of Plano Carpini, ed. Beazley, C. R. (London, 1903), pp. 89, 124 Google Scholar; and The Journey of William of Rubruck, ed. Rockhill, W. W. (London, 1900), p. xv Google Scholar. For Japanese lamellar, see the chapter devoted to this subject in Russell-Robinson, op. cit.

17. On Byzantine armour of wood, see below, n. 60. Wooden armour was used by the nomads and by some north American Indian tribes. See Laufer, op. cit., pp. 204, 276. For further examples, see Stone, G. C., A Glossary of the Construction, Decoration and Use of Arms and Armor (New York, 1934), p. 54 Google Scholar, fig. 71; p. 67, fig. 87.

18. For Scythian examples, see Gardner, E. A., ‘Ornaments and Armour from Kertch in the New Museum at Oxford’, JHS, v (1884), 65 Google Scholar, pl. xlv, 1; Antiquités de la Région du Dnièpre (in Collection B. Khanenko, Kiev, 1899), II, pp. 7, 19, and pl. vii. Also Arwidsson, G., ‘Armour of the Vendel Period’, Acta Arch. Copenhagen, X (1939), 54 Google Scholar, n. 35, where other finds of splint from Kertch are detailed. For Avar splint greaves/shinguards, see Arendt, op. cit., figs, 1b, 3b.

19. The Dura graffito shows a rider with a double row of splints around his trunk, with the area below and above protected by mail or scale armour joined to the splints. His limbs are protected by laminated defences and he wears a conical helm with an aventail of scale or mail. For a similar outfit, compare the royal figures in the Naqsh-i-Rustam reliefs (above, n. 9). Such an outfit was adopted by the Romans during the fourth century: compare the descriptions given by Ammianus and Julian (n. 3) with the Dura graffito. Similar armours have been excavated from seventh-century A.D. Vendel graves in Sweden (see G. Arwidsson, ‘Armour of the Vendel Period’, 38ff.) and from Merovingian graves at Schratzheim in Bavaria (ibid., 32, n. 3). The northern peoples of the migration period were keen to obtain or copy Roman armour, as their adoption of late Roman helmet styles illustrates. See G. Arwidsson, ‘A New Scandinavian Form of Helmet from the Vendel Time’, Acta Arch. Copenhagen, V (1934), 31-59. Compare the parade helmets described in Ammianus and Julian and the second-century helmets excavated from Roman forts at Newstead and Ribchester, in J. Curle, A Roman Frontier Post and its People (Glasgow, 1911), pp. 179-80, pls. xxix, xxx; Webster, op. cit., pl. xxii. Cf. the helmet of Chosroes II at Taq-i-Bostan(in Pope, op. cit., pl. 161a); for a better view and comparison, see Russell-Robinson, op. cit., fig. 11; and some of the Vendel helmets in Arwidsson, ‘A New Scandinavian Form of helmet’; for other attempts to imitate Roman equipment, see Arendt, op. cit., figs. 6 and 7. The composite armour of scale or mail and splints proved to be one of the most durable forms of defence in the east, where it was especially popular in Persia, central Asia, and China: see Arwidsson, ‘Armour of the Vendel Period’, 54, 59; Russell-Robinson, op. cit., pp. 31-3, 61-3.

20. Art of Horsemanship, ed. and trans. Marchant, E. C. (Loeb, London/Cambridge, Mass., 1946), xii, 5 Google Scholar.

21. Russell-Robinson, op. cit., p. 13, fig. 6b.

22. Webster, op. cit., pp. 123-6.

23. Cichorius, op. cit., pls. xii, xiii, xviii, etc.

24. Cichorius, op. cit., pls. ii, iii; Florescu, op. cit., figs. 196, 197, 199, 201, 202, where laminated arm defences are worn by Roman auxiliary troops.

25. See n. 24, and the laminated ‘sleeves’ of the soldiers in the Pergamum reliefs (Russell-Robinson, op. cit., fig. 6b).

26. For Newstead, see J. Curle, op. cit., pp. 157-8, fig. 11; for Carnuntum, see von Groller, op. cit., II (1901), pls. xvii; xx, 6-10; for the Dura graffito, Ghirshman, op. cit., fig. 63c; for the Naqsh-i-Rustam reliefs see n. 9 above; and for those at Firuzabad, see Ghirshman, op. cit., figs. 165, 166. For the Palmyrene relief, see Ghirshman, op. cit., fig. 84.

27. See the silver plate in Le Coq, op. cit., fig. 74, where two dismounted warriors are shown, their legs encased in such defences. See also the references of Julian and Ammianus above, note 3.

28. Op. cit., pp. 276-8.

29. For a further introduction to mail, see Laufer, op. cit., pp. 237-57; on mail and its construction, see Russell-Robinson, op. cit., pp. 10-11, 12, fig. 5; Bivar, A. D. H., Nigerian Panoply (Lagos, 1964); Stone, op. cit., pp. 42430 Google Scholar, 36ff.

30. See Dain, A., ‘Les stratégistes Byzantins’, Travaux et Mémoires, II (1967), pp. 31792 Google Scholar, for the most detailed and comprehensive examination of these tactica.

