zzz: For the most recent summary see S. Hornblower and A. Spawforth (editors) Oxford Classical Dictionary3 (Oxford and New York, 1996), p. 785. See also A. Kappelmacher, Iulius 243, Paulys Real-Encyclopädie der Classischen Altertumswissenschaft 10; W. Eck Iulius 3, Paulys Real-Encyclopädie der Classischen Altertumswissenschaft, Suppl. 14, (1974); PIR 2 IV, I 322; G. Bendz 'Frontin, Kriegslisten' Schriften und Quellen der alten Welt 10 (Berlin, 1963), pp. 1‑4; W. Eck 'Senatoren von Vespasian bis Hadrian' Vestigia 13 (1970), pp. 77‑93, 119‑125, and 131‑137; J. Ward Perkins 'The Career of Sex. Julius Frontinus' Classical Quarterly 21 (1937), pp. 102‑105; and W. McDermott 'Stemmata Quid Faciunt? The Descendants of Frontinus' Ancient Society 7 (1976), pp. 229‑261; B. Campbell 'Teach Yourself How to be a General' Journal of Roman Studies 77 (1987), pp. 13‑29; and W. Eck 'Die Gestalt Frontins in ihrer politischen und sozialen Umwelt' Wasserversorgung im Antiken Rom (München, 1989), pp. 47‑62.

zzz: The now fragmentary Art of Surveying, the lost De Re Militari, the Strategemata, and the De Aquis Urbis Romae.

zzz: What is more, an inductive historiographical method will also be used following that described by W. Lacey 'Agrippa's Provincia' Augustus and the Principate (Leeds, 1996), pp. 117‑131, at p. 117. Various hypotheses will be proposed for the career of Frontinus which will be tested against the context and the rest of Frontinus' career so far as it can be reconstructed. As Lacey states of his hypothesis 'certainty will not be achieved' but what should emerge is a plausible more complete picture of Frontinus' career.

zzz: However, I also hope that I may in some way be like the character of Hannah Jarvis; envision the problem and solve the mystery of the hermit of Sidley Park — so that in my own way I might add intelligently to solving the 'mystery' of the career and influence of Sextus Julius Frontinus.

zzz: A. Devine 'Aelian's Manual of Hellenistic Military Tactics: a new translation from the Greek with an introduction' in The Ancient World, 19 1 and 2, pp. 31‑64. Aelian Tactica Preface 3.

zzz: As we shall see, Frontinus was held in high regard by a succession of emperors for his abilities not only as a soldier and commander but also as an administrator, diplomat, and capable overseer of mechanical and architectural construction.

zzz: The meeting between Aelian and Frontinus is probably dateable to A.D. 96. See below.

zzz: A birthdate of 35 would make Frontinus the same age as Nerva, a man with whom he later had a friendly and important relationship. They may have served on the same vigintivirate board and so perhaps established a relationship which seems to have come into play again in A.D. 96. For their friendship see below and De Aquis, Preface 1 and 64. However, Nerva was a favoured patrician (ILS 273), praetor in 66 (Tacitus Annales XV.72.1), and consul in 71 with Vespasian as cos. III. This was a great honour: R. Syme Tacitus (2 Volumes, Oxford, 1958), p. 653 and n. 5. Therefore, if Frontinus was also favoured, then, as will be argued below, the date of his birth may well be between A.D. 37 and 39 for his appointments to praetor and consul to be at the same age as Nerva. There is, however, no need to make Frontinus the same as Nerva. J. Wight Duff, A Literary History of Rome in the Silver Age2 (London, 1960), p. 338 argues that Frontinus' birth may be guessed to be around 40.

zzz: R. Syme Review of Reidinger Die Statthalter Pannoniens, Gnomon 29 (1957), pp. 518‑519. For Q. Valerius Lupercus Julius Frontinus of Vienna (Vienne) see CIL XII 1859‑1860: Vienna. Anthony R. Birley The Fasti of Roman Britain (Oxford, 1981), p. 70 and n. 4, argues that Sexti Julii are relatively rare but found more frequently in Gaul particularly in Narbonensis. There are nearly 30 Sex. Julii in CIL XII, and nine Frontini. See also Syme Tacitus, p. 790. Birley, Fasti, p. 69, argues that the references in Aelian and Martial (see below) to him residing at Formiae and Anxur/Tarracina respectively, are no reason to assume he was from Latium. L. Schumacher Prosopographische Untersuchungen zur Besetzung der vier hohen römischen Priesterkollegien im Zeitalter der Antonine und der Severer (96‑235 n. Chr.) (Dissertation, Mainz, 1973), pp. 254‑255, cited in Birley Fasti, p. 69 n. 3, regarded it as more likely that Frontinus came from Italy — 'vielleicht aus Latium (Tarracina?).' 'Perhaps from Latium'.

zzz: R. Syme 'P. Calvisius Ruso: One Person or Two?' Zeitschrift für Papyrologie und Epigraphik 56 (1984), pp. 173‑192, at p. 191. On Narbonensis see also R. Syme 'More Narbonensian Senators' Zeitschrift für Papyrologie und Epigraphik 65 (1986), pp. 1‑24, at p. 8.

zzz: The office of praetor urbanus in 70 shows Frontinus' military focused career since it allowed him immediately to become a legatus legionis. See below. Syme Tacitus, p. 790, and Birley Fasti, p. 70, argue that Frontinus was an equestrian, which would magnify the problems of his age and earlier career. See below. Ideally to be praetor at thirty would mean Frontinus was born in 40. If he was not praetor until he was 35 we might need to find reasons for the delay. Perhaps he was engaged in other activities. See below.

zzz: E. Birley 'Promotions and Transfers in the Roman Army: Senatorial and Equestrian Officers' The Roman Army Papers 1929‑1986 (Amsterdam, 1988), p. 95.

zzz: Brian W. Jones The Emperor Domitian (London and New York, 1992), p. 15. See also Brian W. Jones The Emperor Titus (New York, 1984), p. 208.

zzz: Jones Domitian p. 15. The problem was a delicate one, according to Jones, because the governor of Moesia, Aponius Saturninus, had tried to kill Julianus (Tacitus Historiae II.89) and he could not be offended either. In 70 the loyalty of the Danubian legions was vital. Jones argues, without naming the man, that Julianus was the brother-in‑law of Vespasian's powerful freedman finance minister. This was probably Tiberius Julius Augusti libertus, and such men had best not be offended. See P. Weaver Familia Caesaris (Cambridge, 1972), pp. 31‑32, and P. Southern Domitian, Tragic Tyrant (London and New York, 1997), pp. 52‑53.

zzz: Frontinus Strategemata IV.3.14. For discussion on the authenticity of Book IV see below.

zzz: Kappelmacher RE Iulius 243, col. 591. Frontinus was a very new ex-praetor and he is far more likely to have been the commanding officer of a legion than higher in the command structure. It seems more likely that it was as a legatus legionis that this surrender was made to him. Towards the end of the first century the legatus legionis was usually chosen from the ranks of ex-praetors, although he could be chosen from ex-quaestors like Tettius Julianus.

zzz: Ward Perkins 'Sex. Julius Frontinus,' p. 102. Ward Perkins argues that the stratagem suggests command of a legion.

zzz: J. A. Crook Consilium Principis (Cambridge, 1955), p. 168. Crook gives no reasons for this conclusion. However, there may be a case for his assumption. See below.

zzz: Tacitus Historiae IV.86. Jones Domitian, p. 16. See Map 1 for the relative locations of the Lingones, Rigodulum, Lyons, and Oppenheim in Cerialis' campaign against Civilis.

zzz: The favour shown to Frontinus, as far as is discernible, seems to have come about because of his loyalty and ability.

zzz: Brian W. Jones 'Titus and some Flavian Amici' Classical Philology 49 (1954), pp. 454‑462, at p. 460 n. 57.

zzz: Birley Fasti, p. 67. On p. 14 he argues that 'under the Julio-Claudians it was not uncommon for men to command legions before the praetorship, whether as ex-quaestors or as ex-tribunes or as ex-aediles. But this practice virtually died out from the beginning of the Flavian period.' This picture perfectly suits Frontinus' situation.

zzz: Birley 'The Epigraphy of the Roman Army' Roman Army, p. 6. For Frontinus, this selection probably took place under Claudius and continued under Nero. It could possibly have even occurred as late as the last year or so of Nero, ie. under Burrus and Seneca. Note that Frontinus commanded consular armies less than three years after commanding a legion as ex-praetor. This reinforces the likelihood that Frontinus had commanded legions before becoming praetor.

zzz: Birley Fasti, p. 70 n. 8 citing Corpus Agrimensorum Romanorum I.I, C. Thulin (editor) (Leipzig, 1913), pp. 9 and 44 on Spain, and pp. 45 and 48 on Africa.

zzz: A view of him as a plebeian nobilis senator would seem to solve the problem of the unlikelihood of Frontinus being patrician and avoid the complications and obstacles to him being an equestrian. The short gap between his praetorship and consulship almost certainly negates the possibility that Frontinus was a novus homo.

zzz: Birley Fasti, p. 46, argues that legati would normally be selected from men who could readily afford to devote twelve months to their legateship. Birley postulates, p. 68, that Cerialis may not have proceeded to Britain until spring 71 and this would have been near the end of a year legateship of Frontinus. However, he most likely served in Britain as a legatus legionis also. See below and Birley Fasti, pp. 70‑71.

zzz: Vespasian seems to have alienated Nero either by sleeping through, or leaving the room during, Nero's singing recitals: Suetonius Divus Vespasianus 4. Vespasian may have first seen the qualities of Frontinus at the court of Nero between snoozes. Frontinus may have shown some loyalty to Vespasian despite the latter's alienation from Nero. This would make the rapid advancement in the early years of Vespasian more understandable. If Frontinus participated in Vespasian's eastern campaign and showed the emperor-to-be his worth, Vespasian may have rewarded that worth as soon as he was able.

zzz: However, Corbulo only started with X in the first campaign of 58. V and XV came as reinforcements in 62 or later but all were present for the conclusion of the campaign. See Tacitus Annales XIII.40.3, XV.6.5, and XV.25.5. H. M. D. Parker The Roman Legions (Oxford, 1923), pp. 134‑137.

zzz: Parker Roman Legions, pp. 134‑142. Tacitus Historiae II.83. The movements of these legions to Egypt is important for Frontinus' possible connection to Heron of Alexandria. See below.

zzz: Among the factors which justify a closer examination of this conjecture are the connections to Titus and Heron of Alexandria. The other legati in the east also have a connection to Frontinus which seems to strengthen the possibility of his having experience there. See below.

zzz: An analysis of X Fretensis will suffice. Edward Dabrowa, Legio X Fretensis (Stuttgart, 1993), records the prosopographical evidence of only two legati between A.D. 14 and 67, Minucius Rufus (14‑19) and M. Ulpius Traianus (67‑69). There is ample space for Frontinus to have been legatus before 67. The prosopographical evidence of the other legions present a similar situation.

zzz: Syme Tacitus, p. 790. Syme has a problem in that the standard age for a consulship was 42, Tacitus, pp. 654‑655, and thus Frontinus, with less than three years between his praetorship and the consulship of A.D. 73, could only be 42 if born in 31. The standard ages for posts were: military tribunate at circa 20, quaestor at circa 25, and circa 30 for the praetorship. See Syme Tacitus, p. 655. He argues that the standard gap between praetor and consul was twelve years. However, the further back we place Frontinus' birth the more reasons we have to conjecture why he held the praetorship so late. Hence Syme's arguments that Frontinus was an equestrian who was possibly adlected into the senate under Galba. There could be the possibility of a second praetorship or see Frontinus as an ex-quaestor legatus. Nerva, not selected to be emperor in Nero's reign, was nonetheless favoured, and had only a five year gap between his praetorship of 66 and his consulship of 71.

zzz: Frontinus De Aquis Preface 1, and evidenced by the composition of the treatise itself.

zzz: See Syme Tacitus, pp. 86, 592 n. 4, 599, 605, and 616. Even so Frontinus seems to be the odd one out in this group. Syme Tacitus, p. 596, discusses the eclipse of the consular ambitions of Corellius, Verginius and Vestricius Spurinna under Domitian. However, Frontinus also was disappointed if he sought a consulship in the reign of Domitian. Arrius Antoninus' career was in decline under Domitian, as was Verginius Rufus'. However, Frontinus', Corelius' and Spurrina's careers under Domitian all involved Germany and two were also appointed to Asia. See Brian W. Jones Domitian and the Senatorial Order (Philadelphia, 1979), p. 98, 102, 108, and 120. Corellius was proconsul Asiae between 84/84 and 95/95. Eck 'Senatoren,' p. 83.

zzz: A. R. Birley 'The Roman Governors of Britain' Epigraphische Studien 4 (1967), pp. 63‑102, at p. 67. G. Houston, Roman Imperial Administrative Personnel during the Principates of Vespasian and Titus (A.D. 69‑81) (Chapel Hill, 1971), p. 134, argues that 'it seems better to assume with Syme that he simply reached the praetorship late.' Jones Senatorial Order, pp. 51‑52, includes Frontinus as one of Vespasian's patrician appointees to imperial consular provinces and one of nine or ten patricians to govern either Africa or Asia in Domitian's reign.

zzz: It is unlikely that Frontinus was a novus homo. Such a detail would presumably be told us by Frontinus himself if not by Tacitus. Syme Tacitus, p. 654, argues that a novus homo had to wait at least ten years after the praetorship before a consulship. Kappelmacher Iulius 243, col. 591, argues that 'Wenn er 70 Praetor war, so wird er als Nichtadeliger etwa um 30 geboren sein.' 'If he was praetor in 70, he would as a non-noble have been born around 30.' He may still have been a senator. However, McDermott 'Stemmata Quid Faciunt?,' p. 235, argues 'surely he was novus homo as were his son in law and his grandson.'

zzz: We have already seen Birley's assertion, Fasti, p. 70 n. 8, that passages in the corpus agrimensorum show his familiarity with Spain and Africa from appointments to those provinces in the 60s.

zzz: Strategemata IV.1.21. Cf Tacitus Annales XIII.36. A proof of Frontinus' involvement in the campaign could be his naming of the location ad castellum Initia which Tacitus omits. Here we have a concern with discipline which fits with chapter I of Book IV. Tacitus names the officer Paccius Orfitus who caused the squadrons to be routed because he disobeyed orders: paucae e proximis castellis turmae advenerant pugnamque imperitia poscebant 'A few squadrons had come from the neighbouring forts, and because of their inexperience they were demanding battle.' Tacitus also states that these squadrons, which he does not number, unlike Frontinus, were freed on the petition of the whole army. This differs with Frontinus' account.

zzz: MSS have mitterentur. Bennett emended emitteretur, but mitteretur seems a better emendation.

zzz: Strategemata IV.1.28. This incident is not recorded in Tacitus' account but seems to contradict Tacitus' statements at Annales XIII.35 that Corbulo showed no mercy, which Tacitus includes in his general summation of the restoration of discipline. sed qui signa reliquerat, statim capite poenas luebat. 'but the man who had left the standards made immediate atonement with his life.' See below.

zzz: Strategemata IV.2.3. The discipline of IV.1.21 and 28 reaches its culmination here. Whilst no such simple statement as this occurs in Tacitus we can infer from Tacitus' account, Annales XIII.8, and 35, that Corbulo was given initially two legions and an extra auxiliary foot and horse. However, after discharging men incapacitated by age and ill-health, Corbulo received reinforcements of one German legion and its auxiliaries. This does seem to fit with Frontinus' figure. See also Dio Cassius LXII.19.

zzz: Strategemata IV.7.2. See below for the possible significance of this stratagem. Tacitus does not record this. However, R. Davies, Service in the Roman Army (Edinburgh, 1989), p. 126, argues that this stratagem 'can only be a reference to the proved ability of the Roman army to build camps for itself.' Davies p. 126 n. 7 compares this stratagem to Annales XIII.35 where Tacitus describes Corbulo's redisciplining of his troops.

zzz: There could be a connection between this concern with discipline and Frontinus' attitude towards the Parthians. Strict discipline may have been the key to their defeat in his opinion. See below.

zzz: Tacitus Annales XXXV.35. Sed Corbuloni plus molis adversus ignaviam militum quam contra perfidam hostium erat: . . . 'Still, Corbulo's main difficulty was rather to counteract the lethargy of his troops than to thwart the perfidy of his enemies.' What follows is a commonplace of the discipline restoring general. Ipse cultu levi, capite intecto, in agmine, in laboribus frequens adesse, laudem strenuis, solacium invalidis, exemplum omnibus ostendere. 'Corbulo himself, lightly dressed and bare-headed, was continually among his troops, on the march or at their toils, offering his praise to the stalwart, his comfort to the weak, his example to all.'

zzz: Dicebat occurs at III.4.6, III.9.1, IV.1.17, IV.7.1, IV.7.2, and IV.7.3.

zzz: The Loeb note states that this was in A.D. 70. However, Frontinus only seems to have included contemporary stratagems which he himself was present for. See the Cerialis stratagem below. The only account Josephus gives of a Roman attack on the Sabbath is in relation to Pompey at Bellum Judaicum I.3‑5. He again refers to this in Bellum Judaicum II.16 during King Agrippa's plea to the Jews not to make war on the Romans. The results of Pompey's Sabbath attacks, the storming of the temple, would have given Vespasian a very beneficial historical precedent to follow.