31. Anon., in Griechische Kriegsschriftsteller, ed. H. Köchly and W. Rüstow, II (Leipzig, 1855), ii, pp. 42-96: xvi,

32. Wars, i, I, 9-15.

33. This small face-guard is yet another example ot the Byzantines adopting nomad panoply along with nomad tactica. The Chinese T’ang annals record that the Kirghiz also strapped a small shield to their shoulder for a similar purpose. See Laufer, op. cit., p. 204. Persian miniatures of the thirteenth century also illustrate soldiers similarly equipped, according to the Mongol fashion of the time. See Russell-Robinson, op. cit., p. 33, figs. 17a, b. The Byzantines must have adopted this from the Huns.

34. On this, see the anonymous treatise (in Köchly and Rüstow, op. cit., pp. 109-208, i. 7; i. 9; and Bivar, op. cit., pp. 284-5.

35. xvi. 4,

36. xvi. 9. On leather/ felt head armour, see below, n. 130. Felt armour was common in the east, and was used by the Sassanid cavalry. See Laufer, op. cit., pp. 290, 292. See Theophanes, Chronographia(de Boor), p. 318, 28.

37. xvi. 9.

38. On Byzantine measurements, see Schilbach, E., Byzantinische Metrologie (Munich, 1970), pp. 16ff Google Scholar. A dactyl was ¾“, a palaistê or palm 4 dactyls or 3”, and a spithamē or span 3 palaistai, or 9”.

39. See below on the tenth-century treatises.

40. xvi. 8.

41. xvii. 3.

42. xvii. 4.

43. For felt trappers, see the Firuzabad reliefs, especially those depicting Ardashir 1 and Shahpur defeating their opponents, in Ghirshman, op. cit., figs. 165, 166.

44. xvi. 9.

45. See above, n. 6. On the problem of the composition of the treatise, see the summary of Moravcsik, Gy., Studia Bizantina (Budapest, 1967), pp. 2223 Google Scholar; and Dain, ‘Stratégistes’, pp. 344ff.

46. Strat. i, 2. 2. These are ankle-length for the cavalry, but knee-length for the infantry. For the practicability of such long coats, see below, n. 126.

47. vii, 15. 2;x, 1. 4.

48. i, 2. 2.

49. i, 2. 3.

50. i, 2. 2. Aussaresses calculated the lance, from internal evidence, to be about 3.60 m. in length (op. cit., p. 51, n. 3).

51. i, 1. 5; xi, 2. 6.

52. There were two main types of bow-case, that in use with the Iranian peoples, the Chinese, and in Europe; and that used on the steppes by people of Turkic/Mongol stock. The first was a case proper, with a flap, slung from saddle or waist-belt. A number of Byzantine military handbooks refer to the carried by the imperial troops, e.g., Maurice, Strat., ii, 1; i, 2. 2. See also Leo, Tactica (Sylloge Tacticorum Graecorum, III), ed. R. Vari; Leonis Imperatoris Tactica, I and II (Budapest, 1917) v, 3; vi, 2; and Syll. Tact. (Sylloge Tacticorum quae olim ‘inedita Leonis tactica’ dicebatur), ed. A. Dain (Paris, 1937), 39, 4. The nomad bow-case, on the other hand, was rather like a sword scabbard, and could be used only when the bow was unstrung (note Maurice, Strat., xi, 2. 6; Leo, Tact., xviii, 49). See Gy. Laszlo, ‘Contribution à l’archeologie de l’époque des migrations, ii: Le carquois d’arc des Hongrois conquérants’, Acta Archaeologica Academiae Scientiarum Hungaricae, VII (1957), 172ff. For illustrations, see the eighth-century A.D. Pendjhikent frescoes in Belenitsky, op. cit., pl. 137, the first- or second-century A.D. rock drawing at Suljek in Siberia (Laufer, op. cit., fig. 35) and the ninth-century silver dish from Perm (Belenitsky, op. cit., pl. 74). There seems to have been only one basic type of quiver, an elongated trapezoidal case with a cover, shown on most of the illustrations depicting mounted soldiers referred to in this article. See Gy. Laszlo, ‘Études archéologiques sur l’histoire de la société des Avares’, Archaeologia Hungarica (hereafter AH), n.s. XXXIV (1955), 224-6, fig. 61; T. Horvath, ‘Die avarischen Gräbenfelder von Ullö and Kiskövös’, AH, XIX (1935). 33. pl.xxii, 7.

53. i, 2.3; 2.5.

54. i, 2.9.

55. i, 2.8.

56. The evidence suggests that the Avars were mainly responsible for the introduction of lamellar horse armour. In this context, the equestrian relief of Chosroes at Taq-i-Bostan is of some importance, since this is the only representation for this period of such armour for horses. Possibly it was modelled on Avar-Roman fashion passed on during Maurice’s reign, perhaps through the guardsmen sent to Chosroes in 592. See Theophyl. Simocatta, Historia (CSHB), v, 3; 5; 11. At a later point in the treatise, Maurice refers to horse armour as being of either metal or felt (xi, 2. 7), a statement which is supported by the archaeological material. See Russell-Robinson, Oriental Armour, figs. 26, 27, 65, 81; and Ghirshman, Iran; Parthians and Sassanians, fig. 446.

57. On the arrival of stirrups in the west, see White, L., jnr., Medieval Technology and Social Change (Oxford, 1962)Google Scholar, where their early appearance is denied; and the critique of this by P. Sawyer, in R. H. Hilton and P. Sawyer, ‘Technical Determinism : the Stirrup and the Plough’, Past and Present, XXIV (April 1963), esp. 90-5. For the latest survey of the evidence, see Biyar, op. cit., 286-7. Stirrups were certainly an improvement, but they were not vital for shock combat. Lack of stirrups meant that, in order to brace himself for the shock, the lancer had to bend forward and grasp his spear with both hands, thus reducing control over his mount. The majority of illustrations show riders without stirrups in this posture. See Aschik, ‘Chambre sépulcrale’, fig. v; Stephani, Supplement, pl. x; Laufer, Chinese Clay Figures, fig. 35; N. Mavrodinov, ‘Le trésor protobulgare de Nagyszentmiklós’, AH, XXIX (1943), 115; Ghirshman, op. cit., figs. 69, 122, 166, 233; Schmidt, Persepolis, III, pls. 89, 91, 95. With stirrups the rider could engage more effectively in close combat, and at greater speed, without fear of losing his seat through sudden sideways movements. His ability to use a bow was neither improved nor impaired. See Vigneron, P., Le Cheval dans l’Antiquité Gréco-Romaine (des Guerres Médiques aux Grandes Invasions). Contribution a l’Histoire des Techniques, I (Nancy, 1968), pp.255 Google Scholar f.

As a result of the use of stirrups, a new form of saddle appears to have been introduced, and it seems likely that the Byzantines adopted the Avar saddle along with the stirrup. See Gy. Laszlo, ‘Der Grabfund von Koroncó und der altungarische Sattel’, AH, XXVII (1943), 159-70, pls. xiii-xix; idem, ‘Etudes archéologiques’, 279-81, figs. 83-5. See Maurice, Strat., i, 2. 7.

58. See, for example, Strat., 1, 2. 3; 2. 7; etc.

59. ii, 5. 5; iii, 5. 14.

60. See above, note 17; wood was clearly a practical form of defence.

61. The heavy infantry spear could be either the same type as that used by the cavalry or, for particular types of terrain, a shorter version. See Strat., xii, 8. 20-1.

62. For this equipment, see Strat., xii, 8.4.

63. Possibly the crossbow. See J. F. Haldon, : the Byzantine Crossbow?’, University of Birmingham Historical Journal, XII, 3(1971), 155-7. There are a number of problems surrounding this interpretation, however, and further research is needed before a satisfactory answer can be given.

64. xii, 8.5.

65. i, 2. 2; 2. 17. In the anonymous treatise, zaba is used in its strict sense, to mean a cloth or leather coat, worn when mail was not available. The use of the word in the later tenth century supports this (below, note 126). Maurice and other sixth- or seventh-century sources who equate it with lorikion, i.e. mail, are using it in a more general sense. The origins of the word are obscure. It appears, as far as I am aware, in only four other sixth- or seventh-century contexts: see Chronicon Panchale (CSHB), p. 625 (A.D. 532) and p. 719 (A.D. 626); Vie de Théodore de Sykéon (ed. and trans. A.-J. Festugière, Bruxelles, 1970), 28, 3; Justinian, Nov. 85, 4 (edd. Schoell and Kroll).

66. i, 2. 3; ii, 5. 3-5, where a number of field-army units are referred to.

67. E.g. i, 2. 6; xii, 8. 4.

68. Iron head-pieces are specified at Strat., i, 2. 6.

69. See Belenitsky, Central Asia, pl. 74.

70. Belenitsky, op. cit., p. 144.

71. Thordemann, ‘Splint Armour’, 126-7.

72. Arendt, ‘Waffenfund’, fig. 7.

73. Above, n. 12.

74. See Arendt, op. cit., figs, 1b, 3b; and Stein, A., Serindia: Report of Explorations in Central Asia and Westernmost China (Oxford, 1921), V, pls. lxxii, lxxxiii Google Scholar.

75. Le Coq, Bilderatlas, fig. 74.

76. See the Europos, Dura Final Report, VIII, i, pls. liv, lv Google Scholar.

77. For general references to the armour of soldiers in the eighth and early ninth centuries, see for example Theophanes, Chronographia (de Boor), p. 387, 15; Scriptor Incertus de Leone Bardae Filio (in Leo Gram. (CSHB), pp. 335-62), p. 339. 7-8. Theme soldiers were expected to have only the minimum. Cf. Ecloga (in Zepos, JGR, II), xvi, 2.

78. This ivory is now in the Louvre. See Noëttes, Lefèbvre de, L’Attalage: le cheval de selle à travers les âges (Paris, 1931), fig. 346 Google Scholar.

79. Several other ivories which Lefèbvre de Noëttes dated to the seventh or eighth century are later in date. See Goldschmidt, A. and Weitzmann, K., Die Byzantinïschen Elfenbeinskulpturen des X-XIII. Jahrhunderts, I: Kãsten (Berlin, 1930)Google Scholar, no. 32. Cf. Lefèbvre de Noëttes, op. cit., figs. 280, 344.

80. Despite the archaising tendencies of MS. illuminations or ivories, a few details of contemporary significance are often included.

81. In Dufrenne, S., L’Illustration des psautiers grecs du moyen âge (Paris, 1966)Google Scholar. From the Bristol psalter folios 10r, 86r, 89r, 93r, 180v, 231v; from the Paris psalter (MS. grec 20, Bibliothèque Nationale), folios 17V, 18r; from the Mt. Athos psalter (MS. Pantocrator 61, Bibl. Nat.), folios 11v, 30v, 68v, 89r, 109gr, 196r, 197v.