zzz: Tacitus Historiae IV.71.4. See Syme Tacitus, pp. 173‑175. For a summary of the campaign see A. R. Birley 'Petillius Cerialis and the Conquest of Brigantia' Britannia 4 (1973), pp. 179‑190, at pp. 183‑185.

zzz: Ward Perkins 'Sex. Julius Frontinus,' pp. 102‑105. Birley Fasti, p. 70 and n. 10, argues that this suggestion is attractive but cannot be proved. Petersen considered that it was XXI Primigenia: PIR 2, Julius 322.

zzz: See Map 1. Ward Perkins 'Sex. Julius Frontinus,' p. 103. He argues that therefore Frontinus could not be legatus of XXI because the stratagem implies the presence of the army in the territory of the Lingones. This is not necessarily so; Frontinus may have been sent as a subordinate to receive the surrender and this certainly explains tradidit mihi, which is less comprehensible if Cerialis were also present. Ward Perkins' own chronological reconstruction rejects that the surrender of the Lingones took place at Trier because Cerialis was present, and therefore the surrender would not have been to a subordinate.

zzz: See Parker Roman Legions, pp. 143‑145. The arrival of this army caused I Germanica and XVI to swear an oath of allegiance to Vespasian.

zzz: R. Syme, Review of E. Stein Die Kaiserlichen Beamten und Truppenkörper im römischen Deutschland unter dem Prinzipat and E. Ritterling Fasti des römischen Deutschland unter dem Prinzipat, Journal of Roman Studies 23 (1933), pp. 94‑98, at p. 97.

zzz: Syme, Review of Stein and Ritterling, p. 97, commenting on Ritterling p. 56.

zzz: This date seems to have arisen before the secure dating of Frontinus' consulship. The relatively secure dating of that consulship to 73 rules out such a consular command before that date.

zzz: Instead this would therefore place Frontinus in the much more exclusive group of Julius Agricola, Turpilianus, and possibly Sisenna as the only governors of Britain to assume office immediately after their consulship. See below.

zzz: Tacitus Historiae IV.68. Ward Perkins 'Sex. Julius Frontinus,' pp. 103‑104. However, Ward Perkins' reconstruction, which has these two legions making up part of the belated reinforcement expedition of Domitian (Historiae IV.86) is wrong. They actually departed to be part of Cerialis' army at Historiae IV.68. On that original occasion Domitian seems to have wanted to go with them. See below.

zzz: Ward Perkins 'Sex. Julius Frontinus,' p. 104 n. 1. Tacitus Historiae IV.86.

zzz: However, Lyons is not as far north as the territory of the Lingones so, if Domitian only accompanied the army to Lyons, Frontinus' taking of the Lingones' city should have been under the command of Cerialis.

zzz: This may also reflect the date of composition of the Strategemata — after 84, during Domitian's reign and the comment may have been lip service to the emperor. See below.

zzz: Therefore Crook's view, Consilium Principis, p. 168, that Frontinus was a comes of Domitian may have vindication.

zzz: This would (perhaps) imply that the legateships of legions and other appointments were 'up for grabs.' A situation Domitian and Mucianus seem to have promoted if not created.

zzz: The likelihood of such an occurrence is said by Campbell 'Teach Yourself,' p. 14, to be low. This is strange when we consider that the stratagems in Strategemata were taken from history and Frontinus' own experience and were included because of their practicability. Frontinus also included them so that they could inspire and be copied by aspiring generals.

zzz: More probably Frontinus did know of the stratagem but possibly did not include it in his collection because he did not see it in practice. Such a conjecture has nothing to support it but if it were so we might be able to postulate that all of Frontinus' contemporary stratagems warranted inclusion because he had witnessed them. It is also interesting in that Frontinus does not include the Pericles stratagem either.

zzz: Strategemata IV.3.14. See below for the inscription evidence. As the account of the campaign is not full we do not know the complete movements during the campaign, but we can be sure Frontinus could have gone with his legion. See Birley 'Cerialis,' pp. 183‑185.

zzz: Great effort seems to have been taken in appointing the 'right man for the job' as governors of Britain. See below. Ward Perkins' whole purpose was to show that all three Flavianic governors of Britain had previous experience in Britain. This was to prove the Flavianic 'policy' of appointing men 'to posts for which their previous careers had especially fitted them,' p. 105. He gives the examples of Funisulanus Vettonianus' career in Dalmatia, Pannonia, and Moesia Superior; and Sex. Sentius Caecilianus in Africa, as well as Cerialis and Agricola in Britain, to show such a policy.

zzz: Kappelmacher Iulius 243, col. 592. Kappelmacher refutes the evidence that Frontinus' consulship came in 74, which is based on E. Hübner's restoration of CIL VI 2016 . . .ON. . . to read [sex.iulio.fr]ON[tino]. 'Die römischen Legaten von Britannien' Rheinisches Museum für Philologie 12 (1857), pp. 52‑83, at p. 54. As Birley, Fasti, p. 71 and n. 14, points out: so many cognomina include the elements ON it is 'preferable to suppose that Frontinus was consul earlier, in 73 or perhaps even in 72.' 72 would make Frontinus almost superhuman in the time between his praetorship and consulship. Eck Julius 3 (1974), col. 110, argues that Frontinus' consulship can be dated between 71 and 75. However, 73 seems secure and logical. Frontinus cannot have been consul in 74 since he must have already been on his way to Britain to replace Cerialis who was in Rome as consul II on May 1 and must therefore have left Britain earlier. Frontinus would have to have left for Britain before Cerialis' own departure. See CIL III, Vespasianus Veturio IX, p. 852. Ward Perkins 'Sex. Julius Frontinus,' p. 102.

zzz: Jones Titus, p. 138 and n. 137. See n.137 for details. See below for the reign of Titus.

zzz: Birley Fasti, pp. 71 and 77‑79, argues that A.D. 77 is a far more convenient and satisfactory year for Frontinus' tenure to end. For this reason alone it looks suspicious. However Brian W. Jones and D. Mattingly An Atlas of Roman Britain (Oxford, 1990), pp. 73‑74 and table 4.2, accept the earlier dating. The dating of Frontinus' office from 74‑77 would still give Frontinus more than three years as governor, more than the average for appointments to Britain. And at any rate, we have so little information on Frontinus' tenure as governor that such a change in date does not effect our knowledge of that office at all.

zzz: Tacitus Agricola xvii. Kappelmacher Iulius 243, col. 591. Sheppard Frere Britannia3 (London, 1987), p. 62, argues that the Silures 'were to prove themselves the toughest and most successful opponents which the Roman army was to encounter' in Britain. We can therefore probably not over estimate the magnitude of Frontinus' victory over this foe.

zzz: R. Syme 'Flavian Wars and Frontiers' Cambridge Ancient History XI (Cambridge, 1936), p. 152. We shall have to wait and see what a second edition of Cambridge Ancient History XI makes of Frontinus' governance.

zzz: V. E. Nash-Williams The Roman Frontier in Wales2, Michael G. Jarrett (editor) (Cardiff, 1969), p. 5. George C. Boon Isca (Cardiff, 1962), p. 10, argues that Frontinus 'substantially effected the conquest of Wales' leaving only one campaign for Agricola in the territory of the Ordovices.

zzz: See map 3. See also Jones and Mattingly Atlas, p. 72, table 4.1. For the forts themselves and their distribution see map 4 and Jones and Mattingly maps 4.16, 4.17, 4.33.

zzz: Tacitus The Agricola and the Germania, translated by H. Mattingly and S. Handford, (Harmondsworth, 1970), Agricola xvii, p. 68. Also see below.

zzz: Tacitus outlines the governors in Agricola xiv‑xvii. For Cerialis see below.

zzz: See Birley Fasti, pp. 37‑40. Birley argues that although very little is known of Plautius' prior career, the reason for his selection to lead the invasion can be readily explained through traditional political alliance between the Plautii and Claudii. He goes on to argue, p. 39, that Plautius was 'certainly holding office as a consular governor himself early in Claudius' reign' and suggests Pannonia in 39 — which would place him in a significant position at Claudius' accession in January 41, and in 42 during the coup of Scribonianus, legate of neighbouring Dalmatia.

zzz: Tacitus Annales XII.31‑39. Birley summarises the possibilities for his consulship, which we know he held with P. Suillius Rufus, but offers no alternative to Syme's suggestion of A.D. 45. Birley Fasti, p. 41 and nn. We have records of the pair's office: AE 1949.250 and 1973.152. Birley argues that 42, 44, and 46 should be ruled out because there is no room for Scapula in the consular Fasti of those years.

zzz: Birley Fasti, p. 43. A.D. 47 is ruled out as the year for his consulship since inscriptions date to August and November of the year of his consulship and he was already in Britain by years end. R. Syme 'Domitius Corbulo' Journal of Roman Studies 60 (1970), pp. 27‑39, at p. 28, argues that he might have been consul in 45, 'a notion supported by the guess that Scapula . . . had won merit . . . under Claudius in the campaign of 43.' Birley argues that exercitu ignoto 'untried army' (Annales XXI.31.1) rules out the possibility that Scapula had served with the army in Britain previously, but Tacitus' judgement in the Agricola implies that he had served as a soldier, most probably in Britain because it is unlikely that the governor of such a new province would be chosen from the ranks of the inexperienced especially with Caratacus still on the loose. The 'untried army' may refer to its operations in the territory of the allies. Birley, Fasti, p. 42, argues that 'Scapula may have been chosen for Britain on merit, but one imagines that the influence of L. Vitellius, then at its height, may have been a factor.'

zzz: Frere argues, Britannia, p. 67, that it was probably Cartimandua's surrendering of Caratacus which led to the breach with her husband.

zzz: Tacitus Agricola xiv. See Birley Fasti, pp. 44‑49. Birley argues, p. 48, that the situation in Britain upon his arrival were similar circumstances 'in many respects to those which Gallus had had to cope with in the Crimea in 45.' Gallus' activities in A.D. 45‑46 involved the installation of the Bosphoran king Cotys, a special mission to the Black Sea, and his annexation of the Thracian kingdom which involved warfare. Gallus was therefore prepared for what he would encounter in Britain especially in dealing with rival Brigantian claims of Venutius and his recent ex-wife Cartimandua to the throne. Gallus may even have had earlier experience in Britain as praefectus equitatus during Claudius' invasion. Frere Britannia, p. 66. Birley rejects this possibility, Fasti, pp. 46‑47, because it is unlikely that a consular of four years would have had such a role. Birley instead postulates that he was a cavalry commander under Tiberius. Gallus was curator aquarum from A.D. 38 to 49, although Frontinus refrains from making any comment on him whatsoever at De Aquis II.102, despite the possibility that he was appointed to this usually consular post before he was consul — a post which he seems to have held in 39. See Birley Fasti, p. 47.

zzz: Tacitus Annales XII.39.2. ita Silurum nomen penitus extinguendum. Tacitus relates the marked obstinacy of the Silures (ac praecipua Silurum pervicacia). Such a comment is useful for assessing the accomplishment of Frontinus.

zzz: Tacitus Annales XII. 40. See Frere Britannia, pp. 66‑67 and Peter Salway Roman Britain (Oxford, 1981), pp. 107‑108 and nn.

zzz: Tacitus Annales XII.40. nam Didius, senectute gravis et multa copia honorum, per ministros agere et arcere hostem statis habebat. 'since Didius, retarded by his years and full of honours, was content to act through his subordinates and to hold the enemy at a distance.'

zzz: On Veranius see Eric Birley 'Britain under Nero: the significance of Q. Veranius' in Roman Britain and the Roman Army (Kendal, 1976), pp. 1‑9. At p. 5 Birley argues that it was Claudius' decision to maintain the status quo in Britain under Gallus. However, the change of emperor and the uncertainty of the province seem more likely reasons. A change of heart by Nero would necessitate a change in governor, since it would be difficult for Gallus to change gears from guarding the frontiers to an all out aggressive conquest.

zzz: Tacitus Agricola xiv. Didium Veranius excepit, isque intra annum extinctus est. 'Didius was followed by Veranius, who died within the year.' Birley, Fasti, p. 50, argues that it is unclear whether Tacitus' intra annum means before the end of the calendar year or in less than twelve months. Birley favours the latter. Birley Fasti, p. 50, citing A. Gordon Quintus Veranius Consul A.D. 49 (1952), AE 1953.251 (Rome), argues that the date of Veranius' appointment should probably be considered as 57, with Veranius dying in office in 58, and not as 58 because Suetonius Paullinus arrived in Britain in 58. Syme Tacitus, pp. 765‑766. Birley argues that the Greek inscription partially recording Veranius' career, given in full on p. 50, points to 57 as being the year of his appointment since only it ties in with the congiarium of Nero. Having Paullinus arrive in A.D. 58 would also allow for the problem of Tacitus' date for Boudicca's revolt of 61 to be acceptably changed to 60 still during Paullinus' third campaigning season. See Syme Tacitus, pp 765‑766 and nn, and Birley Fasti, p. 56.

zzz: Birley Fasti, p. 51, argues that this was due to Veranius' father's service to Germanicus as legatus in the east in A.D. 18.

zzz: Birley Fasti, pp. 53‑54. Corbulo's campaigns began in A.D. 58 and it may be that in 57 the decision was reached to adopt a new aggressive policy — and that this decision also led to the replacement of Gallus.

zzz: Frere Britannia, p. 69. He states that for the thirteen years A.D. 47‑60 there was permanent military activity on the Welsh frontier. This he tries to explain away by arguing that conquest was not intended. Instead we should consider that a permanent solution was sought and that Veranius bragged he could have achieved it, Paullinus almost did, and that it was finally attained under Frontinus.

zzz: Tacitus Annales XIV. 29.1. However, Birley 'Veranius,' p. 1, cites C. E. Stevens 'The will of Q. Veranius' Classical Review n.s.1 (1951), pp. 4‑7, who argued at p. 6 that this boast was Nero's reason for reversing his decision on abandoning Britain.

zzz: Tacitus Annales XIV.29.2. On Paullinus' being extracted from retirement see Syme Tacitus, p. 387. Corbulo himself had been extracted from retirement and his success may have meant that with the death of, and therefore vacancy created by, Veranius he was replaced by another who fitted the Corbulo mould.

zzz: Tacitus Agricola xvi. Birley Fasti, p. 57, argues that although he is criticised by Tacitus, Paullinus does fare better than Petillius Cerialis. See below.

zzz: Tacitus Historiae II.37.1. For the consulship of A.D. 66 see Salway Roman Britain, p. 123. Birley, Fasti, p. 56, however, argues that the consulship was to a namesake, 'presumably his son.' He still maintains that Paullinus was not out of favour.

zzz: Frontinus De Aquis 102.110‑111. The corollary between governors of Britain and men who held the cura aquarum would make an interesting study but one impractical to go into here. Turpilianus' consulship of 61 also supports the later dating of Paullinus' tenure from 58‑61 and then Turpilianus' office. Turpilianus may have played a part in uncovering the conspiracy of Piso, earning the ornamenta triumphalia. See Birley Fasti, p. 58.

zzz: Dio Cassius LXIII.27.1. He was then executed by order of Galba in A.D. 68; no descendants are recorded. Tacitus Historiae I.6.1, I.37.3; Plutarch Galba 15.2, 17.3. See Birley Fasti, p. 58.

zzz: Birley Fasti, pp. 59‑60. Birley argues that Trebellius is described as segnior — more sluggish than the inactive Turpilianus.

zzz: See Tacitus Annales XV.46.2 and XVI.28.3. Suetonius Nero 13.2 records the closing of the temple of Janus which is dateable numismatically to 64. See Birley Fasti, p. 61 n. 19.

zzz: Tacitus Agricola xvi. This would seem at least tacitly to support Birley's argument.

zzz: He was not allowed to govern in absentia like Cluvius Rufus, another supporter of Vitellius.

zzz: Tacitus Agricola xvi. Trebellius recovered his dignity under Vespasian, as magister of the Arval Brethren in A.D. 72 possibly until 75. Birley Fasti, p. 62 and n.

zzz: Tacitus Annales XV.3.1. He may also have been proconsul of Macedonia after his consulship of A.D. 66. See Birley Fasti, p. 63.

zzz: Birley, Fasti, pp. 64‑65, argues that Josephus Bellum Judaicum VII. 4.2 suggests that Cerialis was designated governor before his appointment to deal with Civilis and so Bolanus would have been disinclined to commence a new campaign since he knew he was being replaced.

zzz: Tacitus Agricola xvii. Bolanus was rewarded with patrician rank in A.D. 73/74 and later became proconsul Asiae. As Birley argues, Fasti, p. 65, 'these items may serve to indicate that his achievements in Britain had not been negligible in the eyes of an emperor who knew that province well.' Bolanus' son Crispinus was the addressee of Statius' Silvae and this poem provides an hyperbolic examination of Bolanus' career stating for instance, 5.2.32‑40 and 47, that he was Corbulo's second in command.

zzz: Tacitus Agricola viii. On Cerialis see Birley Fasti, pp. 66‑69, which is based on Birley 'Cerialis,' pp. 179‑190.

zzz: Tacitus Historiae III. 59.2. See Birley Fasti, p. 66. He argues that the Caesii are well attested in Umbria. At 'Cerialis,' p. 182, Birley argues that Cerialis' command of cavalry immediately after his revelation also imply knowledge of local terrain.

zzz: Tacitus Historiae III.59.2. This close connection to Vespasian by A.D. 69, has been thoroughly examined by Gavin Townend 'Some Flavian Connections' Journal of Roman Studies 51 (1961), pp. 54‑62, who at p. 59 concludes that he must have been Vespasian's son-in‑law, married to Vespasian's only daughter Domitilla. She was dead by 69 but was later deified and had had a daughter. It is therefore likely, and Tacitus' account backs up, that Cerialis continued to be favoured under the Flavians.

zzz: Tacitus Historiae III.79.1. Birley Fasti, p. 68 n. 19, argues that this was his hallmark and cites Tacitus Historiae IV.76.3, 77.2, 78.2, V.20.1; and Annales XIV.33.1, as evidence.

zzz: Birley Fasti, p. 68 n. 25 although when we refer to his own note (Birley 'Cerialis,' p. 184) he argues that there is only 'a slight implication' that luck rather than Cerialis' generalship won the day. The passage referred to is Cerialis' final victory at Historiae V.81.2.

zzz: Tacitus introduces Cerialis as nec ipse inglorius militiae 'and being himself not without reputation in war.' This cannot be regarded, even given the defeat that follows, as critical. Tacitus Historiae IV.77.2 speaks of Cerialis' felix temeritas 'successful temerity.' Luck was necessary to any general so to speak of it pejoratively as Tacitus seems to, is self defeating. Cerialis should be included as one of the great generals of Agricola xvii and indeed when Tacitus states that Cerialis would have eclipsed the credit of any successor, if not for Frontinus, we cannot regard this as criticism of Cerialis. Therefore, despite the fact that defeats and incompetent 'lucky' victories are ascribed to Cerialis during the Civilis campaign and are attributed to his lack of caution, patience, and his failing to take advantage of a situation, (see Historiae IV.79.4, IV.78.2, V.15.2, V.18.2, V.20.2, V.21.2‑3, and V.22.1‑3), Tacitus cannot help but acknowledge that he was a good general.

zzz: Birley 'Cerialis,' p. 188. He argues that Tacitus' account makes it inescapable that Tacitus loathed Cerialis, and that what is said of Cerialis' achievements in the Agricola may 'represent less than justice.' He argues that Tacitus is therefore likely to have undervalued Cerialis' achievement in Britain in the lost books of the Historiae. His final sentence, p. 190, concludes that even if we had more of Tacitus' account of Cerialis' governorship of Britain, it 'might need to be treated with considerable reserve.'

zzz: The case may also be that Tacitus formed an ill opinion of Cerialis after composing the Agricola — in which case his attitude towards him in the Historiae may perhaps be tempered with that in the earlier work.

zzz: Birley Fasti, pp. 377‑378 and table 7. M Trebellius Maximus could have been from Italy of Gaul.

zzz: However, if the above reconstruction is correct he is the only governor of Britain with such recent experience before his appointment. This fact may either discredit the hypothesis or argue that Frontinus was truly exceptional — something his career itself supports.

zzz: Gallus possibly in the Crimea or Thrace, Veranius in Lycia, Paullinus in Mauretania, Trebellius in the Taurus mountains, and Bolanus possibly in Corbulo's eastern campaign. Cerialis probably had some experience of mountain warfare in Germany.

zzz: Where Frontinus got such experience could be either in Germany with Cerialis, or earlier with Corbulo in the east like Bolanus.

zzz: See below. The German appointment may have arisen because of his success in Wales itself.

zzz: On Isca see Nash-Williams Frontier, pp. 8, and 10‑14. See also V. E. Nash-Williams The Roman Legionary Fortress at Caerleon, Monmouthshire (Cardiff, 1940). On the via Julia see W. Camden Camden's Britannia (London, 1695), p. 602. Williams speaks of the 'several directions of the Via Julia,' David Williams The History of Monmouthshire (London, 1796), p. 42. These directions are presumably from Isca outwards. Cf I. Margary Roman Roads in Britain3 (London, 1973), pp. 323‑324 and figure 13. See map 5, reproducing figure 13. We can see that certainly road 60a is the via Julia but, according to Williams, the via Julia would have to incorporate also 60b, and 62a, (and probably 60aa — see below). George C. Boon, The Legionary Fortress of Caerleon-Isca (Caerleon, 1987), p. 14, has pointed out that Caerleon was a 'nodal point of the road system, and a key element was the bridge over the Usk.' It seems fair to assume that the via Julia included this bridge which must have been begun at the same time as the fort, in the campaign of Frontinus.

zzz: See Jones and Mattingly Atlas, maps 4.33 and 4.34 and Nash-Williams Fortress, Map A. Williams Monmouthshire, p. 42, argues that the directions of the via Julia presuppose that Frontinus took possession of all the Silures' major posts which he argues the location of Isca probably was.

zzz: V. E. Nash-Williams The Roman Frontier in Wales (Cardiff, 1954), p. 71.

zzz: Nash-Williams Frontier (1954), p. 63. This fort's position 'marked it as the strategic centre of south mid-Wales' and given that its base-fortress was Isca/Caerleon it would seem natural to associate its first occupation with Frontinus.

zzz: Nash-Williams Frontier (1954), p. 54. In the second edition Nash-Williams identifies Caersws I and II and argues that both may have been pre-Flavian, Frontier, pp. 66‑70.

zzz: Boon Isca, p. 10. These sea-borne landings are argued to have been a part of the conquest especially in the south, perhaps as far west as Neath. For water communications and transport see Nash-Williams Frontier, p. 146.

zzz: Nash-Williams Frontier (1954), pp. 111‑112. He argues that this plan accorded to Vegetius' De Re Militari IV.1. Much of this was edited from the second edition. See Johannes Kromayer and Georg Veith Heerwesen und Kriegführung der Griechen und Römer (Munich, 1928). See below for Frontinus' influence on the castramentation chapters of Vegetius.

zzz: Frere Britannia, p. 87, and Salway Roman Britain, p. 138. However, water pipes were still being laid in A.D. 79.

zzz: O. A. W. Dilke The Roman Land Surveyors (Newton Abbot, Devon, 1971), p. 203. See Frere Britannia, p. 87. See below.

zzz: Syme Tacitus, p. 790 argues 'that consular author, who was singularly reticent about contemporary warfare (Britain is absent).' See Birley Fasti, p. 70 and n.7.

zzz: Kappelmacher Iulius 243, col. 592; A. Dederich 'Bruchstücke aus dem Leben des Sextus Julius Frontinus' Zeitschrift für die Alterthumswissenschaft 105 (1839), p. 838. The stratagem follows Darius' deception of the Scythians by leaving dogs and asses in camp. When the Scythians heard these braying and barking, they imagined Darius was still there. I.5.25, Herodotus IV.135.

zzz: See Iuli Frontini Strategemata, R. I. Ireland (editor) (Leipzig, 1990), apparatus criticus I.5.26, where the emendation to Silures is credited to Dederich.

zzz: References to the Ligures in historical narratives do not contain the stratagem. See Birley 'Cerialis,' p. 189.

zzz: Birley 'Cerialis,' p. 190. The reason for this argument is that at Agricola xx Tacitus refers to multae civitates/many tribes giving hostages to Agricola beyond Brigantia, whereas Brigantia is referred to as a single civitas. The language implies previous contact with Rome, therefore under either Cerialis or Frontinus.

zzz: Pliny Historia Naturalis IV.102. W. S. Hanson and D. B. Campbell 'The Brigantes: From Clientage to Conquest' Britannia 17 (1986), pp. 73‑89, at pp. 87‑88.

zzz: Hanson and Campbell 'Brigantes,' p. 88. They argue that Frontinus, as a distinguished writer in the field of military science, is unlikely not to have built on the success of Cerialis in the north.

zzz: Birley Fasti, p. 71. He argues that archaeological evidence is consistent with Roman forces having been at Carlisle during his tenure, if not before.

zzz: Hanson and Campbell 'Brigantes,' p. 88. They continue on p. 89 that too many of the plethora of forts in the north of Wales and Britain are attributed to Agricola and that a re-assessment (as yet still not forthcoming) of the archaeological evidence is required to try and extract the true role of Cerialis and Frontinus in the consolidation of the Brigantian territory which until that time will be overshadowed by the evidence of the role of Agricola.

zzz: Williams Monmouthshire, pp. 36‑37. This road seems to correspond with road 60aa in Margary Roman Roads, pp. 323‑324, figure 13, and map 5.

zzz: Tacitus Annales I.11. Tacitus adds the comment incertum metu an per invidiam 'due to fear or jealousy.'

zzz: Syme Tacitus, p. 218. See A. R. Birley 'Roman Frontiers and Roman Frontier Policy: Some Reflections on Roman Imperialism' Transactions of the Architectural and Archaeological Society of Durham and Northumberland 3 (1974), pp. 13‑25.

zzz: Frere Britannia3, pp. 29‑30, assumes that the fragmentary Tim. . . in the text can be taken as Tincommius.

zzz: In a paper Ancient Roman Water Supply and Aqueducts delivered before the Auckland Classical Association on March 4 1997.

zzz: Ward Perkins' article was to prove that Frontinus, like his proconsular successor and predecessor, had experience in Britain and so was prepared for the position.

zzz: See Onasander Strategikos Prooemium 7‑10. Onasander disclaims any originality for himself; his comments do not seem to be those of a specious literary device. Onasander's work is a collection of generalship principles, and, as Frontinus and Vegetius tell us, would be useful and practical for that reason alone. Onasander has had very little attention paid to him as a military tactical writer because of his common-sense approach to the subject. He is considered of little worth because his treatise advises that which is so obvious and is couched in terms of general principles and not historical example. See J. Hornblower Hieronymus of Cardia (Oxford, 1981), p. 199. However, in terms of 'obvious' generalship, most if not all of the military disasters of history could have been avoided if such 'obvious' factors had not been ignored or neglected. What is more, as Onasander himself tells us, these general precepts are drawn from historical examples. Stratagems were themselves intended to communicate a military principle and so we cannot deride Onasander for communicating those principles directly.

zzz: Onasander Strategikos Prooemium 4. To de suntagma qarrounti moi loipon eipein wV strathgwn te agaqwn askhsiV estai palaiwn te hgemonwn kata thn sebasthn eirhnhn anaqhma 'It remains for me to say with good courage of my work, that it will be an exercise for good generals, and an object of delight for retired commanders in these days of imperial peace.'

zzz: Indeed N. Milner Vegetius and the Anonymous De Rebus Bellicis (D.Phil Thesis, Oxford, 1991), p. 298 and n. 266, argues that Onasander influenced Frontinus. See E. Sander 'Frontin als Quelle für Vegetius' Philologische Wochenschrift XLIX (1929), pp. 1230‑1231, at p. 1231; D. Schenk Flavius Vegetius Renatus: Die Quellen der Epitoma rei Militaris, Klio 22 (1930), pp. 28‑83, at pp. 81‑83; and F. Lammert Review of G. Bendz Die Echtheitsfrage des vierten Buches der Frontinischen Strategeamta (Dissertation, Lund, 1938), Philologische Wochenschrift LIX (1939), pp. 234‑239, at pp. 236‑237. Onasander's statement that kaqaper gar, ei tiV en polemoiV autoV strateusamenoV sunetaxato toionde logon, ouk an para touto httonoV hxiouto marturiaV, oti mh monon fusikhV agxinoiaV idian euresin eishnegkato strathghmatwn. 'if a general after experience in the field had composed such a work, it would not be considered of less value because he introduced and commemorated in his work, not only the personal discoveries of his native wit, but also the brilliant deeds of other generals,' Onasander Strategikos Prooemium 10, may have been one inspiration for Frontinus to write. We can consider these to be the stratagems of Frontinus himself; and those of Domitian, Corbulo, and Vespasian, to have been observed by Frontinus.

zzz: Indeed 'no mere closet philologist, at all events, may deny the possibility that it may prove useful to the professional soldier.' W. Oldfather, Onasander Introduction, Aeneas Tacticus, Asclepiodotus, and Onasander, (On the Defence of Fortified Positions, Tactics, and The General respectively), translated by members of The Illinois Greek Club, (London and New York, 1923), p. 351. And there have been professional soldiers who considered it of great use. Maurice de Saxe, Marshal of France, writing just before his death in 1750, held the treatise in the highest regard and in his Mes Rêveries 'declared with pleasure that he owed his first conceptions of the conduct of a commander-in-chief to Onasander.' Baron de Zur-Lauben Le général d'armée, par Onosander p. 5, quoted in Onasander Introduction, p. 351. Zur-Lauben is an Onasander enthusiast whose opinion is probably too high. Even if Maurice de Saxe is remembering Onasander somewhat too fondly in his retirement we cannot deny that he may have become interested in generalship through reading Onasander. Maurice commanded during the reign of Louis XV. He was made lieutenant general in 1733 and Marshal of France in 1745 during the war of the Austrian Succession, achieving victory over the British at Fontenoy. In 1746 he overran the Netherlands.

zzz: W. Smith Dictionary of Greek and Roman Biography and Mythology, Volume III, (London, 1849), p. 31. The Strategikos was a 'celebrated work on military tactics' and all 'subsequent Greek and Roman writers on the same subject made this work their text-book.'

zzz: E. McCartney Review of Aeneas Tacticus, Asclepiodotus, Onasander, Classical Philology 20 (1925), pp. 183‑185, at p. 185. For example McCartney points out Onasander's understanding of the use of noise and the importance of 'spick-and‑spanness' in chapters XXVIII and XXIX, and argues that chapter XII, 'stresses the need of having an army breakfast early when in the vicinity of the enemy' for which McCartney proffers the battle of Trebia, Polybius III.71‑72, and an engagement in Spain between Hasdrubal and Scipio, Polybius XI.22.

zzz: A. N. Sherwin-White The Letters of Pliny (Oxford, 1966), IV.8.3. S. Gsell Essai sur le Regne de L'Empereur Domitien (Paris, 1894), in Studia Historica 46 (Rome, 1967), p. 322. See Jones Titus, p. 125 and n. 67, and Jones Senatorial Order, p. 16 n.68.

zzz: Jones Titus, p. 137‑138. Such a hypothesis could provide insights into the career of Frontinus. All three men had been legati at approximately the same time, and Pansa, Traianus, and Titus had experience in the east. Traianus and Titus had been legati together in Judea in Vespasian's campaign. If Frontinus was involved in the campaigns he may have been taken by Titus to Ptolemais in A.D. 67 as part of legio V, X, or XV. Traianus had been legatus of legio X and Pansa legatus of possibly VI in the east and it would therefore seem appropriate if Frontinus had also commanded in the east for Jones' hypothesis. It therefore does not seem unreasonable to assume that Frontinus had military experience in the east before 70, with either Corbulo or Vespasian. Therefore, if Jones' hypothesis is reasonable, Frontinus can be assumed to have known Trajan's father well — giving a solid grounding for Trajan's later respect for Frontinus and his writings. We may also be able to strengthen a connection between Frontinus and the east. See below.

zzz: Jones 'Flavian Amici,' pp. 456‑458. For example Q. Petillius Cerialis Caesius Rufus, Epirus Marcellus, L. Junius Q. Vibius Crispus, M. Pompeius Silvanus Staberius Flavinus, T. Aurelius Fulvus, Q. Julius Cordinus, C. Rutilius Gallicus, Gnaeus Domitius Lucanus and his brother Gnaeus Domitius Tullus.

zzz: Jones 'Flavian Amici,' p. 460. Jones lists examples, n. 50: Traianus legatus legionis 67/68 consul suffectus 70. T. Aurelius Fulvus legatus legionis III Gallicae 64/69 and consul suffectus 69 or 70. L. Annius Bassus leg. leg. XI Claudiae 69 and consul suffectus 71. Sex. Vettulenus Cerialis legatus legionis V Macedonicae 67/70, X Fretensis 71, and consul suffectus ?73. Cn. Pompeius Collega legatus legionis IV Scythicae 69/70 and consul suffectus ?73. C. Dillius Aponianus legatus legionis III Gallicae 69 or 70 and consul suffectus ?73. Jones adds to this list in Titus, p. 138 n.137, to include M. Hirrius Fronto Neratius Pansa legatus legionis ?VI Ferrata and consul suffectus 73 or 74. For Frontinus to receive the fasces in 73, it seems almost certain that this would also make him a part of this group legionary commanders who were being rewarded for their service before 70. See above.

zzz: See Sherwin-White Letters of Pliny, IV.8.3; Gsell Domitien, p. 322; Jones Titus, p. 125 and n. 67, and Jones Senatorial Order, p. 16 n.68.

zzz: Everett L. Wheeler 'The Modern Legality of Frontinus' Stratagems' Militärgeschtliche Mitteilungen 43.1 (1988), pp. 7‑29, at p. 12, argues that the prestigious proconsulate of Asia was the capping off of Frontinus' career. Wheeler also argues, p. 12, that Frontinus commanded in Lower Germany 82‑84 and may have served in the Chattan war. However, Jones Domitian, p. 26 and n. 15 argues that Frontinus was one of Domitian's comites during 82/83. Jones, Senatorial Order, p. 21, argued that it is 'possible that Domitian appointed Sex. Julius Frontinus to Lower Germany, but he had already governed Britain during Vespasian's reign; furthermore, T. Vestricius Spurinna may have followed Frontinus in Lower Germany — but not before the fourth year of the reign.' See below and Jones Domitian, p. 59 and n.57. Wheeler, p. 12, and Eck 'Senatoren,' pp. 80‑81 and 137, date the Asian appointment to 86/87. See below.