82. Cf. the Firuzabad reliefs referred to above, n. 43.

83. See De Cerimoniis (CSHB), pp. 500, 5-12; 505, 14-18; 506, 12; 14-15.

84. For illustrations, see nn. 80, 81.

85. Belenitsky, Central Asia, pl. 110.

86. This banded armour seems to be a form of lamellar in which the rows of lamellae are lacquered in different colours. For further examples see Stein, op. cit., V, pl. lxxiii, a Buddhist painting on silk of the ninth or tenth century from Tun Huang, Turkestan; note that in this painting, ordinary lamellar is shown quite clearly on other figures.

87. See Le Coq, op. cit., figs. 32, 33. The tubular lower arm-guards worn by one of the Kizil figures and shown on the Mug shield are the earliest examples of a fashion which was later to become general in Persia and the Middle East. See Russell-Robinson, Oriental Armour, pp. 26, 28ff.; Stone, Glossary, pp. 107-9, fig. 140. Cf. other frescoes from Pendjhikent dating to the eighth century, showing nobles with tubular or mail arm- and wrist-guards. Some also seem to have mail shoulder-guards. See Belenitsky, Central Asia, pl. 137. Combination armours became a feature of Persian and middle-eastern armour, and were also common in China. See Laufer, Chinese Clay Figures, pp. 275ff. For the early stages of this development, see Stein, op. cit., V, pls. lxiii, lxxxiv, lxxxvii/006, xc, xcii/0026, etc. For Iranian and later Turkish examples, see Russell-Robinson, op. cit., pp. 33ff. The origins of the style lie in the wearing of a separate jacket of leather or a cuirass of lamellar above the mail shirt, a tradition which prevailed in Byzantium as the klibanion.

88. See the soldiers on the Nagyszentmiklós jug, in Mavrodinov, N., ‘Le trésor protobulgare de Nagyszentmiklós’, AH, XXIX (1943), 1201 Google Scholar, fig. 77.

89. See Ghirshman, , Iran: Parthians and Sassanians, fig. 433; Belenitsky, op. cit., pl. 137 Google Scholar.

90. This may be the result of Byzantine influence; for whereas the pendant aventail, attached to the helmet, seems to be an eastern (Persian) development, the hood seems to be a western (Byzantine) feature, passed on to the peoples of western Europe.

91. See Mavrodinov, , ‘Le trésor protobulgare’, pls. iii, iv Google Scholar; and Aschik, , ‘Chambre sépulcrale’, pl. iv Google Scholar.

92. For descriptions of the arms and equipment of western armies, see Maurice, , Strat., xi, 3 Google Scholar, especially para. 3. 2. The same passage, with additions, is quoted by Leo in Tact., xviii, 80f. For further comment, see Oman, The Art of War, I, pp. 76ff.; 126ff.; II, pp. 3-5; Lot, L’Art militaire, I, 103ff.; Hewitt, J., Ancient Armour and Weapons in Europe (London, 1855), I, pp. 93ff.; 129ff Google Scholar.; Smail, R. C., Crusading Warfare (1097-1193) (Cambridge, 1956), pp. 112ff Google Scholar. See also a number of MS. illuminations illustrated in Barber, R., The Knight and Chivalry (London, 1970)Google Scholar, figs. 1, 2, 8. For the kite-shaped shield, see Hewitt, op. cit., pp. 143ff. For examples of the heaviest armour worn by western soldiers in the eleventh century, see the Bayeux tapestry.

93. Leonis VI Sapientis Problemata, ed. A. Dain (Paris, 1935); Nicephori Praecepta Militaria, ed. J. Kulakovskij, in Zapiski Imperatorskoj Akademii Nauk, VIII ser. (classe phil.-hist.), VIII (1908), 9. There are also a number of lists of military equipment in De. Cer., pp. 669ff. These include some of the items referred to below, but generally deal with more specialized weapons which I do not discuss here, such as boarding pikes.