zzz: Birley Fasti, p. 71‑2 and nn., Crook Consilium Principis, p. 168, Syme Tacitus, p. 214. Jones Domitian, pp. 58‑59 and nn. W. Eck Die Statthalter der germanischen Provinzen vom 1.-3. Jahrhundert (Cologne, 1985), pp. 141‑142, considers that Frontinus may have been governor of Lower Germany in 73/4, presumably after his consulship and before he went to Britain. However, Birley Fasti, p. 70 n. 11, says that such a position seems ruled out by the discovery that A. Marius Celsus was governor of Lower Germany in 72/73. See above. Brian W. Jones, 'The Dating of Domitian's War Against the Chatti' Historia 22 (1972), pp. 79‑90, at p. 90, argues that 'with the deaths of Vespasian and Titus, only Domitian was left and he was known to lack military experience.' This is the background to the observable fact that men like Frontinus were indispensable under Domitian.

zzz: Crook Consilium Principis, p. 168, n. 176, argues Frontinus was the amicus of the Flavians, Nerva, and Trajan. J. Devreker 'La Continuité dans le Consilium Principis sous les Flaviens' Ancient Society 8 (1977), pp. 223‑243, at pp. 226‑228, cited in Jones Titus, p. 125 n.68, argues that Frontinus was an amicus of Domitian, Vespasian and Titus. Most recently the revisionist Southern, in Domitian, p. 40, also implies that Frontinus was an amicus of Domitian but not included in Juvenal's fourth satire.

zzz: K. Waters 'Traianus Domitiani Continuator' American Journal of Philology 90 (1969), pp. 385‑405, at p. 390. Waters argues for administrative continuity between the reigns of Domitian and Trajan and argues that men like Frontinus, Julius Ursus, Pedanius Fuscus, and Iavolenus Priscus, who were energetic and able supporters of Domitian continued in favour under Trajan, 'and in many cases their sons or heirs.' This is indeed true of Frontinus in his son in law and descendants. See McDermott 'Stemmata Quid Faciunt,' pp. 229‑261.

zzz: CIL XIV.42 for cos. II and CIL VI.2222, SEG I.329 and XVIII.294 for cos. III. See also Pliny Panegyricus LXI.6.

zzz: Syme Tacitus, pp. 16‑17 and 35. See Pliny Panegyricus LX.5‑6. See also A. N. Sherwin-White Fifty Letters of Pliny2 (London, 1969), IX.19.1.

zzz: Sherwin-White Letters II.7.1-II.7.2. See also IV.8.3. He then explicitly postulates Frontinus may have had a say in the succession, p. 749. What is more E. Smallwood Documents Illustrating the Principates of Nerva Trajan and Hadrian (Cambridge, 1966), p. 3, argues that Spurinna replaced Frontinus as suffectus III.

zzz: Vegetius II.3. Vegetius' source for this claim may have come from the constitution of Trajan which he refers to as a source at I.8. However, as Milner points out in his note, Trajan may have only been introduced to please Theodosius I who wanted to be thought of as a second Trajan, because at I.27 only the constitutions of Augustus and Hadrian are mentioned. What then might this imply for II.3? Vegetius may have considered that Theodosius should read Frontinus and so invented Trajan's esteem or he may have known Theodosius had read Frontinus himself and was providing what he knew would be a welcome positive reinforcement of Theodosius' own opinion. However, we have seen that Trajan did hold Frontinus in high regard and so Vegetius' claims for Frontinus' writings are probably true.

zzz: See below. Indeed this respect for Frontinus by Trajan, Alphonse Dain, Histoire du Texte D'Élien le Tacticien (Paris, 1946), p. 18, seems to attribute to the fact that Frontinus dedicated his De Re Militari to Trajan. This is, however, unlikely.

zzz: Scriptores Historiae Augustae Severus Alexander 65.5. et id quidem ab Homullo ipsi Traiano dictum est, cum ille diceret Domitianum pessimum fuisse, amicos autem bonos habuisse. 'And this very thing was told to Homullus by Trajan, who said that Domitian was, indeed, a most evil man but had righteous friends.'

zzz: Pliny Panegyricus XV.1. See Syme Tacitus, pp. 10‑12, 18‑19, and 217‑235.

zzz: Pliny Panegyricus XIV.5, talks of Trajan commanding a series of campaigns after Saturninus' revolt. After 89 nothing is known of Trajan's military activities until 96 and these campaigns' identification remains conjectural. It is also not known exactly where Trajan served prior to being summoned from Spain in command of legio VII Gemina.

zzz: Frontinus had probably returned from his proconsulship of Asia by then. See below.

zzz: Pliny Epistularum IV.8.3 with Sherwin-White ad loc. Kappelmacher Iulius 243, col. 593: 'Wann Frontinus Augur wurde, ist nicht zu ermitteln.' 'When Frontinus was augur, cannot be established.' However, Houston Administrative Personnel, p. 720, and Schumacher Priesterkollegien, p. 254, cited in Birley Fasti, p. 72, argue that he was made an augur under Vespasian or Titus.

zzz: Pliny Epistularum IV.8. Sherwin-White Letters, Introduction, p. 80, and ad loc. Verginius had also regularly nominated Pliny. See Epistularum II. i.

zzz: Jones Senatorial Order, p. 3. Frontinus Strategemata I.1.8, I.3.10, II.3.23, and II.11.7.

zzz: G. Perl 'Frontin und der "Limes": Zu Strat. 1,3,10 und 2,11,7' Klio 63 (1981), pp. 563‑583, at p. 563. 'He speaks of the emperor in the appropriate way, and he reports in context and without exaggeration, although the frequent naming of Domitian and the spirited praise of his measures depicts a dutiful respect during the lifetime of Domitian. His brief mentions are all the more important as the main sources are not available, and the available authors are, on the one hand nothing more than exaggerated panegyric on the victorious emperor, and on the other those writing after his death reduce his activities to mere farce.'

zzz: See Jones 'Dating of Domitian's War,' pp. 88‑90. Jones uses this stratagem to date the beginning of the campaign to spring 82, and dates its end to summer 83 on the evidence of II.11.7. He argues that the Chatti always seemed to have attacked the Romans at times of weakness, such as in A.D. 9 and 69/70. Therefore in 81 would be another such opportunity.

zzz: Frontinus' stratagem is the earliest reference to the limes of Domitian. Jones 'Dating of Domitian's War,' p. 80 n.9.

zzz: Jones Domitian p. 130 and nn. 25‑27. However, Jones misreads Syme who argues that Domitian, over a front of one hundred and twenty miles 'drove military roads deep into the broken and wooded country that hitherto had secured them immunity (the Chatti) and thus opened access to their fortresses.' Syme Cambridge Ancient History XI, pp. 162‑163. Southern Domitian, p. 85, argues that such a penetration to Kassel, an area in the Chattan heartland perhaps vital to their economic or spiritual survival circa 120 miles north-east of Mainz, would make sound military sense.

zzz: However, as the Roman mile was 1000 double paces, therefore 2000 paces. Even if Roman marching paces were reasonably small, 75 kilometres seems far too small a distance, averaging only approximately 37 centimetres per pace. 175 kilometres, or thereabouts, would make each pace about 87 centimetres. This is seemingly an unfortunate typographical error although it does occur twice in the same sentence.

zzz: B. Isaac 'The meaning of the terms limes and limitanei' Journal of Roman Studies 78 (1988), pp. 125‑147, at p. 127. H. Schönberger, 'The Roman Frontier in Germany: an Archaeological Survey' Journal of Roman Studies 59 (1969), pp. 144‑197, at p. 159, offers another interpretation. He cites E. Fabricius ORL A Strecke 3 (1936), p. 45, arguing that Domitian had limites laid out over a distance of 120 Roman miles. Fabricius argued that this interpretation suited the layout in the Taunus and Wetterau. Schönberger argues that these roads were protected with wooden watchtowers 500‑600 metres apart although on level ground they could be up to 1000 metres apart. Southern Domitian, pp. 84‑85, also argues that limites were not frontiers but roads. She also argues that Fabricius' interpretation, that the frontier referred to the 120 miles between the Taunus and Wetterau, is invalid because the Wetterau was already occupied under Vespasian. That does not necessarily mean that a frontier or a road had already been built.

zzz: E. Luttwak The Grand Strategy of the Roman Empire (Baltimore and London, 1976), p. 214 n. 103. See also H. Düntzer 'Domitian in Frontins Strategemata' Bonner Jahrbucher H 96/97 (1896), pp. 172‑183, at pp. 182‑183. Most recently see Southern Domitian, pp. 84‑85.

zzz: Perl, 'Frontin und "der Limes",' p. 566. 'The emperor Caesar Domitianus Augustus had several broad paths to a (total) length of 120 miles (177.5 km) hacked out of the forest . . .' Perl bases this on the above interpretation of limes and not as it clearly means in I.5.10 (per limitem erumpere) a 'feste Reichsgrenze,' 'fortified imperial boundary.' However, Ireland and Bennett and McElwain have ab altera limitem agere coepit, tamquam per eum erupturus; which Bennett translates 'he began to build a road, as if intending to make a sally by this.' The construction of a wall in this context would also make sense for the stratagem.

zzz: Syme Cambridge Ancient History XI, pp. 182‑183. Schönberger 'The Roman Frontier in Germany,' p. 159.

zzz: See Isaac 'Limes and Limitanei,' pp. 125 and 130‑131. Isaac also provides an interesting corollary between limes and dolabris which could show influence from Frontinus. See below.

zzz: See A. Dederich 'Bruchstücke aus dem Leben des Sextus Julius Frontinus' Zeitschrift für die Alterthumswissenschaft 106‑107 (1839), pp. 841‑850, at pp. 843‑844.

zzz: This situation is not, as Dederich 'Bruchstücke,' p. 844, argues, exactly the same situation as the stratagem because the two constructions are described differently. What is more Dederich ascribes the second to the Dacian campaign not the Chattan one.

zzz: See below. Dederich considers the preface 'Bruchstücke,' p. 842, 'incontrovertible.'

zzz: If the text is interpolated it may have been written at a time when limites did mean a defensive structure, ie. after the first three centuries AD. See below.

zzz: Strategemata II.3.23. Southern Domitian, p. 84, argues that this makes it sound 'as though Domitian was at the scene of the battle, making spontaneous decisions, unless Frontinus has stretched the truth a little.' Despite her revisionist standpoint, Southern seems not to take advantage of Jones' view that Frontinus is the most balanced account of Domitian's campaign.

zzz: See B. Levick 'Domitian and the Provinces' Latomus 41 (1982), pp. 50‑73, at p. 64. Jones, 'Dating of Domitian's War,' p. 85, argues that Frontinus' is the earliest reference to Domitian's title Germanicus and that 'obviously' Frontinus' survey is chronological. Therefore this last stratagem implies that the title was conferred after hostilities had ceased. Jones also argues that this stratagem is associated with the period of consolidation and reorganisation after the campaign.

zzz: Germania xxxvii and Agricola xxxix state: proximus temporibus triumphati magis quam victi sunt. 'in recent times, they (the Germans) were more triumphed over than conquered;' and inerat conscientia derisui fuisse nuper falsum e Germania triumphum. 'In his heart was the consciousness that his recent counterfeit triumph over the Germans was a laughing-stock.' Pliny Panegyricus XVI.3: Accipiet ergo aliquando Capitolium non mimicos currus nec falsae simulacra victoriae. 'And the day will come when the Capitol shall see no masquerade of triumph, the chariots and sham trappings of false victory.' This statement may have had a wider reference, see Gsell Domitien, p. 196 n. 1. At 20.4 Pliny compares Domitian to the very barbarian hordes from whom he was fleeing.

zzz: Luttwak Grand Strategy, p. 92 n. 104. He also argues that Domitian's war 'established a frontier on the crest of the Taunus Mountains, which dominate — and could now protect — the fertile Wetterau. The result of this was to deny access to areas previously vulnerable to attack by the Chatti. See Perl, 'Frontin und "der Limes".'

zzz: This stratagem also provides an interesting connection between Frontinus and Vegetius. The latter reports at III.6 that melius est praecedere cum securibus ac dolabris milites et cum labore aperire vias. 'It is preferable that soldiers lead the way with hatchets and pickaxes and laboriously open up roads.' See also Josephus Bellum Judaicum 3.115‑126. Milner argues for the influence of Frontinus on Vegetius, especially Books III.6-IV.30, Vegetius and the Anonymous, pp. 255‑258; 292‑293 and 311. See also Schenk Flavius Vegetius Renatus, pp. 39‑81. See Appendix II.

zzz: Despite the problem of the exact interpretation of Frontinus' I.3.10 the campaign would have involved all three of these operations to some extent.

zzz: Jones Domitian, p. 59 and n. 58. Jones refers to I.1.8, Domitian's acting for the good of the provinces, and II.11.7, where Frontinus says how Domitian earned his title 'Conqueror of Germany' by beating the enemy, and refers to the fame of Domitian's justice.

zzz: Luttwak Grand Strategy, p. 92 and n. 103. He is referring to I.3.10. See above.

zzz: Campbell 'Teach Yourself,' p. 28. Campbell then examines the stratagems in turn.

zzz: Indeed, if the fourth book was added later (by Frontinus himself) as seems most likely (see below), he would have had time to consider such a comparison. This may help to date Frontinus' writing of the Strategemata before the end of Domitian's reign. See below.

zzz: Frontinus Introduction, p. xvi. Cf Gromatici Veteres Volume I, C. Lachmann (editor) (Berlin 1848), pp. 1‑58.

zzz: Dederich 'Bruchstücke,' p. 843. The text has not been referred to since Bennett and McElwain.

zzz: The reason for this may well be that of manuscript transmittance. The work itself was not included in many manuscripts, the preface in fewer still, so it could well be that the preface was not a priority for collectors of land surveying texts. As a result of such omission the preface may have become considered interpolated and so omitted today. See Dederich 'Bruchstücke,' pp. 843- 849.

zzz: He has probably therefore also edited and emended the text to make it comprehensible. Indeed, he argues 'Bruchstücke,' p. 844, that the vulgate is corrupt and therefore able to be altered. 'Die Worte sind also, wie sie in der Vulgata stehen, ganz gewiss verdorben. Man könnte ändern: statim (mihi) a septentrionali plaga (i.e. Dacia) Romam transire permisit.' Dederich seems to narrow in his interpretation of septentrionali plaga for it to mean northern Dacia. It is more likely that northern Europe is meant and therefore Germany. See below.

zzz: Dederich 'Bruchstücke,' p. 843 n. 4. Frontinus has an undatable Dacian stratagem at I.10.4. Scorylo dux Dacorum, cum sciret dissociatum armis civilibus populum Romanum neque tamen sibi temptandum arbitraretur, quia externo bello posset concordia inter cives coalescere, duos canes in conspectu popularium commisit iisque acerrime inter ipsos pugnantibus lupum ostendit, quem protinus canes omissa inter se ira adgressi sunt. Quo exemplo prohibuit ad impetu Romanis profuturo. 'Scorylo, a chieftain of the Dacians, though he knew that the Romans were torn with the dissensions of the civil wars, yet did not think he ought to venture on any enterprise against them, inasmuch as a foreign war might be the means of uniting the citizens in harmony. Accordingly he pitted two dogs in combat before the populace, and when they became engaged in a desperate encounter, exhibited a wolf to them. The dogs straightway abandoned their fury against each other and attacked the wolf. By this illustration, Scorylo kept the barbarians from a movement which could only have benefited the Romans.' This could be a stratagem of a loyal Dacian commander during the armis civilibus, possibly in 69. The stratagem itself therefore may have been one which was used diplomatically (by Frontinus himself?) during the Dacian war to remind the Dacians what they should do. Even if the stratagem was not recent it could still have been used for this purpose.

zzz: This makes Frontinus' appointment to Asia all the more interesting. See below.

zzz: Postquam ergo maximus Imperator victoriam Daciam proxime reseravit, et statim ea ad septentrionalem plagam transire permisit . . . 'Therefore when our most great Emperor had opened the way to his Dacian victory, and then by achieving it had cleared the way through to Northern Europe, . . .'

zzz: Dederich 'Bruchstücke,' p. 844. One possibility for the rejection of this passage is that Frontinus here refers to vallorum when he describes the construction of limites in the stratagem. The arguments above envisage that limites were not in any way ramparts or defensive works but roads alone. However, some still hold the view that these were roads with towers along them. This preface may give weight to the latter argument that limites were indeed ramparts or roads with protective towers. Alternatively Frontinus may have used limites in the de agrorum qualitate passage and a later scribe may have been confused by such an early use of the term and emended it to vallorum which was less ambiguous.

zzz: Dederich 'Bruchstücke,' p. 846. Dederich does not examine Asia, and the lack of reference to that post is disconcerting.

zzz: The position of governor of Lower Germany between 78 and 82 was suggested by Syme, Review of Stein and Ritterling, p. 97. See below.

zzz: Indeed this would support the view that he had begun the De Re Militari soon after his return from Britain.

zzz: Goesius De Re Agraria p. 142 n. cited by Dederich 'Bruchstücke,' at p. 844.

zzz: See above and At postquam primum hosticam terram intravimus, statim coelestia Caesaris nostri opera mensurarum ratione exercere coepi. 'And after our first entry into enemy territory, I began to do the celestial work of our Emperor in the capacity of land-surveyor.'

zzz: See Pliny Historia Naturalis II.18, Velleius Paterculus II.66.3 and II.104.3, and Tacitus Annales I.10.

zzz: Frontinus may, however, have been suffering from ill health and so retired after Asia to recoup. See below.

zzz: Although this is not true of Frontinus, he may have considered at the time that his political pinnacle had been reached in A.D. 87.

zzz: Dederich, 'Bruchstücke,' p. 841, although without reference. See J. Sandys Latin Epigraphy: an introduction to the study of Latin epigraphy (Cambridge, 1927), p. 311. Another possibility is votum soluit posuit . . . 'fulfilled her vow with this building.'

zzz: CIL VI 20483. McDermott 'Stemmata Quid Faciunt?,' p. 254, rejects the possibility that this is the same Julia who married Q. Sosius Senecio. See CIL VIII.7066 for family.

zzz: Wheeler 'Modern,' p. 12. See above. For the campaign see Jones Domitian, pp. 128‑131, and Parker Roman Legions, pp. 150‑151. There are major problems with the sources which make knowledge of the exact movements of the legions impossible.

zzz: Syme Tacitus, p. 214 and n. 3. Frontinus is the only adviser Syme deems it necessary to name. This is a more plausible scenario than to have Frontinus as governor of Lower Germany between 78 and 82.

zzz: We have seen above Birley's argument, 'Cerialis,' p. 187, that Cerialis, if active in 83 even if not consul, could have 'played a key role as an adviser on Domitian's German campaign.'

zzz: See Jones Senatorial Order, p. 3. If Frontinus was responsible for the strategy then his balanced account of the conduct of Domitian may be easily explained.

zzz: Frontinus' continued presence would certainly add plausibility to the idea that the strategy for the campaign was, at least to some extent, his.

zzz: Düntzer 'Domitian,' p. 183. ' This may make a closer participation of Frontinus in the war against the Chatti more than probable. This changes the balance in the matter of the question raised by Asbach over Zwanziger's objection . . . ie. the question whether Frontinus had a command in the war against the Chatti. There is not any specific evidence, and therefore there is nothing to contradict the supposition, which very much commends itself on its own grounds, that it is far more likely that he was placed at the side of Domitian in his first military expedition as an 'experienced military man,' which is what Tacitus calls him.' Düntzer cites Asbach Westdeutsche Zeitung V (1886), p. 369.

zzz: See Frontispiece. Eck 'Die Gestalt Frontins,' p. 54 Bild 1. CIL. XIII. 8624. See Ward Perkins 'Sex. Julius Frontinus,' pp. 102‑105. Frontinus Introduction, p. xvii. Syme, Review of Stein and Ritterling, p. 97. Eck conjectures the addition of [leg(ati) Aug(usti)?] to the end of the inscription and translates it '(Weihung an) Iuppiter den Besten und Größten, Iuno und Minerva fur die Unversehrtheit (des kaiserlichen Beauftragten) Sextus Julius Frontinus.' 'Dedication to Jupiter the best and greatest, Juno and Minerva for the successful discharge of the imperial appointment of Sextus Julius Frontinus.'

zzz: Ward Perkins tries to use it as evidence of Frontinus' presence in the Civilis campaign of 70.

zzz: Ward Perkins 'Sex. Julius Frontinus,' p. 103 and n. 2, assigns the command of II Adiutrix to Frontinus for that campaign, and argues that as a legatus legionis in Lower Germany in 70 his presence offers a satisfactory solution to the inscription, although he himself admits that the inscription is therefore a little surprising since legio II Adiutrix was based at Batavodurum (further north than Vetera. See map 2) (Tacitus Historiae V.20) in late 70. The problem, which he calls 'hardly serious', may be avoided all together if Frontinus did not command II Adiutrix or if the inscription does not date to circa 70. As we have seen above, command of II Adiutrix does seem attractive and therefore it is more likely that the inscription does not date to A.D. 70.

zzz: Oppenheim is very near Mogontiacum, the city via which Cerialis went to Trier. See above and map 1. Therefore, the daughter of Frontinus may have been present also which supports Frontinus' presence with Cerialis early in the campaign. A daughter of Frontinus, himself only in his thirties, would probably have been too young to go on campaign, but as the daughter of the legatus legionis she was still due a great deal of deference and the inscription might very well have been erected on her behalf. A legatus could certainly take his family along on campaign. Her presence may imply that Frontinus was indeed older, the daughter of a man nearer forty is more understandable on campaign with her experienced and valued father. Such a problem is avoided with a later date and with her father as a governor of Lower Germany or as a senior comes of Domitian.

zzz: Frontinus may be stating that the enemy did not venture forth to oppose him during his operations in which case his family would have been safe anyway.

zzz: We may therefore be able to restore a period of literary composition to Frontinus after his return from Britain. See below.

zzz: See Map 6. See Birley Fasti, p. 71 and n. 19. D. Magie, Roman Rule in Asia Minor (New Jersey, 1950), is of little use in this case. In appendix II, p. 1582, he only gives governors for 82‑83 and 83‑84 then considers that Civica Cerialis was governor in 89. Magie also places the governorship of Frontinus in 82‑83, a date shown to be wrong by the Phrygian gate inscription. See below.

zzz: The traditional date assigned to Cerialis' proconsulate is 88/89, Brian W. Jones 'C. Vettulenus Civica Cerialis and the 'False Nero' of A. D. 88' Athenaeum 61 (1983), pp. 516‑521, at p. 517 n. 4; but Eck 'Senatoren,' p. 86, argues for 87/88. What is more the two possibilities of Cerialis' successor (L. Mestrius Florus and M. Fulvius Gillo) are firmly dateable to 88/89.

zzz: Jones Domitian, p. 158 and n. 68 has for arguments for 87 and 88. The first to argue for 87 seems to have been R. Gephardt 'C. Suetonii Tranquilli Vita Domitiani: Suetonius' Life of Domitian with Notes and Parallel Passages (Dissertation, University of Pennsylvania, Philadelphia, 1922), p. 67.

zzz: The connection between the last two pretenders and the Parthians, and therefore the ramifications for Frontinus, will be dealt with below. See Tacitus Historiae I.2.1 mota prope etiam Parthorum arma falsi Neronis ludibrio. 'even the Parthians were almost roused to arms through the trickery of a pretended Nero.'

zzz: Suetonius Domitian 10.1‑2, Tacitus Agricola xli.1. See Jones 'Vettulenus,' pp. 516‑521, and Domitian, p. 158.

zzz: Jones 'Vettulenus,' p. 520. Jones suggests that the Parthians may have already been in negotiation with the Dacians as they were in Trajan's later war. The Parthians may have wanted to take the chance of embarrassing Rome, with little risk, by supporting the pretender as they had done with Terentius Maximus in 80.

zzz: See P. Gallivan 'The False Neros: A Re-Examination' Historia 22 (1973), pp. 364‑365; and P. Gallivan 'Suetonius and Chronology in the 'De Vita Neronis'' Historia 23 (1974), pp. 297‑318, at p. 318. Gallivan argues that the false Nero arose in A.D. 88 but he also argues that the interpretation of adulescente could mean any age between fifteen and thirty. Therefore twenty years need not be a strict figure.

zzz: If we consider that Frontinus was a legatus Augusti pro praetore after the departure of Domitian of Lower Germany then the appointment of T. Vestricius Spurinna to replace him falls between 84 and 86, and would fit with Frontinus' being sent to Asia. See Jones Senatorial Order, pp. 21 and 120. It was Spurinna who succeeded Frontinus as consul in 98 and 100. See above.

zzz: The fact that the brief account of Cerialis' execution as governor of Asia follows almost immediately on from chapter 8 in Suetonius implies that Cerialis was put to death as a result of Domitian's conscientious dispensing of justice.

zzz: By this date it may well be that the procurator was effectively the deputy governor, in which case Cerialis' replacement was not so drastic.

zzz: Jones 'Vettulenus,' pp. 517 and 520‑521. This also reinforces Suetonius' passage.

zzz: Dio Cassius LXVI.19.3. Jones also adduces Dio Cassius LXXIX.18.1‑3 for the kind of support given to the false Alexander in 221.

zzz: Frontinus seems not to have succumbed to this adverse situation. Indeed his administrative and military skills, or possibly his reputation alone, may have prevented any deterioration of the situation. Cerialis was less capable as an administrator, commander, and probably also as a diplomat. He may therefore have colluded with the false Nero, and Italus probably informed Domitian of any cooperation.

zzz: Jones Domitian, p. 139, argues that this took place between 17 March and 13 May of 86. See CIL XVI.32 and XVI.33. The appointment of Frontinus to Asia must have followed very soon after this.

zzz: J. Evans 'Tacitus, Domitian and the Proconsulship of Agricola' Rheinisches Museum für Philologie 119 (1976), pp. 79‑84.

zzz: Evans 'Tacitus,' p. 82. Evans rejects the view that Cerialis was a vir militaris; see especially n. 21. However in the context of the false Nero, he almost certainly would have been. See Jones 'Vettulenus,' p. 518.

zzz: Evans 'Tacitus,' p. 83. He argues that Frontinus gained the appointment as reward for 'outstanding service to the Flavian dynasty in three military theatres.'

zzz: Jones Senatorial Order, p. 120. Jones 'Vettulenus,' p. 518. Cerialis' brother Sextus had been legatus legionis of V Macedonica from 67‑70 during Vespasian's campaign in Judea, and was then legatus legionis of X Fretensis in 71. In 75 he held the proconsulship of Moesia. Jones suggests that he was then legatus Augusti pro praetore Iudaeae circa 70‑71.

zzz: Jones, 'Flavian Amici,' p. 460 n. 50, has '?73' for the consulship but in Senatorial Order, p. 120 cites Eck's date of 72, 'Senatoren,' p. 93.

zzz: Tacitus Annales III.32. See H. Traub 'Agricola's Refusal of a Governorship' Historia 24 (1975), pp. 255‑257.

zzz: Tacitus Annales III.35. Traub 'Agricola's Refusal,' p. 256 n. 12; F. Marsh The Reign of Tiberius (London, 1931), p. 149.

zzz: However, Eck 'Senatoren,' pp. 86 and 140, and Jones 'Vettulenus,' p. 519 argues that the proconsul in 88/89 was M. Fulvius Gillo.

zzz: This might therefore imply that Cerialis did indeed collude with the pretender. The false Nero was held at bay by Frontinus until 87, and then again during Italus' and Patruinus' tenures in 88. Indeed he was dealt with by the arrival of Florus in the late second half of 88. Domitian's execution of Cerialis as a promoter of revolution may therefore have been completely justified. The increase of troops in Syria must have occurred sometime before November 88 since winter would seem a very late time to change governor. It might however rob the pretender and the Parthians of an opportunity to attack with the change of governor occurring then. We may therefore conjecture that the change of governor in 87 was also late in the year.

zzz: If Domitian regarded Frontinus, an accomplished general and administrator, as a threat then the posting to Asia may have been a deliberate decision to give Frontinus a non-military command which would necessitate his retirement afterwards. This 'packing off' of Frontinus to where he would be considered less of a challenge to the emperor may also have been the decision behind Frontinus' limites project in Germany — giving Frontinus extended service after the military campaigning in a corner of the new province whereby he would not be prominent in the minds of Rome.

zzz: However, it seems most likely that Cerialis would have been appointed with the expectation that he follow on from the good work of Frontinus before him, and that there was no need for Frontinus to hold the post again. What is more Frontinus had presumably laid effective administrative ground work for his successor to operate with and so could vacate the post having achieved his goal.

zzz: His observations may have added to the opinion he had formed during Corbulo's campaign.

zzz: Silvae 4.4.61‑64, translated by J. Mozley, (London and Cambridge, Massachusetts, 1955). See A. Hardie Statius and the Silvae. Poets, Patrons and Expideixis in the Graeco-Roman World, ARCA Classical and Medieval Texts, Papers and Monographs 9 (Liverpool, 1983), pp. 164‑171 especially p. 166.

zzz: Jones Domitian p. 159 and especially n. 75. Jones argues that there is no evidence of plans for a grandiose eastern campaign. 'For one thing, Domitian could not afford it.' Jones uses Grosso's arguments, 'Aspetti della politica orientale di Domiziano, II' Epigraphica 17, pp. 33‑78, pp. 36‑55, to the effect that at Silvae 4.4.30‑31, one can gain the opposite impression. A. Bosworth 'Arrian and the Alani' Harvard Studies in Classical Philology 81 (1977), pp. 217‑255, at p. 227, argues that Roman presence in the Caucasus seems assured, and that it was the natural sphere of operations of the Cappadocian army. See N. Debevoise A Political History of Parthia (Chicago, 1938), p. 210; A. Schieber The Flavian Eastern Policy (Dissertation, Buffalo, 1976), pp. 145‑146; Bosworth 'Arrian and the Alani,' p. 227.

zzz: Mionnet III 206, 1121, 210, 1155, 1156; suppl. VI 312 ff. BMC Ionia p. 250 nos. 133‑137. Kappelmacher Iulius 243, cols. 592‑593. See Figure 1. Eck 'Die Gestalt Frontins,' p. 55 Bild 2. Eck has µanJu(patw) Fronteinwµ and translates it 'unter dem Prokonsul Frontinus.' 'Under the proconsul Frontinus.'

zzz: See Figures 2 and 3 Eck 'Die Gestalt Frontins,' p. 56, Bild 3a and 3b. Eck 'Senatoren,' pp. 77‑78. C. Jones, Review of W. Eck Senatoren von Vespasian bis Hadrian, Gnomon 45 (1973), pp. 689‑690, improved the inscription thus: TRIB POTEST [VI IMP XIII]II COS XII PP dhmar[cik]hV e[xousiaV to V autokratori to id, upatw to ib, pa]tri patridoV Eck dated the proconsulship to 86/87 but Jones' arguments state that 85/86 is still more likely. Eck pointed out to Birley, Fasti, p. 71 n. 19, that Frontinus was certainly in Asia in 86 because Domitian is cos. XII on the Hierapolis inscription. Also Eck, 'Senatoren' p. 82, has shown that for Domitian's Asian proconsuls thirteen years elapsed between their first consulship and appointment as proconsul Asiae. Therefore for Frontinus 73 to 86. See also Evans 'Tacitus,' p. 81 n.10. Evans also accepts Frontinus' proconsulate as 86/87. Agricola's refusal of the governorship of 90/91 also fits with this equation since he had been consul in 77. CIL III 7059 has: brit[ANNICO PONT MAX TRIB POTE]st porta[M ET TU]rres brettaik[* ARCIEREI MEGISTW DHmArcikHS E]xousiaV [KAi TO*S PU]rgoV Jones, Review of Eck, p. 689, complains of the unexplained discrepancies between publications of this inscription, stating that he and Bowersock, in 1966 copied POTES IIII to correspond to the Greek TO D UPA. This does not appear in the other published versions of the inscription above.

zzz: No conclusions seem to have been drawn from the existence of this gate but such building projects, and probably the roads that accompanied them, were the kinds of schemes (both civil and military) that military men were necessarily good at. We can only speculate whether this was a civil or a military but it would seem most likely that the Phrygian Hierapolis gate was a non-military construction.

zzz: This is true but Wheeler seems to have ignored hindsight by arguing that Frontinus capped off his career in 87 when we know he had (almost) another career ahead of him.

zzz: Martial Epigrammata X.lviii, Volume II, translated by D. Shackleton Bailey, (London and Cambridge, Massachusetts, 1993).

zzz: Translations of this epigram do not use Frontinus, instead they insert Faustinus. MSS. of Martial all have Frontine, M. Valerii Martialis Epigrammaton Libri J. Borovskij (editor) (Leipzig, 1976), and M. Valerii Martialis Epigrammata D. Shackelton Bailey (editor) (Stuttgart, 1990), X.lviii. The relationship between Martial and Faustinus may well be the reason. Martial was Faustinus' client. He was a rich patron with villas at Tivoli, Tarracina, Trebula and Baiae (described in II.lviii), Martial honoured Faustinus in books III and IV and gives him 19 epigrams up to book X 'after which he drops from sight.' J. Sullivan Martial: The Unexpected Classic (Cambridge, 1991), p. 18. Faustinus was expected to disseminate Martial's works to friends (VII.lxxx), protect the poet from attacks (VII.xii), and encourage him in general. The previous references probably led to the emendation to Faustinus, an emendation with no good reason.

zzz: Kappelmacher Iulius 243, col. 594. 'That a man and particularly a literarily active man, so seen would be honoured by Martial is no surprise.'

zzz: Martial himself gives us a picture of the blissful life at Formiae, Epigrammata X. 30. Peter White, Promised Verse (London and Cambridge, Massachusetts, 1993), pp. 25‑26, uses this epigram and translates the last lines thus: 'To hang by your door night and day is not the only way to be a friend — besides, poets cannot afford that loss of time. By the rites of the Muses which I practice, by all the gods, I swear I am your friend even if I do not pay court to you.' This seems to overrate the relationship between Frontinus and Martial although it would explain the lack of epigrams addressed to him. It also seems to make the reference to 'your door' too specific to Frontinus and not a complaint about the lot of the poet generally. However, White points out, Promised Verse, p. 26, that few poets had 'enough social independence to limit their services to great friends as Martial tried to do.' Frontinus may have been one of these great friends.

zzz: Martial Epigrammata X.xlviii. . . .saturis mitia poma dabo, de Nomentana vinum sine faece lagona, quae bis Frontino consule trima fuit. 'When my guests are satisfied, I shall offer ripe fruit and leesless wine from a Nomentum flagon twice three years old in Frontinus' consulship.' or '. . .which was three years old in Frontinus' second consulship.' This line has created problems because there is no evidence for bis=iterum in the classical period so it is usually taken with trima. However, Martial is post-classical and can therefore use bis for iterum with no problems. Therefore bis Frontino consule should translate 'when Frontinus was consul for the second time.'