94. Syll, Tact., 39, 2.

95. Ibid., loc. cit.

96. Leo, Tact., vi, 2; Syll. Tact., loc. cit.

97. Leo. Tact., vi, 2; Syll. Tact., 38, 5; 39, 2.

98. See De Cer., pp. 500, 8; 505, 11, 16-17; and Niceph., Praecepta, pp. 1. 24; 2. 11; 11. 32.

99. See Niceph, ., Praecepta, p. 11.33 Google Scholar;p. 12. 11; and the previous references to Leo and the Syll. Tact. The origins of the sabre in Byzantium are obscure, but we must in all probability attribute it to the appearance of Chazar and other Turkic soldiers in the imperial service during the second half of the ninth century. See Genesius (CSHB), p. 89, 9; Vita S. Euthymii (ed. P. Karlin-Hayter, B, xxv-xxvii (1955-7), pp. 8-152), p. 10; Theoph. cont. (CSHB), p. 358; Georg. Mon. cont. [CSHB), p. 853; Leo Gram. (CSHB), p. 267. On the long Avar sabre, see Fettich, N., ‘Das Kunstgewerke der Avarenzeit in Ungarn’, AH, I (1926), 14 Google Scholar, fig. 12; Marosi, A., Fettich, N., ‘Trouvailles avares de Dunapentele’, AH, XVIII (1936), 11 Google Scholar, pl. i; 49, 55-6; Gy. Laszlo, ‘Études archéologiques’, 228-9, 232-3, pls. xlvi, li—liii. From the later seventh century the shorter curved Turkish sabre developed. See Zakharov, A., Arendt, W., ‘Studia Levedica’, AH, XVI (1935), pt. 2 Google Scholar: ‘Turkische Sãbel aus den viii-ix Jahrhunderten’, and pls. iii, vi, viii; N. Fettich, ‘Die Metallkunst der landnehmenden Ungarn’, AH, XXI (1936), pls. xv, xxxi, xli/2, lxxii, xci/55; L. Kiss, ‘Der altungarische Grabfund von Geszteréd’, AH, XXIV (1938), pls. x, xi. For the waist-harness, see Gy. Laszló, ‘Etudes archéologiques’, figs. 47, 60, 61, 64, 79, 80, 83. Cf. the Perm plate and the Pendjhikent frescoes in Belenitsky, Central Asia, figs. 74, 137. For the two-edged sword see A. Bruhn-Hoffmeyer, Middelalderens Tvaeggede Svaerd (Copenhagen, 1954), I, pp. 200-4. He wrongly considers the spathion to have been a short sword. Although longer swords were known to the Byzantines, used by the ‘Franks’ (see Leo, Tact., xviii, 82), the gladius or short thrusting sword adopted by the Romans from the Iberian Celts had been abandoned by the end of the third century A.D. Vegetius describes two types known to him : the gladius major or spatha (the spathion herouliskion of Maurice, Strat., xii, 8. 4), and the gladius minor or semispatha (Veg., ii, 15). The latter does not appear in any of the later sources that I have examined, but see Justinian, Nov. 85, 4—. Kulakovskij considered that the sabre may have been introduced at a much earlier date, but the evidence he cites does not warrant this conclusion. See Niceph., Praecepta, pp. 42-4.

100. Tact., v. 2.

101. Tact., v, 2; vi, 1.

102. Tact., vi, 25.

103. E.g. Strat., i, 1. 5; i, 2. 2;xi, 2. 6; xii, 8. 4; etc.; Leo, Tact., v, 2; vi, 2; 3; etc.

104. Syll. Tact., 38, 3; 6; 39, 1; 8.

105. Leo, Tact., v, 2; vi, 2; Syll. Tact., 38, 3; 39, 1;8.

106. Syll. Tact., 38, 6; 39, 8; Leo, Tact., vi, 7; Problemata, xi, 35; 40; De Cer., 669, 20.

107. Leo, Tact., vi, 26.

108. Veg., II, 15: ‘hastili pedum trium semis, quod tunc vericulum, nunc verutum dicitur’.

109. The mattiobarbulus was a lead-weighted javelin. See Veg., i. 17: ‘Plumbatarum quoque exercitatio, quos mattiobarbulos vocant’; iii, 14: ‘Quartus idem ordo construitur de his, qui alacriter verutis vel mattiobarbulis, quas plumbatas nominant, dimicant’. See ii, 15, for these weapons in action.

110. See John Malalas (CSHB), p. 163, 3; Leo, Tact., vi, 26; 32; Problem., xii, 4-5; 7; 21; 40 (verutum); Tact., vii, 3; Problem., xii, 4–6; 21 (matzubarboulon); Dain, Naumachica (Paris, 1943), i, 14-16; 70 (excerpts from Leo’s naval section) ;Syll. Tact., 38, 3; 47, 16; 22-4; Niceph., Praecepta, p. 3. 15f.;p.4. 11f. (menaulion). The latter were to be made of cornel wood, mature oak, or ‘artzikidion’, rather than wood which splinters easily. See Syll. Tact., 38, 3; Niceph., Praecepta, p. 4. 11f. Cf. also De Cer., 669, 20; Theoph. (de Boor), 221, 3; De Administrando Imperio [DAI] (Jenkins/Moravcsik), I, 116, 9.

111. Niceph., Praecepta, p. 3. 15f.; 26f. There seems no reason to doubt that Leo is copying Maurice where the martzubarboulon is concerned. Cf. Maurice, Strat., xii, 8.2; 4; 5; Leo, Tact., vii, 3; Problem., xii, 6:7. It was certainly a javelin and not, as Aussaresses considered, a mace: op. cit., p. 53. He presumably based his translation on the fact that where Maurice has martzubarboulon listed among the extra material carried with each unit, Leo writes of axes (tzikouria) and maces (bardoukia, matzoukia). See Strat., xii, 8. 6; Tact., vi, 27). Even so, it is quite clear from other contexts that the weapon was not a mace. See, for example, Strat., xii, 8. 2 (Tact., vii, 3); xii, 8. 4 (Problem., xii, 6); xii, 8. 5 (Problem., xii, 7). An additional confusing factor is that martzubarboula are kept in leather cases in the same way as maces: see Strat., xii, 8. 5; Tact., vi, 11; 26; etc. This seems in fact to be a reference to a leather sheath or case in which a group of javelins were held, in much the same way as the heavy javelins of Vegetius’s time were kept in the hollow of the shield. See Veg., i, 17.