zzz: McDermott 'Stemmata Quid Faciunt,' p. 256. He states that Duff's argument is 'by no means a necessary interpretation of ll. 5‑6 of this epigram.'

zzz: A. Trevor Hodge Roman Aqueducts & Water Supply (London, 1992), p. 16. In Roman Water Supply, Professor Hodge stated that his school Latin teacher first said that the De Aquis was one of the driest works in Latin literature. Hodge added the rejoinder that whilst his teacher was a learned man he was wrong in this respect — the De Aquis was absolutely, without doubt, the driest in Latin literature. It was in this same paper that Professor Hodge included Frontinus in the ranks of the senatorial nobles appointed to top jobs under Nerva and Trajan.

zzz: See below. See maps 7 and 8. The proximity of Anxur to Formiae would easily allow for Frontinus to have estates in both regions. Anxur-Tarracina and Formiae are 20 miles apart on the via Appia past the Pomptine marshes. Frontinus could spend days of leisure at these, and probably other villas, where he could entertain such men as Aelian and Martial and possibly Nerva and Trajan. Aelian says that he was able to spend some days at Fromiae with Frontinus, implying a previous relationship enabling such a stay.

zzz: McDermott 'Stemmata Quid Faciunt,' pp. 229‑261. McDermott also assigns less certain estates to the family, in Tusculum, Bologna, Volturnum, and probably in Asia Minor — possibly Laodicea or Castabala but more likely in Phrygian Apamea.

zzz: Devine argues, Aelian Introduction, p. 31, that the Preface suggests that Frontinus was no longer writing at the time of his and Aelian's meeting. Nothing in Aelian seems to support such a supposition. We know that Frontinus wrote the De Aquis (Preface 1) after that appointment, so following A.D. 97. epi tou qeou 'in the time of the deified' Nerva would suggest that 96 is correct and so Frontinus had not finished writing.

zzz: Indeed Aelian may only be treating Frontinus with respect because of the high regard Trajan held him in. However, the Preface does not smack of insincerity and there would seem little use in evoking a meeting with a man, albeit a respected one, ten years after his death, especially when there was, presumably, no corroboration for Aelian's story of the meeting, for no other purpose than Aelian ingratiating himself with the emperor.

zzz: See Dain Histoire du Texte, p. 21. He adds the pertinent if frustrating point 'mais il faut savoir se résigner', 'but it is necessary to know how to resign oneself' and indeed one does know when to give up. At p. 23 he again shows a necessary indecisive character 'il vaut mieux se résigner à ignorer' 'it is best to resign oneself to ignorance.'

zzz: Dain Histoire du Texte, p. 21. 'Aelian will be to us just the 'Tactician': fortunately by this title we avoid confusion with the other Aelians.'

zzz: OCD3, p, 18, identifies all these works by the same author Claudius Aelianus (A.D. 165/70‑230/5) and so he cannot have been the author of the Tactica.

zzz: Philostratus Vita Sophistae 624‑625. However, this man lived to be sixty but was alive during the reign of Heliogabalus and so cannot be the author of the Tactica.

zzz: Suidas sv. AilanoV: Aelian of Praeneste to whom Suidas gives the titles of both sophist and pontifex. Suidas himself probably confused this Aelian with the author of the Tactica. What is more Aelian of Praeneste is dated to the reign of Hadrian, the incorrect addressee of the Tactica, and so identification with the author in the past may have seems attractive and appropriate.

zzz: Dain Histoire du Texte, p. 23, 'we are dealing with a man known in literary circles.'

zzz: See below. Dain argues that we cannot date the treatise accurately, Histoire du Texte, p. 23, but the circumstances of Trajan's Parthian war would seem to give us a reasonably narrow window for its publication.

zzz: See below. Frontinus may have possibly foreseen the inevitability of a war with Parthia during his time in Asia. He may therefore have prepared Aelian for the possibility of a forthcoming use for his treatise. Such an image does seem very human. A belle-lettriste worried about a project and being encouraged by a learned and respected senator and friend that it would indeed be useful.

zzz: Dain Histoire du Texte, p. 19. F. A. Lepper, Trajan's Parthian War (Oxford, 1948), p. 29, argues that by autumn A.D. 114 Trajan was already in Armenia. Lepper argues that a departure date from Rome on about 27 October 'very suitable.' Indeed R. P. Longden 'Notes on the Parthian Campaigns of Trajan' Journal of Roman Studies 21 (1931), pp. 1‑35, at p. 1, argues that 'no one now doubts that Trajan left Rome in the autumn of 113.'

zzz: The meeting may be a literary convention for Aelian to give himself some credibility but seems authentic. The main examination of Aelian's Tactica in English is Devine. See also P. Stadter 'The Ars Tactica of Arrian: Tradition and Originality' Classical Philology 73 (1978), pp. 117 -128. See Dain Histoire du Texte, pp. 18‑19.

zzz: Aelian Tactica Preface 6. Indeed, a slightly different view of his work's application to Frontinus is also very human. He may have disagreed or missed the point specific to Frontinus' encouragement as so often happens when young people, who consider themselves learned, discuss their intentions with more learned superiors. Often the approach that the superior takes is inaccessible or inexplicable to the junior and they often consider that approach to be some strange tangent. The junior will then either interpret to his own satisfaction, or ignore the suggestion. It could well be that which happened between Frontinus and Aelian especially if Aelian as a Greek and antiquarian missed the point made by Frontinus, a practical and experienced general who waxed lyrically about the need for a disciplining phalangite treatise. Aelian may have realised in 113 the point or approach Frontinus' meant and therefore, after long delay, published his work. And also Aelian has left in his own antiquarian idea convinced of its validity — again a very human approach.

zzz: Corbulo's campaigns between A.D. 58 and 64 involved no pitched battle but phalangite tactics may still have been in operation. However, Parker Roman Legions, p. 258, argues that Tacitus Annales XIII.38.6 gives the impression that the ordinary Republican formation was adhered to.

zzz: Pliny Panegyricus LXII.2. Cf LXI.6. See Birley Fasti, p. 72. See also Epistularum II.1.9, Dio Cassius LXVIII.2.3, McDermott 'Quid Stemmata Faciunt,' p. 256. Pliny's comment is that this commission Hoc est igitur hoc est, quod penitus animo Caesaris insinuavit. 'This, then, is what recommended them so warmly to Caesar.' However, Frontinus' probable earlier contact with Trajan in whatever form seems just as likely for Trajan's esteem.

zzz: R. Syme 'The Imperial Finances Under Domitian, Nerva and Trajan' Journal of Roman Studies 20 (1930), pp. 55‑70. The via Appia was repaired and a new road from Puteoli to Naples was built. Frontinus himself was to extend the Anio Novus and Aqua Marcia: De Aquis 88.

zzz: Indeed Syme 'Imperial Finances,' p. 63, argues that the plebs received more cash from Nerva in sixteen months than they did in almost as many years from Domitian. Syme's comment is 'it looks very much like bribery.' Not only that but, at p. 59, Syme points out that upon his accession Nerva gave a congiarium of 650 denarii whilst in fifteen years as emperor Domitian had given only 225 denarii in total.

zzz: Indeed the efforts of the economic commission were feeble if their considered purpose was to refill the treasury. Syme argues, 'Imperial Finances,' p. 59, that the economies of the commission 'are so trivial as to be laughable.' At p. 65 Syme argues that this commission was 'not a feeble and tardy attempt to fill a Treasury which Domitian had left empty, but a mild palliative to the extravagance of those who came after.'

zzz: De Aquis 102. Hodge, Roman Aqueducts, p. 16, argues that Frontinus was appointed curator aquarum by Trajan. Birley Fasti, p. 72 n. 21, argues that if the office had been held by Aviola since 74 this would explain why Frontinus found so much neglect and so many abuses. See De Aquis 110‑118. Frontinus' tenure of this office certainly seems to point to the need for reform. Kappelmacher Iulius 243, col. 593, argues that Frontinus 'als Curator Aquarum führte er eine Reihe von Neuerungen ein, die auf eine gerechte und fur die Allgemeinheit ersprießliche Verwaltung schließen lassen.' 'As curator aquarum he led forth a range of innovations, which were carried out with a judicious concern for the common good.' See De Aquis 64. 'schützt den Staat vor Diebstählen und untüchtigen Beamten.' 'He protected the state from theft and incompetent officials.' See De Aquis 87, 112, 114, and 117.

zzz: Hodge, Roman Aqueducts, p. 17, argues that the work throws a great deal of light on Frontinus' character and the standards of public service he set and lived up to.

zzz: CIL XV 7474. Frontinus Introduction, p. xvii. See also CIL IX 6083.78. ∫SEXTI JULI FRONTONI∫. The explanatory note of this inscription in the CIL is so elementary that it gives no clue to the significance of this inscription although it does describe the spelling as sic. Civita d'Antino [apud Manfredum Ferrante]. 'At the house of Manfred Ferrante.' It seems to have been some kind of domestic instrument.

zzz: Bennett and McElwain translate this passage 'which might serve to guide me in my administration.'

zzz: Hodge Roman Aqueducts, p. 17. n. 16, argues that Frontinus' publication of his successes in technical reforms in a public service and in a political context reminds him of Mussolini and the Italian Fascists' boast of making the trains run on time. However, the policy move seems to have been an entirely sensible one and we need not be so harsh in judging Nerva for an increase in effective and efficient public service administration.

zzz: Ward Perkins 'Sex. Julius Frontinus,' pp. 102‑105. Ward Perkins lists as his examples of such a policy both Cerialis and Agricola, and also Funisulanus Vettonianus and Sex. Sentius Caecilianus. He speaks, p. 105, of the 'tendency of the Flavian emperors to appoint men to posts for which their previous careers had especially fitted them.'

zzz: See above. However, at De Aquis 64, Frontinus talks of optimi diligentissimique Nervae principis 'that best and most industrious emperor Nerva;' while at De Aquis 93 he talks of imperatorem Caesarem Nervam Traianum Augustum; and at 102 of M. Cocceius Nerva, divi Nervae avus 'Marcus Cocceius Nerva, the grandfather of the Deified Nerva.' Bennett and McElwain's note on De Aquis 64 states that Trajan not Nerva is meant, but Nerva may still have been alive at the time of writing chapter 64 yet not at 93. We may therefore have a work in progress which was added to as necessity arose; an expectable situation in a work for his own guidance. Certainly this reinforces the view that sees Frontinus writing whilst in office, which he may have also done for his earlier writings. Note also Frontinus' claim that the work was for him to re-examine for policy formation.

zzz: Birley Fasti, pp. 71‑72, argues that after his year in Asia, Frontinus probably occupied himself largely with writing.

zzz: Frontinus De Aquis Preface 1. See also Preface 2 where he states that there is no surer way to discover what ought to be done and what ought to be avoided than to familiarise oneself with the business entrusted. Also that it is disgraceful for a decent man to conduct his business by subordinates and not know the details of the business himself. See also Frontinus Introduction, p. xxvi. This reinforces the probability that Frontinus read Onasander before going to Britain, not to mention innumerable works by authors which are no longer extant.

zzz: He may therefore have written the De Re Militari in office, although exactly which military office is harder to postulate. But the governorship of Britain was his most important military post to date in 74 and so he may have written it when entering into that office, although he was already militarily experienced. This would fit with his next statement in Preface 2: In aliis autem libris, quos post experimenta et usum composui, succedentium res acta est; huius commentarii pertinebit fortassis et ad successorem utilitas, sed cum inter initia administrationis meae scriptus sit, in primis ad meam institutionem regulamque proficiet. 'Now in the case of my other books which I have written after practical experience, I consulted the interests of my successors. The present treatise may also be found useful by my successor, but it will serve especially for my own instruction and guidance, being prepared, as it is, at the beginning of my administration.' It would seem most likely then that he wrote the De Re Militari after governing Britain.

zzz: It is probable that Frontinus and Pliny the Elder knew each other as men prominent in the reign of Vespasian, although Pliny was approximately ten years older. If the reconstruction of Frontinus being an equestrian is correct then their connection may have been stronger since Pliny was procurator of Africa in circa A.D. 59, and later in Spain, both areas which Frontinus seems to have also have had experience with. See above. In the same vein Lucius Junius Moderatus Columella, an equestrian from Cadiz in Spain and a contemporary of Pliny the Elder, who probably published the De Re Rustica after circa A.D. 65, states in his preface to Book V that he was encouraged by his commander Marcus Trebellius to continue an interest in agriculture and take up land measurement. See M. Beagon Roman Nature (Oxford, 1992), p. 6. Such pursuits encouraged by a military commander may also have bearing on Frontinus' interest in surveying and land measurement.

zzz: Pliny Epistularum III.5.8‑19. Pliny the Younger seems to have considered that his uncle possessed a passion for saving time. The whole letter makes an interesting parallel to Frontinus in that it begins by enumerating the writings of Pliny the Elder and the reasons for their composition.

zzz: Kappelmacher Iulius 243, cols. 595‑605. Frontinus Introduction, p. xiv.

zzz: The scribe of the earliest MS, dated by Devine Aelian Introduction, p. 34, to the reign of Constantine VII Porphyrogenitus (reigned A.D. 912‑959) may have possibly known of Fronto's greater appropriateness to the writings of Homer as opposed to Frontinus'. See below.

zzz: Fronto's Correspondence has references to Iliad I.24, II.223, III.112, VI.236, VI.408, VIII.311, IX.203, IX.312, XIV.350, and XXIII.282; Odyssey I.58, VI.106, III.117, X.29, X.31, X. 46, XI.108, XII.338, XII.359, XII.364, XII.370, and XII.372. See M. Cornelius Fronto Correspondence, 2 Volumes, edited and translated by C. Haines, (London and Cambridge, Massachusetts, 1928 and 1929).

zzz: Lachmann Gromatici Veteres I, pp. 1‑58. Lachmann's texts may combine the book on agriculture with the surveying text but Wheeler 'Modern,' p. 12 argues that it was a separate work.

zzz: See below. However, the work was considered certain by Kappelmacher, Iulius 243, col. 603. See also A. Dederich 'Bruchstücke aus dem Leben des Sextus Julius Frontinus,' Zeitschrift für die Alterthumswissenschaft 106‑107 (1839), pp. 1077‑1080.

zzz: Dilke Surveyors, p. 126. Dilke dates the work on surveying to the reign of Domitian, p. 41. Lachmann also edited those parts of the texts that he considered to be interpolated from the commentary of Aggenius Urbicus. Kappelmacher Iulius 243, cols. 595‑597.

zzz: Frontinus Introduction, p. xiii. Frontinus' predecessor in fostering Pliny's career, Verginius, was also educated in Alexandria and it seems safe to assume that these two patrons of Pliny were most probably friends. The dating of Heron is problematic. There is little evidence to suggest a date for him other than after circa 150 BC and before A.D. 250, respectively after Apollonius, whom he quotes, and before Pappus, who cites him. However, in 1938 O. Neugebauer pointed to the eclipse of the moon in Heron's Dioptra, as being only possible in A.D. 62 (O. Neugebauer Über eine Methode zur Distanzbestimmung Alexandria-Rom bei Heron, cited in A. Drachmann The Mechanical Technology of Greek and Roman Antiquity (Copenhagen, 1963), pp. 9‑10, and 12). This seems to have been accepted since. See Drachmann pp. 9‑10, 12 and 19‑21.

zzz: Dilke Surveyors, p. 129. Dilke gives no arguments for this assumption. We can argue against this if we believe in 'the ready-reckoner I have always kept' — ie. if we think, as the disputed de agrorum qualitate preface posits, that he kept a technical notebook throughout his life (possibly from when he was in Heron's class onwards).

zzz: Dilke, p. 203, without reference and somewhat enigmatically, argues that 'recent research suggests that the part played by [Frontinus] in the planning of Verulamium may have been underestimated.' See above.

zzz: Frontinus Strategemata II.3.21. This is stated to be application of Homer Iliad IV.299‑302.

ιππηας μεν πρωτα συν ιπποισιν και οξεσφι,
πεζους δ᾽ ἐχοπιθε στησεν πολεας τε και εσθλους,
ερκος εμεν πολεμοιο. κακους δ᾽ ἐς μεσσον ελασσεν,
οφρα και ουκ εθελων τις αναγκαιη πολεμιζοι.

zzz: Vegetius I.5. This reference is to Iliad V.801 TudeuV toi mikroV men ehn demaV alla machthV. 'since Tydeus was a small man for stature, but he was a fighter.' Translation Lattimore.

zzz: See below. Aelian Tactica Preface 1. Polybius XVIII.29 includes a quote from Homer Iliad XIII.130‑132 whilst describing the phalanx: ασπις αρ ασπιδ ερειδε, κορυς κορυν, ανερα δ᾽ ανηρ. ψαυον δ᾽ ιπποκομοι κορυθες λαμπροισι φαλοισι νευοντων. ως πυκνοι εφεστασαν αλληλοισι. 'locking spear by spear, shield against shield at the base, so buckler leaned on buckler, helmet on helmet, man against man, and the horse-hair crests along the horns of their shining helmets touched as they bent their heads, so dense were they formed on each other.'

zzz: For example Webster Roman Imperial Army, pp. 221‑225. Vegetius himself states at I.8, Haec necessitas conpulit euolutis auctoribus ea me in hoc opusculo fidelissime dicere, quae Cato ille Censorius de disciplina militari scripsit, quae Cornelius Celsus, quae Frontinus perstringenda duxerunt, quae Paternus diligentissimus iuris militaris adsertor in libros redegit, quae Augusti et Traiani Hadrianique constitutionibus cauta sunt. Nihil enim mihi auctoritatis adsumo sed horum, quos supra rettuli, quae dispersa sunt, uelut in ordinem epitoma conscribo. 'This requirement [of inquiring after the military system of the Roman People] made me consult competent authorities and say most faithfully in this little book what Cato the Censor wrote on the system of war, what Cornelius Celsus, what Frontinus thought should be summarised, what Paternus, a most zealous champion of military law, published in his books, and what was decreed by the constitutions of Augustus, Trajan and Hadrian. For I claim no authority to myself, but merely write up the dispersed material of those whom I have listed above, summarising it as if to form an orderly sequence.' This seems to justify the view that Vegetius was simply an abbreviator of earlier works. However he should not be derided for it since he admits it himself and it was the express purpose of his work to do so. The problem is therefore to assess from which earlier work(s) each section comes and the degree to which this can be done.

zzz: Cato ille Maior, cum et armis inuictus esset et consul exercitus saepe duxisset, plus se reipublicae credidit profuturum, si disciplinam militarem conferret in litteras. Nam unius aetatis sunt quae fortiter fiunt; quae uero pro utilitate reipublicae scribuntur aeterna sunt. Idem fecerunt alii conplures, sed praecipue Frontinus, diuo Traiano ab eiusmodi conprobatus industria. Horum instituta, horum praecepta, in quantum ualeo, strictum fideliterque signabo. 'Cato the elder, since he was unbeaten in war and as consul had often led armies, thought he would be of further service to the State if he wrote down the military science. For brave deeds belong to a single age; what is written for the benefit of the state is eternal. Several others did the same, particularly Frontinus, who was highly esteemed by the deified Trajan for his efforts in this field. These men's recommendations, their precepts, I shall summarise as strictly and faithfully as I am able.'

zzz: Vegetius I.8 n. 3. However, Milner does argue that 'the suspicion must be that Vegetius' real information was too jejune to be based on the original works by Frontinus or Paternus.' Vegetius and the Anonymous, p. 292. See Below.

zzz: Milner Vegetius and the Anonymous, p. 311. There would still be room for a Tactica to be composed since Frontinus' De Re Militari is unlikely to have taken on that form.

zzz: G. Gundermann 'Quaestiones de Iuli Frontini Strategematon Libris' Fleckeisen Jahrrbucher Supplementband 16 (1888) p. 318. See Frontinus Introduction, p. xx. For the Chattan campaign see Jones Domitian, pp. 128‑131.

zzz: Sherwin-White Letters IV.8.3. Wheeler 'Modern,' p. 12, argues that it was probably composed between 84 and 88. However, if Wheeler's own career reconstruction is correct then Frontinus would have been to busy in office to devote time to literary composition.

zzz: Everett L. Wheeler Stratagem and the Vocabulary of Military Trickery, Supplement to Mnemosyne 108 (New York, 1988), p. 2.

zzz: More recently Campbell 'Teach Yourself,' p. 15, also accepts the fourth book's authenticity.

zzz: An opposite view, to the effect that Frontinus may have written the Strategemata out of a belief in the likely merit of the work itself, is expressed by Campbell, who says 'Teach Yourself,' p. 14, 'it is not true to say that the Strategemata are merely an appendix to this work.' Campbell, referring to Webster The Roman Imperial Army, London, 1969, p. 221 who takes a position similar to Wheeler's, referring to Frontinus' Strategemata as 'an appendix.'

zzz: Frontinus Strategemata Preface 1.1. It may also have stood on its own merits hence its survival.

zzz: All of the prefaces in the genre of military tactical writing that survive contain at least some real statement of purpose or objective. See below.

zzz: Frontinus Strategemata I Preface. This paragraph is considered by some scholars to be interpolated by the author of Book IV. Cf IV Preface.

zzz: Frontinus Strategemata I Preface. This could be the reason that Lydus lists a work On the Office of the General: it and the Strategemata are possibly one and the same, since the Strategemata was for use by men in the office of general.

zzz: Campbell 'Teach Yourself,' p. 14. Campbell cites examples of similar tactics being used: for example, feigned retreats and ambuscades in Josephus Bellum Judaicum II.634‑637, and III.186‑187, p. 24. Also we can use the Cerialis stratagem from the Civilis campaign which is indeed a stratagem in contemporary warfare, even if Frontinus does not report it.

zzz: See above. Ioannes Lydus On Powers or The Magistracies of the Roman State, translated by A. Bandy, (Philadelphia, 1983), De Magistratibus I.47. For the date of composition see p. xxvii‑xxxviii. Such a passage has been used to suggest that Cato and Celsus wrote tactical works. However, the fact that they attest the name veteran does not prove that they did, even if the other authors in Lydus' list are predominantly military tactical writers. Proof of such a tactical work by Cato comes from Vegetius and Fronto and will be discussed below; proof of such a work by Celsus comes in Vegetius also and later in Lydus himself III.33-III.34.

zzz: Frontinus does mention veterans twice in the Strategemata I.3.2 and II.3.7 but does not mention adoratores. Nor do the references to veterans provide the distinction Lydus claims. Also, in the existing corpus of Cato's writings, no reference to veterans exists. It may very well have come from the De Re Militari or some other lost work.

zzz: Wheeler 'Modern,' p. 12 and n. 29. For this juristic genre see F. Schulz Geschichte der römischen Rechtwissenschaft (Weimar, 1961), pp. 309‑315.

zzz: Knowledge of the Latin language had been in decline in the eastern empire since, according to Lydus, the reign of Theodosius II, A.D. 408‑450. De Magistratibus II.12 and III.42. See Introduction p. xix. The only other Latin military writers, if we disregard the passage as proof of Cato's military writing, are Paternus, whose Tactics/De Re Militari is quoted at I.9, Celsus, who is called the Roman tactician at III.33‑34 and who wrote a treatise on attacking the Persians suddenly and by surprise, and Vegetius whom Lydus refers to as following Frontinus in II.47; the only reference to Vegetius. Such a reference may indicate only that Lydus knew that Vegetius followed and used Frontinus, not that Lydus had read Vegetius, or that he had access to him. The other references to Cato at I.2 and I.5, use Cato's number of 439 years between Aeneas' arrival in Italy and the foundation of Rome, and that Romulus was not ignorant of Greek, stated in Cato's Περὶ ῾Ρωμαϊκης Ἀρχαιότητος or [On Roman Antiquities]/Origines. Neither are from the De Re Militari, yet they still display Lydus' Latin erudition. The attestation of the names veterani or adoratores does not, therefore have to have come from Cato's De Re Militari. Occurrences of those words occur elsewhere in Cato's writings in zzz.

zzz: Ioannes Lydus De Magistratibus III.3. We would expect to find a corresponding passage to this in the Strategemata II.3 De Acie Ordinanda 'On the disposition of troops for battle.' Yet when we examine these stratagems we see a great many variations on strict battlefield formation. Frontinus does not stress the position of the overall commander as being in the centre. As regards cavalry for example II.3.3 has cavalry on the right wing, II.3.17‑18 have cavalry in the rear, and II.3.20 has cavalry on both flanks. The only stratagem which demonstrates a predominance of cavalry on the left wing is II.3.22a: Pompey's disposition at Pharsalus. II.3.22b is Caesar's disposition to oppose Pompey whereby he put his cavalry on the right wing mixed with infantry. And we know who won Pharsalus. It would seem unlikely that Frontinus would suggest as stringent a formula of battle-line dispositions as Lydus proposes. Frontinus' Strategemata stresses expediency and adaptability to suit each situation. Therefore Lydus' use of Frontinus in this instance would seem to arise from either a misunderstanding of Frontinus or a lapse in memory. What is more it probably also shows that Lydus had not read the Strategemata since, expediency and adaptability are the work's major lessons and should not have been readily forgettable.

zzz: See especially Wasserversorgung im Antiken Rom IV; Harry B. Evans Water Distribution in Ancient Rome. The Evidence of Frontinus (Michigan, 1994); R. Lanciani I commentarii di Frontino intorno le acque e gli acquedotti, MemLinc. ser. 3.4 (1881; reprint Rome, 1975); and Hodge Roman Aqueducts.

zzz: Hodge Roman Aqueducts, p. 16. Hodge then explains the weaknesses of Frontinus. 'Sometimes his expositions on hydraulics are confused or inaccurate, because they reflect a Roman technical knowledge that was itself incomplete or erroneous, and also that his treatment of the subject has major gaps.' Hodge argues that Frontinus was a bureaucrat, not an engineer, and that he gives the view from head office, not the view of a man on the job.

zzz: That is not to say that the men in the field would not have to change their ways after following years of management which was now being replaced because of its shortcomings.

zzz: Frontinus De Aquis II.118. Southern, Domitian, p. 64, argues that 'Domitian is therefore dismissed and, while not quite labelled a thief, certainly receives no credit for the fact that without his careful attention to detail Nerva would not have had an income to restore to the people.'

zzz: Hodge Roman Aqueducts p. 18. Pliny Epistularum IX.xix.1. Hodge also quotes the epitaph of Christopher Wren's as being appropriate. 'Si monumentum requiris, circumspice.'

zzz: His longevity under such changeable imperial regimes under whom other very capable men became unstuck might be cause to say that Frontinus was a genius of sorts, possibly a genius of survival.

zzz: The Calvisii Rusones were connected to Frontinus, Syme Tacitus, p. 793. There are two possibilities for the acquisition of the names Julius Frontinus by P. Calvisius Ruso. First, he may have received them in Frontinus' will and secondly, and more likely, through maternal ascendance. R. Syme 'P. Calvisius Ruso. One Person or Two?' Zeitschrift für Papyrologie und Epigraphik 56 (1984), pp. 173‑192, at pp. 176‑177, and 192. See also Birley Fasti, p. 72.

zzz: Anthony Birley Marcus Aurelius2 (London, 1987), genealogical table D, and p. 245. See R. Syme 'The Testamentum Dasumii: Some Novelties' Chiron 15 (1985), pp. 41‑63, at pp. 46‑47, 55, and tables I and II, pp. 62‑63. The reason to assume that P. Calvisius Ruso Julius Frontinus' wife was a Dasumia is to explain Scriptores Historiae Augustae Marcus Aurelius I.6. cuius familia in originem recurrens a Numa probatur sanguinem trahere, ut Marius Maximus docet; item a rege Sallentino Malemnio, Dasummi filio, qui Lupias condidit. 'His family, in tracing its origin back to the beginning, established its descent from Numa, or so Marius Maximus tells, and likewise from the Sallentine king Malemnius, the son of Dasummus, who founded Lupiae.' See Table I.

zzz: For Asclepiodotus see the Loeb Introduction on Asclepiodotus; and Posidonius Volume I, The Fragments2, Cambridge Classical Texts and Commentaries 14 (Cambridge, 1989); and Volume II, The Commentary (i) Testimonia and Fragments 1‑149, Cambridge Classical Texts and Commentaries 14A (Cambridge, 1988), L. Edelstein and I. Kidd (editors), pp. 31‑33.

zzz: We have no information at all about this work nor the circumstances of its composition. This does give a second sophistic connection for the writing of Greek military treatises not just Tactica.

zzz: B. van Groningen 'General Literary Tendencies in the Second Century A. D.' Mnemosyne 18, pp. 41‑56. 'It is an essentially weak literature. Weak in its intellectual and emotional elements,' p. 51. 'Such a literature has no single valuable purpose,' p. 52. He attributes this weakness to anything that looks for inspiration to the past. He argues, p. 56, that the neglected second century deserves the neglect to which it has been subjected.

zzz: George Kennedy 'Sophists as Declaimers' in G. W. Bowersock (editor) Approaches to the Second Sophistic, Papers Presented at the 105th Annual Meeting of the American Philological Association, (University Park, 1974), p. 17. See also Graham Anderson The Second Sophistic (London and New York, 1993), p. 9.

zzz: E. Bowie 'Greeks and Their Past in the Second Sophistic' chapter VIII in M. Finley (editor) Studies in Ancient Society (London, 1974), pp. 166‑209, at p. 166. Authors of the second century tried to write in the Attic dialect according to classical models like Thucydides, Plato, and Demosthenes. The second sophistic has been considered from this linguistic perspective alone. Bowie 'Greeks and their Past,' p. 166. See Anderson Second Sophistic, pp. 86‑100, H. Rose A Handbook of Greek Literature (London, 1934), pp. 396‑421, A. Lesky A History of Greek Literature (London, 1966), pp. 829‑845, Cambridge Ancient History XI, XVII II 'Atticism and the Second Sophistic Movement,' pp. 678‑690, and B. van Groningen 'General Literary Tendencies,' p. 49.

zzz: For example C. P. Jones Culture and Society in Lucian (London and Cambridge, Massachusetts, 1986). Bowersock Greek Sophists examines the sophists from the point of view of Philostratus, and even Anderson Second Sophistic examines individual authors.

zzz: Philip De Lacy 'Plato and the Intellectual Life of the Second Century A. D.' in Bowersock Approaches, p. 4. Also Anderson Second Sophistic, p. 8.

zzz: Aelian Tactica Preface 6. Aelian Tactica Introduction, p. 31. See also McCartney, Review, p. 183, for Aelian's freely confessed antiquarian interest.

zzz: See Polyaenus Book IV.3 and Preface I and IV. On Polyaenus see J. Melber 'Über die Quellen und den Wert der Strategemensammlung Polyäns' Jahrbücher für classische Philologie 14, Supplement, (Leipzig, 1885), pp. 417‑688; Rolly Janet Phillips The Sources and Methods of Polyaenus (Cambridge, Massachusetts, 1970); and F. Martín García Lengua, Estilo y Fuentes de Polieno (Dissertation, Madrid, 1980); Polyaenus Stratagems of War, translated by R. Shepherd, (London, 1793), reprinted (Chicago, 1974); Polyaenus Stratagems of War, 2 Volumes, translated by Everett L. Wheeler and P. Krentz, (Chicago, 1994); P. Stadter Plutarch's Historical Methods: An Analysis of the Mulierum Virtutes (Cambridge, Massachusetts, 1965); Polyaenus Strategematon Libri VIII et Incerti Scriptoris Byzantini Saeculi X. Liber de re Militari, edited by E. Woelfflin, J. Melber, and R. Vari, (Stuttgart, 1887/1901, reprinted 1970); and Pauly-Wissowa RE II.21.2, Polyainos (8).

zzz: Flavius Arrianus TECNH TAKTIKA and EKTAXIS KATA ALANWN, translated by J. De Voto, (Chicago, 1993), Introduction, p. ii.

zzz: Arrian Tactica 32.3. Bowie 'Greeks and their Past,' p. 192, argues that Arrian's Tactica reflects a harmony of the Greek past with the Roman present.

zzz: Stadter 'The Ars Tactica of Arrian,' p. 126. 'Arrian neither retreats into his Greek heritage nor flees from it; proudly and consciously he incorporates it into his active life as a Roman imperial official.' See Stadter Arrian, p. 49, for the combination of Arrian's practical and antiquarian interests.

zzz: A reason for the possible esteem of Arrian will be examined below. Polyaenus of Macedon, whose Strategemata is the only other to survive from antiquity, does not try and assert the superiority of his Strategemata over that of Frontinus because, in addressing his work to the emperors Marcus Aurelius and Lucius Verus, such a move might have only served to weaken his chances of getting their attention. Domitia Lucilla was still alive in 155, but was dead by Marcus' accession to the throne in 161 (Birley Marcus, p. 107). Therefore to deride a member of her family would have been an unheard of social and political faux pas. Polyaenus actually seems to avoid all cause for comparison between himself and Frontinus, focusing on classical Greek history and organising his eight books according to protagonist, unlike Frontinus who focused on Roman history and organised his work according to subject.

zzz: Laelius married Pompeia Sosia Falconilla, Frontinus' great-great-grand daughter. See McDermott 'Stemmata Quid Faciunt?,' pp. 238‑240.

zzz: Aelian Tactica Preface 4‑5. Aelian's Tactica is the fullest surviving. Aelian Introduction p. 31. Indeed Oliver Spaulding, himself a colonel in the US Army, holds Aelian's treatise in some esteem ('Ancient Military Writers' Classical Journal 28 (1932), pp. 657‑669, at p. 664). He argues that 'it is the earliest attempt at a study, not of tactics proper, but of the history of tactics.'

zzz: Aelian may have considered the treatise too petty and insignificant for an experienced commander like Trajan.

zzz: See Dio Cassius LXVIII.29.1. ' "pantwV an kai epi touV IndouV, ei neoV eti hn, eperaiwqhn." IndouV te gar enenoei, kai ta ekeinwn pragmata epolupragmonei, ton te Alexandron emakarize. kaitoi elege kai ekeinou peraiterw prokecwrhkenai, kai touto kai th boulh epesteile. . . ' "I should certainly have crossed over to the Indi, too, if I were still young." For he began to think about the Indi and was curious about their affairs, and he counted Alexander a lucky man. Yet he would declare that he himself had advanced farther than Alexander, and wrote in those terms to the senate, . . .'

zzz: Edward Gibbon The History of the Decline and Fall of the Roman Empire (Volume I, London, 1764; D. Womersley (editor) (Harmondsworth, 1995), p. 35.