112. Tact., v, 2; vi, 25-6; xix, 57 (Naumachica, I, 65).

113. Syll. Tact., 38, 1; Niceph., Praecepta, p. 2. 3-4; 10.

114. Syll. Tact., 39, 1; 8; Niceph., Praecepta, p. 13. 24-6. Another word which appears for shield is (De Cer., pp. 579, 2; 5; 670, 7; DAI, I, 26, 31; 51, 82-4), used of the shields carried by sailors and marines, made of hide. See Reiske in De Cer., II, p. 682. would appear to be the same as dorkai (De Cer., p. 669, 19); while are hide/wood shields with metal plates stitched on, the polished circular iron shields of Leo, Tact., v, 3. See De Cer., p. 669, 19; DAI, II, p. 201.

115. Leo, Tact., v, 3.

116. Tact., v, 3; vi, 2.

117. The lorikion reaches to the knees for the infantry (Syll. Tact., 38, 4) and to the ankles for the cavalry (Tact., vi, 2; Syll. Tact., 39, 1). For and see De Cer., pp. 669, 16-17; 670, 9-10. For the term lorikion as equivalent to mail, see Syll. Tact., 38, 7; 39, 1; 6; Niceph., Praecepta, p. 12. 31-2. The term lorica meant a mail shirt in the west also. The assize of arms of Henry II in 1181 orders that knights should be equipped with loricas, cassidas, and carry clypeos. See Stubbs, W., Select Charters (Oxford, 1870-1960), p. 183 Google Scholar.

118. Tact., v, 3; vi, 4; Syll. Tact., 38, 4; 7; 39, 1.

119. See Syll. Tact., 30, 2; 31, 1; Tact., vi, 30; 34, where both authors paraphrase the treatises of Aelian and Arrian, but interpret their sources by making additions to the text, such as the klibanion, which they describe as of either iron or horn, Cf. Aelian, Takt. Theor., II, 7; 11 (in Kõchly and Rüstow, op. cit., II, i, pp. 236-469); Arrian, Tactica, 3, 2; 4, 1 (in Arriani Nicomediensis Scripta Minora, ed. R. Hercher [Leipzig, 1885], pp. 104-39). In the Sylloge (39, 6), horse armour is constructed either of lorikia, or of klibania of iron or horn, . See also Niceph., Praecepta, p. 11. 20f. The word derives from the Latin clibanarius, itself of Persian origin, and refers to heavy cavalry armour. See Bivar, op. cit., 277, n. 28. Reiske, basing his argument on a false etymology supplied by John Lydus, suggested that it came from chalybinon, i.e. steel. See De Cer., II, p. 583; J. Lydus, De Mag. (CSHB), p. 158.

120. Dain, Naumachica, I, 14.

121. In Goldschmidt and Weitzmann, op. cit., no. 122; for other tenth-century examples, ibid., nos. 1a, 4, 8d, 10a-d, 11, 12a-b.

122. Diehl, Ch., Peinture Byzantine (Paris, 1933), pl. lxxxiii Google Scholar; Nersessian, S. Der, Aght’amar, Church of the Holy Cross (Cambridge, Mass., 1965), pls. 23, 27 Google Scholar; for the frescoes, see Restié, , Byzantine Wall Paintings in Asia Minor (1967), pls. 21, 28, 31 Google Scholar (late eleventh century); 183 (c. 1190-1200); 230, 237 (c. 1200–10); 246, 247 (c. 1070); 288 (c. 1040-50). A strikingly similar armour can be seen in Buddhist paintings of the ninth or tenth century at Tun-Huang in Chinese Turkestan. See Stein, op. cit., V, pl. lxxiii; the beast-headed demon carrying the banner of the Vaisravana wears a lamellar coat constructed very similarly to that shown on Goliath at Aght’amar. Members of the imperial household had gold armour. See De Cer., pp. 500, 5-12; 505, 14-18; 506, 12; 14-15.m Other klibania were divided into ‘ordinary’ and ‘better’, see DAI, 51, 83-4.

123. Leo, Tact., v, 3; Syll. Tact., 39, 1; Niceph., Praecepta., p. 111 of.;De Cer., p. 670, 3.

124. Leo, Tact., v, 3; xix, 3; Syll. Tact., 38, 4; 7; Niceph., Praecepta., p. 1. 16-17; p.12, 5f. Cf. De Cer., pp. 749, 772. Both versions of the padded coat were popular in the east. It is quite likely that the idea was adopted as a result of Arab or Persian influence in the first place. It was passed on in the twelfth century to the west under the name of wambais or gambeson. Saladin is reported to have given such a garment to Richard I : ‘unum alcottonem (Fr. hacqueton; Arab, al-qutun, cotton) satis levern, nullo spiculo penetrabilem’ (Oman, op. cit., II, pp. 3f.) See also Chronicon Colmariense (in Monumenta Germaniae Historica, XVII, ed. Pertz [Hanover, 1861]), s.a. 1298, p. 264.

125. Although soldiers wearing knee-length, long-sleeved coats of lamellar are illustrated in an eleventh-century MS. See Diehl, , Peinture Byzantine, pl. lxxxii Google Scholar. Stein, Cf., Serindia, V, pls. lxxv/004 Google Scholar; c; and a stucco figure from a Buddhist shrine at Kara Shahr (Chinese Turkestan) dated c. 750-850, ibid., pl. cxxxv. Compare also Aschik, Chambre Sépulcrale, pl. iv.