zzz: For scholarship on Arrian's Tactica see Stadter 'The Ars Tactica of Arrian,' pp. 117‑128; Everett L. Wheeler 'The Occasion of Arrian's Tactica' Greek, Roman, and Byzantine Studies 19 (1978), pp. 351‑365. Also P. Stadter Arrian of Nicomedia, pp. 41‑45. See also Flavius Arrianus TECNH TAKTIKA Introduction, pp. i‑iii.

zzz: Arrian Tactica 32.2‑3. Wheeler's argument that this Roman infantry treatise is the first half of the Tactica are flawed (Wheeler 'The Occasion of Arrian's Tactica,' p. 356). The terms of the infantry treatise seem to point to a description of Hadrian's military reforms. This view is based on the argument that Arrian's description of the cavalry exercises was a description of those reforms (Wheeler 'The Occasion of Arrian's Tactica,' p. 357). These cavalry reforms are praised because of Hadrian's incorporation of foreign practices such as the mounted archers of the Parthians and Armenians, and the light spear cavalry of the Sauromatai and Celts. Arrian Tactica 44.1. However, for Wheeler ('The Occasion of Arrian's Tactica,' pp. 357‑358) to argue that this cavalry 'exercise' was simply a description of the armatura or ludi castrenses to celebrate Hadrian's 20th regnal year would mean that it would have had no purpose in a tactical work, and there would be no reason for Arrian to claim that the Tactica had a practical purpose or to include reforms.

zzz: A. Dent 'Arrian's Array' History Today 24.8 (1974), pp. 570‑574, at p. 573 argues that Arrian wrote the Ectaxis in Greek because it was the one language that all his troops would understand. The same can be said for Arrian's Tactica, perhaps reinforcing its practical nature.

zzz: Everett L. Wheeler 'Flavius Arrianus: A Political and Military Biography' Dissertation Abstracts International 39 (1978), p. 400. Stadter Arrian, p. 43 n. 35 (p. 207) cites F. Kiechle 'Die 'Taktik' des Flavius Arrianus' Bericht der römische-germanischen Kommission des Deutschen Archäologischen Instituts 45 (1965), pp. 87‑129, who argues that Hadrian revived the phalanx for use against the Sarmatians and Parthians' heavy cavalry. Bosworth 'Arrian and the Alani,' pp. 242‑243, has rejected this view because of the defensive nature of Arrian's formation and the attacking purpose of the phalanx. See below.

zzz: Even if the relationship is not strong, as some historians have argued, it does not prevent the Tactica from being practical since Arrian admits this anyway. It is only the degree to which the Tactica is practical in a contemporary sense that is affected.

zzz: Stadter Arrian, p. 46, and B. Bachrach A History of the Alans (Minneapolis, 1973), p. 8 state that they are heavy cavalry armed with lances whilst Dent 'Arrian's Array,' p. 571, and Bosworth 'Arrian and the Alani,' p. 244, state that they were mounted archers.

zzz: Arrian Ectaxis 26. Here Arrian refers to the Alani by their ancient name the Scyths.

zzz: Arrian has his cavalry behind the phalanx to do this and prevent an outflanking manoeuvre. Bachrach History, p. 8, notes that the favourite stratagem of the Alans was the feigned retreat. Arrian seems aware of this and this is probably where Bachrach got his argument from.

zzz: Arrian Ectaxis 11‑12, 14, 25, and 30. Bosworth 'Arrian and the Alani,' p. 233.

zzz: At the battle of Gabiene in 316 BC Eumenes' phalanx withstood a cavalry charge from approximately 4500 of Antigonus' cavalry. H. Delbruck The History of the Art of War, Volume I (Antiquity), translated by W. Renfore, (London, 1975), p. 240; Diodorus of Sicily XIX.43.5. Arrian himself recorded in his Events after Alexander that after Alexander's death a Macedonian phalanx opposed a cavalry charge by drawing up its ranks into close formation and unleashing a deadly missile barrage. W. Goralski 'Arrian's Events After Alexander: Summary of Photius and Selected Fragments' The Ancient World 19 (1989), Section 9: Eumenes Battles Neoptolemus and Craterus, PSI XII 1284, p. 95. See also Stadter Arrian, p. 48 n. 49 (p. 208). Alexander himself had used missiles to repulse enemies. Arrian Anabasis IV.4.4‑5. Arrian himself may reflect his fascination with Alexander in his use of phalanx tactics, but they had proven successful and so their adoption was completely justified. So too, Trajan may have adopted Alexander's tactics against the Parthians since they were successful.

zzz: Everett L. Wheeler 'The Legion as Phalanx' Chiron 9 (1979), pp. 303‑318.

zzz: Wheeler argues, 'Flavius Arrianus,' p. 400, that the 'Roman phalanx' appeared in Roman military history increasingly frequently from the late Republic on. No doubt this was expanded in the thesis proper.

zzz: Plutarch Crassus 23.3, cf Onasander Strategikos 21, Aelian Tactica 37.1, Asclepiodotus Tactica 11.2‑3, 10, 22. Wheeler 'Legion,' pp. 307‑308. See Dent 'Arrian's Array,' p. 570.

zzz: De Bello Africo 15 and 17. Wheeler argues, 'Legion,' p. 307, that by the time of Caesar testudo had come to mean, 'not only the closely packed formation for assault on city walls, but also a tightly massed defensive disposition.' Labienus' cavalry forced Caesar's men to close ranks and form a defensive position until Caesar made his cohorts form two lines of defence facing both directions from behind which Caesar was able to make attacks with volleys of missiles and thus put the enemy cavalry to flight. Such a formation, a densely packed line of infantry able to withstand a cavalry barrage and facilitate such a counter attack, is very similar to the function of Arrian's formation in the Ectaxis.

zzz: Dio Cassius XL.22.2, Plutarch Mark Antony 45.2‑3. See Wheeler 'Legion,' p. 308. This is of interest because Frontinus reports this stratagem of Antony's: Strategemata II.3.15. M. Antonius adversus Parthos, qui infinita multitudine sagittarum exercitum eius obruebant, subsidere suos et testudinem facere iussit, supra quam transmissis sagittis sine militum noxa exhaustus est hostis. 'When Mark Antony was engaged in battle with the Parthians and these were showering his army with innumerable arrows, he ordered his men to stop and form a testudo. The arrows passed over this without harm to the soldiers, and the enemy's supply was soon exhausted.'

zzz: Wheeler 'Legion,' p. 310. See Dio Cassius LXII.8.2‑3 where he describes the Roman formation as drawn up in three phalanges. Cf Tacitus Annales XIV.34. Wheeler argues that Tacitus' vagueness 'leaves open the question of a phalangite formation.' See Parker Roman Legions, p. 258.

zzz: Tacitus Annales XIV.34, Agricola xxxv.1‑2. Wheeler 'Legion,' pp. 310‑311. Tacitus' reason for this formation, having the auxilia absorb most of the fighting so as to avoid spilling Roman blood, is rejected by Parker Roman Legions, p. 258.

zzz: XV was also taken on Trajan's Parthian war, Parker Roman Legions, p. 159. See Dio Cassius LV.23. If Frontinus did serve in some capacity in or as commander of XV Apollinaris in either the Parthian campaign of Corbulo or in the Jewish war, quite possibly he may have had some part in its acquiring phalanx tactic skills.

zzz: Bosworth 'Arrian and the Alani,' pp. 244‑245. See Wheeler 'The Occasion of Arrian's Tactica,' p. 357.

zzz: Bosworth 'Arrian and the Alani,' p. 244, and Lepper Trajan's Parthian War, p. 176.

zzz: Wheeler 'Legion,' p. 313, argues that Trajan faced 'the same basic style of warfare as Arrian foresaw encountering against the Alani, a style for which the phalanx offered the best means to employ the heavy infantry of the legions.'

zzz: Wheeler 'Legion,' p. 313. However, see above for Vegetius' inclusion of Trajan only to please Theodosius I.

zzz: Dio Cassius LXVIII.23.1. Cf Aelian Tactica 36‑37, Asclepiodotus Tactica 11, and Arrian Tactica 28‑29. See Wheeler 'Legion,' p. 313. Wheeler argues that this reference might be out of place because Trajan was probably already interested in phalanx tactics before the Parthian campaign.

zzz: It was argued above that Frontinus was probably in contact with Domitian during the latter's Dacian campaign and may even have had some input into its strategy. Such a possibility adds strength to the tenor of Frontinus' possible discussion about Dacia with Trajan, since Frontinus probably had some experience of warfare in that province. The practical applicability of phalanx tactics in Dacia may well have been a subject.

zzz: The use of an Alexander fascination by later emperors for tactics in their Parthian campaigns will be discussed below.

zzz: Although Trajan arrived after the fighting had stopped, he still commanded legio VII to bring it from Spain.

zzz: However, we have seen that Parker (Roman Legions, p. 258), rejects the view that Corbulo used phalanx tactics. Given the lack of a full account of Corbulo's campaign, we may not presume to make such an outright rejection.

zzz: Polyaenus has the express purpose of assisting Marcus Aurelius and Lucius Verus in this war. Polyaenus Strategemata I Preface 1 and 13.

zzz: Dio Cassius Epitome of LXXII.7.2‑4. See also Campbell 'Teach Yourself,' p. 29. Dio speaks of the Romans forming a compact body.

zzz: Bosworth 'Arrian and the Alani,' p. 246, referring to Dio Cassius LXIX.15.1.

zzz: He may also have written a poliorcetic work separate from the Tactica which is referred to by Athenaeus Mechanicus at 6.1, and 31.6‑10. See Wheeler Stratagem, p. 13 -14, especially note 40.

zzz: I.1.1, I.2.5, II.4.4, II.7.14, III.1.2, III.10.1, IV.1.16, IV.1.33, IV.3.1, IV.7.12, IV.7.31, and IV 7.35. Some of these stratagems could come from histories of Cato's conduct since the stratagems are repeated in Plutarch, Appian, Dionysius, Zonaras, Valerius Maximus, Pliny the Elder, and Livy. Especially I.1.1, I.2.5, II.4.4, IV.1.33, IV.3.1, IV.7.31, and IV.7.35. They might just as equally have come from Cato's De Re Militari, in which he most probably included anecdotes from his own experience.

zzz: Strategemata IV.1.16 is however listed as a fragment of Cato's De Re Militari in M. Catonis Praeter Librum De Re Rustica Quae Exstant H. Jordan (editor) (Stuttgart, 1967), pp. 80‑82. None of the other 11 Cato stratagems in Frontinus is listed.

zzz: However, see now the conspectus locorum in Ireland Strategemata. Kappelmacher Iulius 243, cols. 599‑600. Kappelmacher refers to the conspectus locorum in Gundermann Iuli Frontini Strategematon Libri IV (Leipzig, 1888). This has now been superseded by Ireland.

zzz: However, we must use more primitive methods of estimation where it comes to stratagems which do not have an extant source, especially in the case of Greek collections of which none survive before Polyaenus.

zzz: Webster Roman Imperial Army, pp. 221‑225. Webster argues that the coherent passages are from Frontinus.

zzz: Chapter 7 is an analysis of Vegetius' sources which is summarised below.

zzz: Milner Vegetius and the Anonymous, pp. 311‑312. Frontinus' placement of the fortress at Isca seems to follow the castramentation rules of Vegetius and we may therefore see in that Frontinus' practical application of such rules — something he certainly would have included in his De Re Militari. Given the large number of camps Frontinus constructed in Wales he would have tested every aspect of Vegetius' castramentation chapters — the camp should be able to control, not be controlled by nearby hills or mountains, it should not be liable to flooding, and appropriate to the size of the units occupying it.

zzz: There are some Roman women's stratagems included in the illustrious women's stratagems. Cloelia VIII.31, Porcia VIII.32, and Epicharis VIII.62. The stratagem of Rhome is included in VIII.25.

zzz: There is no need to assume with Wheeler, Polyaenus Introduction, p. ix, that Polyaenus was Bithynian and simply pretending to be Macedonian.

zzz: The only stratagem which post dates Augustus comes from the illustrious women's stratagems concerning Epicharis' defiance of Nero's torture in A.D. 65. VIII.62

zzz: In an inexplicable mistake, the first translator of Polyaenus into English, Richard Shepherd, Archdeacon of Bedford, in 1793 remarked that Polyaenus was regarded in so high a light as a military writer and his Strategemata esteemed of so great utility 'that Frontinus, a Roman knight, stimulated by the reputation it had obtained the author, published a new performance, of the same nature, and under the same title of military stratagems.' Shepherd Polyaenus Stratagems of War; Preliminary Discourse, p. xix. Shepherd then stated that, in accordance with Casaubonus, a comparison between the two collections would strengthen the argument for the superiority of Grecian over Roman writers. Such an opinion is certainly not held today in the case of Polyaenus and Frontinus. Despite this completely uninformed opinion of Shepherd's which, one presumes, would have been patently obviously wrong, even in 1793, it records one note of interest, that being that Frontinus is considered as equestrian.

zzz: J. Melber 'Über die Quellen,' p. 662. 'Mit aller Entschiedenheit muß also die Ansicht aufrecht erhalten werden, daß Polyän, abgesehen von Suetons Leben des Cäsar und Augustus, keine römischen Quellen benutzt hat.' 'The view that Polyaenus used no Roman sources apart from Suetonius' lives of Julius Caesar and Augustus must be maintained with full rigour.'

zzz: Phillips Sources and Methods, p. 193 n. 1. She earlier says 'the overwhelming predominance of Greek material in Polyaenus is largely explained by his lack of familiarity with the Latin language,' p. 30. See also Wheeler 'Modern,' p. 12.

zzz: E. Champlin Fronto and Antonine Rome (London and Cambridge, Massachusetts, 1980), p. 42.

zzz: Phillips Sources and Methods, p. 31. Phillips' techniques of compilation assigned to Polyaenus are probably also wrong. See below.

zzz: Martín García Polieno, p. 1129‑1141. Phillips Sources and Methods, p. 193, assumes that these intermediary sources were earlier collections used by both Frontinus and Polyaenus although this seems to contradict her statements throughout the work that Polyaenus used Greek historians directly and that, on p. 194, she argues that Frontinus used collections more frequently than Polyaenus.

zzz: Phillips Sources and Methods, p. 194. Valerius Maximus only comes in as a major source for Book IV. See Frontinus Introduction p. xxii especially nn.2‑3.

zzz: For example Polyaenus VIII.3.2 and Frontinus II.5.1; VIII.5 and II.7.1, VIII.6 and I.1.4; VIII.7.1 and IV.4.1; VIII.10.2 and II.4.6; VIII.10.3 and II.2.8; VIII.11 and III.3.2; VIII.16.1 and II.1.1; VIII.16.2 and IV.1.1; VIII.16.4 and IV.1.5; VIII.16.6 and II.11.5; VIII.16.7 and II.7.4; VIII.17 and I.1.1; VIII.21 and IV.7.22; VIII.22 and I.11.13; VIII.23.4 and II.1.16; VIII.23.7 and III.17.6; VIII.23.15 and IV.5.2; VIII.23.16 and II.8.13; VIII.23.25 and IV.7.32. Whilst these may not all come from the same source to assume that where they do Frontinus preferred to follow a Greek collection is just silly.

zzz: This can be shown simply in that Valerius Maximus has a short collection of 7 Roman stratagems in Book VII.4 which Polyaenus follows closely in his Roman stratagems. He copies the first two not only in content but also in order, and of the seven Polyaenus uses five. However, Polyaenus omits or changes some of the details given by Valerius here. This should not be seen as a sign of a different source but rather that the interpretation of Polyaenus as a slavish copier of earlier sources is wrong. He had a discernible editorial technique and often used his own opinions on anecdotes.

zzz: Not just Phillips is guilty of this. The authors who deride him as an inferior historical source also miss the purpose in his writing in the first place. G. Maslakov 'Valerius Maximus and Roman Historiography. A Study of the exempla Tradition' Aufsteig und Niedergang der Romischen Welt II.32.1 (1984), pp. 437‑496, at p. 457, argues of Valerius that, like Polyaenus, he can sometimes be the only source for an historical event but this is coincidental to his purpose and aims. Phillips Sources and Methods, pp. 204‑211 discusses Polyaenus' main sources; Ephorus, Thucydides, Philistus, Timaeus, Hieronymus, Herodotus, and Plutarch. He also used Frontinus and Suetonius and other previous collections such as Valerius Maximus. A source that Phillips seems to have neglected is Appian whom Polyaenus has a close relationship to. See Polyaenus Introduction p. xv., Campbell 'Teach Yourself,' p. 16 n. 14. However, Maslakov 'Valerius,' pp. 457‑458, argues that an analysis of sources is unwarranted because the elucidation of the aims of the author and his shaping of material must be taken into account also. The same can be said of Polyaenus; Phillips does not anywhere consider the military tactical genre, let alone see it as the important and chosen genre of Polyaenus.

zzz: Phillips Sources and Methods, p. 1. Whilst Polyaenus may be of use to us today for this reason it is not his chief value which instead lies in his being the only Greek stratagem collection to have survived and as such gives us a glimpse into the survival of Greek military literature and, what is more, the second sophistic's adaptation of such literature.

zzz: We can repeat that it is McDermott's main point that there is only one senatorial family which can be traced from the Julio-Claudian dynasty through to the Severan — seven generations, and that is the family of Sextus Julius Frontinus. McDermott 'Stemmata Quid Faciunt,' p. 229.