126. See Leo, Tact., vi, 25; Syll. Tact., 39, 2; Niceph., Praecepta, p. 1. 20f.; p. 11. 7f.; De Cer., pp. 669, 18; 672, 5-6; 674, 5. For splint arm-guards, see G. Arwidsson, ‘Armour of the Vendel Period’, figs. 8, 12; and the Nagyszentmiklós figures. Those made of iron may have been of the tubular type referred to above (n. 87) which developed in Persia and central Asia from the seventh and eighth centuries. Leo and the Sylloge use the term for leg-defences, which suggests that these also were of metal splints, although the word is a classical term for moulded greaves: Tact., vi, 25; Syll. Tact., 38, 5. For lamellar arm-defences in the steppe areas, see Arendt, op. cit., figs, 1b, 3b; Stein, op. cit., V, pls. lxxii, lxxxiii, lxxxiv, lxxxv, lxxxvii/005. The upper sleeves attached to klibania appear to be of lengthened lamellae or splints, fixed to the main cuirass by thongs or a common backing—see Diehl, Peinture Byzantine, pl. lxxxiii. In the Praecepta the author states that the padded hangings attached to the klibania and the sleeves should have zabai affixed. Here the word no longer means a coat of armour, but the material itself—the face-guards of cloth attached to the helmets are also referred to as zabai—and is used in what would appear to be its original sense. In the sixth-century anonymous treatise, zaba is distinguished as a padded cloth coat; although in the treatises of Maurice and Leo and in other sixth-century references, it is used as a general term or the equivalent of lorikion. Cf. Niceph., Praecepta, p. 11. 13f.; anon., xvi, 9; Vie de Théodore de Sykéon (ed. Festugière), 28, 3: Justinian, Nov. 85, 4: See also n. 65 above.

127. Aventails are called . If of mail, they are qualified as If these were not available, then iron gorgets, lined with wool, were worn instead. See Leo, Tact., v, 3. The latter reference occurs also in Maurice, but the former, to mail, does not; and it seems likely that while the mail aventails were a contemporary usage, the iron gorget is merely copied from Maurice. Such gorgets are not referred to in other tenth-century sources. Helmets with aventails are referred to as in Syll. Tact., 38, 5; 39, 3; and as in the Praecepta, p. 11, 12f.

128. The Sylloge refers to the first type as (38, 7; 39, 9), and to the second as (39, 8). Other terms for helmets are (De Cer., 669, 18) and (Anna Comnena, Alexiad, ed. Leib, I, 5. 7). Reiske (De Cer., III, 790) considers that these are two different types of helmet with face-guards, following Du Cange (Alexiad [CSHB], II, 423-4), but there is no reason to doubt that the two expressions refer to the same thing—a helmet with a mail or lamellar aventail covering face and neck, which could be drawn up to the crown when not in battle.

129. Niceph., Praecepta, p. 1. 23-4; De Cer., 353, 16; 670, 3; anon., De Obsidione Toleranda, ed. H. Van Den Berg (Leiden, 1947), p. 48. The term was applied to a closed, spherical cap of felt, and referred also to a type of royal headgear, a closed diadem. On its origins and development as such, see DAI, II, 652-5; A. Papadopoulos, EEBS, V(1928), 293-9. It seems also to have served as a sort of standard for signalling purposes. See Leo, Tact., xix, 42.

130. Leo, Tact., v, 3; vi, 8; Syll. Tact., 39, 6; Niceph., Praecepta, p. 11. 16-22. The Sylloge (loc. cit.) adds that the horse armour could also be of mail, although this must have been unusual. Cf. De Cer., pp. 81-2, for armoured and caparisoned horses on ceremonial occasions.

131. For the staff-sling, see Veg., i, 16; iii, 14; 24; for the sling in the tenth-century treatises, see Leo, Tact., v, 3; vi, 25; 26; Problem., xii, 5; 6; 7; etc; Syll. Tact., 38, 10; Niceph., Praecepta, p. 2. 2-3.

132. Leo, Tact., vi, 27; xiv, 84 Syll. Tact., 39, 3 cf. Dain, Naumachica, I, 7 (Leo, Tact., xix, 7); VI, 6 De Cer., p. 670, 16-17 pp. 671, 5; 673, 2 Theoph. cont., p. 232, 1-2 Niceph., Praecepta, p. 2. 1; p. 11. 30f.

133. Leo, Tact., vi, 5 (Maurice, Strat., i, 2. 4); Syll. Tact., 39, 4; Niceph., Praecepta, p. 12. 4f. For the ‘curved bows’ see the fragment of Mutanabbi in Vasiliev, A. A., Byzance et les Arabes (Brussels, 1950), II, ii, p. 344 Google Scholar, v. 36.

134. Illustrated in Lefèbvre de Noettes, op. cit., fig. 347. For other examples of long mail coats, see the references in note 81 above, illuminations of the later ninth and tenth centuries.

135. See notes 123, 125.

136. See Estopañan, C., Skyllitzes Matritensis, I (Barcelona/Madrid, 1965)Google Scholar, figs. 11-13, 33-34, etc. Compare, for example, the miniatures in the Psalter of Theodore of Caesarea (now in the British Museum), illustrated in Oman, Art of War, I, pls. v, vi, dated to the middle of the eleventh century.

137. See Vasiliev, op. cit., 2, ii, p. 333, vv. 16–17, P. 347, v.44, and p. 420.

138. On the organization of the late Roman field and frontier armies, see Jones, A. H. M., The Later Roman Empire, 284-602 (Oxford, 1964), pp. 654ff Google Scholar.; Stein, E., Studien zur Geschichte des Byzantinischen Reiches (Stuttgart, 1919), pp. 117ff Google Scholar.; Hannestad, K., ‘Les forces militaires d’après la guerre Gothique de Procope’, Classica et Medievalia, XXI (1961), 13683 Google Scholar; Maspéro, J., et dans l’armée byzantine’, BZ, XXI (1912), 97109 Google Scholar; Teale, J. L., ‘The Barbarians in Justinian’s Armies’, Speculum, XL (1965), 294322 Google Scholar.

139. On the gradual collapse of the sixth-century administration and the development of the early tnemata see, inter al.: Karayannopoulos, J., Die Entstehung der byzantinischen Themenordnung (Byzantinisches Archiv, Heft 10, Munich, 1959)Google Scholar; idem, ‘Über die vermeintliche Reformtätigkeit des Kaiser Herakleios’, JÕBG, x (1961), 93-7; Pertusi, A., ‘La formation des Thèmes byzantines’, Berichte zum XI Internationalen Byzantinisten Kongress (Munich, 1958), I, 140 Google Scholar. On the recruitment of the tagmata, see Ahrweiler, H., ‘Recherches sur l’administration de l’Empire byzantin aux IXe-XIe siècles’, Bulletin de Correspondance Hellénique, LXXXIV (1960), 24ff Google Scholar.; cf. Theoph., pp. 462, 463 for the recruitment of soldiers from the Anatolik Theme into the Scholai under Constantirte V. For a reassessment of the military value of the Themes, see Kaegi, W. E., ‘Some Reconsiderations on the Themes : seventh-ninth centuries’, JÖBG, XVI (1967), 3953 Google Scholar. For arms production, compare the system outlined in novel 85 of Justinian with that described at De Cer., p. 657, 12ff.

140. A recent assessment of this warfare has been made by Ahrweiler, H., ‘L’Asie Mineure et les invasions arabes (VIIe-IXe siècles)’, Revue Historique, CCVII (1962), i, 132 Google Scholar; for techniques of frontier warfare, see Leo, , Tact., xvii Google Scholar; xviii, 138ff.; Nicephori Phocae De Velitatione Bellica, in Leo Diac. (CSHB), pp. 179-258.

141. On the appearance of Turkic mercenaries in the Caliphate and later in Byzantium, see for example Theoph. cont., p. 126, 23f.; Genesius, p. 68; and references in note 100 above. For the increase in the number of mercenary troops hired for campaigns, see Theoph. cont., p. 81; p. 112; Georg. Mon. cont., p. 793; Symeon Mag., p. 624; p. 627. The revival of an offensive policy is described in G. Ostrogorsky, History of the Byzantine State (Oxford, 1968), pp. 210ff.; and note the request of Theophilos to the King of the Franks for military aid in 839 : Theoph. cont., p. 135, 1-6. For the reorganization of the high command and the effect on the theme forces, see Ahrweiler, ‘Recherches’, 46ff.; 56ff.; N. Oikonomides, Les Listes de Préséance Byzantines des 1Xe-Xe Siècles (Paris, 1972), pp. 329, 333ff. On the increase in the number of tagmata, see Ahrweiler, ‘Recherches’, pp. 27ff.; and Oikonomides, op. cit., pp. 327-8, 330, 332-3; and for Nikephoros II’s reforms, see Ahrweiler, ‘Recherches’, pp. 16ff.

142. For the new heavy cavalry and army in action see, for example, the accounts of the battles of Dristra, in Leo Diac, pp. 140-1; 153-6, etc. For archers, see the Arab references in Vasiliev, op. cit., 2, ii, pp. 339, 344, 385. Note that the Arabs (excluding their Turkish mercenaries) are not archers: ibid., p. 336. Nikephoros’ attitude to the west in Liutprandi Legatio ad Nicephorum Phocam (in Liutprandi Episcopi Cremonensis Opera, ed. Joseph Becker, Hanover/Leipzig, 1915), xi. For the revival of interest in military theory, see Dain, A., ‘Les stratégistes byzantins’, Travaux et Mémoires, II (1967), p. 353 Google Scholar, and on the comparative merits of the treatises examined, ibid., pp. 343-6, 354-8, 370-1.

143. Cf. Basil I’s reorganization in Theoph. cont., pp. 265, 3-16; 266, 6-9.

144. For detailed comparison of, for example, types of arm-guards, where the Byzantines followed steppe fashion to a large extent, see above, note 127.

145. For the defeat of 934, see Macoudi, , Les Prairies d’Or, II Google Scholar, ed. de Maynard, Pavetde Courtelle, C. Barbier (Paris, 1863), pp. 623 Google Scholar. See also Kaegi, W. E., ‘The Contribution of Archery to the Conquest of Anatolia’, Speculum, XXXIX (1964), 96108 Google Scholar; Ostrogorsky, op. cit., pp. 320ff.; Ahrweiler, ‘Recherches’, 46.

146. For the latest examination of this topic and on the question of the political activity of the Byzantine army, see Walter E. Kaegi, ‘Patterns of Political Activity of the Armies of the Byzantine Empire’, in On Military Intervention, ed. Morris Janowitz(1971), esp. pp. 22f.