^{page 1 note 2} Cf. BJ 41, 6: nobilitas factione magis pollebat, plebis vis soluta atque dispersa in multitudine minus poterat; Cic., Rep. I, 44Google Scholar; 69; III, 44 and esp. 23: cum autem certi propter divitias aut genus aut aliquas opes rem p. tenent, est factio, sed vocantur optimates. Alluding to the doctrine that amicitia subsists only between boni, Sallust insinuates that the self-styled boni merely form a factio; these men are the nobles (cf. BJ 40–1; Hist. I, 12Google Scholar), between whom and the people the state is torn (cf. Cic., Sest. 96Google Scholar); neither he nor Cicero knows of factions of the Metelli, etc., within the nobility. See also Hellegouar'ch 99 ff.; Henderson, M. I., JRS XLII, 115Google Scholar.

^{page 1 note 3} Party Politics in the Age of Caesar, 7–8.

^{page 1 note 4} Roman Revolution, 157; cf. ch. II.

^{page 1 note 5} Amic. 19–32; cf. 51.

^{page 1 note 6} Ibid. 18–20; 50; 65; 79–84; cf. Arist., EN 1156b 6 ff.Google Scholar

^{page 2 note 1} Schulz, F., Principles of Roman Law, 237Google Scholar, aptly cites Verr. II, 3, 122 and 152Google Scholar.

^{page 2 note 2} For mores cf. Nepos, , Att. 5, 3Google Scholar; though Atticus' sister married Quintus Cicero, he was more intimate with Marcus, ut iudicari possit plus in amicitia valere similitudinem morum quam affinitatem. Common studio, v. infra; cf. also Cluent. 46: iam hoc fere scitis omnes quantam vim habeat ad coniungendas amicitias studiorum ac naturae similitudo; examples, Lig. 21 (Tubero); Fam. I, 7, 11Google Scholar; 9, 23–4 (Lentuli); III, 13, 2 (App. Claudius); V, 13, 5; 15, 2 (Lucceius); XIII, 29, 1 (Plancus); de fato 2 (Hirtius); Ac. praef. and I, 1 (Varro). Hellegouar'ch, 174 ff., misses the common sense of intellectual pursuits. Frankness, , Fam. XI, 28, 8Google Scholar. Courtesy in complaints, III, 11, 5. Image: Lossmann, 33 ff., compares Arist., MM 1213 a 7 ff.Google Scholar Second self: Fam. VII, 5, 1 (Caesar)Google Scholar; Att. III, 15, 4 (Atticus)Google Scholar; ad Brut. 23 ( = 1, 15), 2 (Brutus); Att. IV, 1, 7Google Scholar (Pompey on Cicero); cf. Lossmann, loc. cit.

^{page 2 note 3} Kultur der ciceronischen Zeit, I, 55 ff.Google Scholar

^{page 2 note 4} RE VII A, 1163 ff.Google Scholar (Philippson).

^{page 2 note 5} 1, 3, 11; contra Philippson (1164) he clearly held that Cicero used Theophrastus, but he need not be right.

^{page 2 note 6} Cf. Pöschl, V., Röm. Staat u. gr. Staatsdenken bei Cicero, 23, n. 27Google Scholar on the de Republica.

^{page 2 note 7} Panaetius, the main source of 1–11 (cf. III, 7: quem nos correctione quadam adhibita potissimum secuti sumus; Att. XVI, 11, 4Google Scholar), is criticized explicitly in I, 7; 9–11; 152; 161; II, 16; 86; III, 33.

^{page 2 note 8} Amic. 18–19; 21.

^{page 3 note 1} de invent. II, 167Google Scholar; cf. 157.

^{page 3 note 2} EN 1156b 16 ff. It is characteristic of one difference between Greek and Roman society that Cicero does not bring up Aristotle's point that friendship for utility is proper to the business man.

^{page 3 note 3} Rep. I, 7; 13; Fam. IV, 4, 4; Brut. 308, etc.; for later studies in leisure Ac. I, 11; Att. II, 16, 3, etc. For other Romans see Zeller, E., Phil. d. Griechen III I⁴, 550–6Google Scholar with numerous references, esp. Ac. II, 4 f. (Lucullus); Fam. IV, 3, 3, cf. 2, 2 (Ser. Sulpicius); XV, 4, 16 (Cato, M., whom in Fin. III, 7Google Scholar Cicero depicts in Lucullus' library multis circumfusum Stoicorum libris).

^{page 3 note 4} RE VII A, 995Google Scholar.

^{page 3 note 5} Amic. 26. Cf. Hellegouar'ch, 146 f., for amor as equivalent to amicitia, e.g. Att. XIV, 13B, 1Google Scholar.

^{page 3 note 6} Tusc. Disp. I, 1–3Google Scholar; cf. Rep. III, 7Google Scholar; Leg. II, 62Google Scholar.

^{page 3 note 7} Rep. I, 13Google Scholar; Leg. III, 14Google Scholar.

^{page 4 note 1} Rep. I, 13Google Scholar; III, 4–6. Hence it is a public service to expound Greek doctrines, ND I, 7Google Scholar; Divin. II, 1–4Google Scholar; Fin. I, 10Google Scholar; Offic. I, 155Google Scholar.

^{page 4 note 2} Rep. I, 33Google Scholar; ND I, 7Google Scholar; Offic. III, 5Google Scholar.

^{page 4 note 3} Rosc. Am. III.

^{page 4 note 4} E.g. Sall., Cat. 20, 4Google Scholar: idem velle atque idem nolle, ea demum firma amicitia est; Fam. V, 2, 3Google Scholar; V, 7, 2; XI, 28, 1 (Matius refers to perpetua benevolentia); Planc. 5, etc. Cf. Hellegouar'ch, 181 ff.; but his view (183): ‘Voluntas est done le terme propre à désigner la notion d'“opinion politique”’ is at once too narrow and too weak; voluntas connotes the will to realize one's opinions and they need not be political.

^{page 4 note 5} On these and other such terms see Hellegouar'ch's analyses, 68 ff. Consuetudo could naturally be hampered by separation, cf. Fam. XV, 14, 2Google Scholar and p. 5 on Matius.

^{page 4 note 6} Fam. III, 1, 1Google Scholar (Appius dear to Cicero propter multas suavitates ingenii, officii, humanitatis tuae); 10, 9–10; 11, 13, 2.

^{page 5 note 1} Fam. V, 7Google Scholar; cf. Fin. III, 8Google Scholar on Lucullus mecum et amicitia et omni voluntate sententiaque coniunctus; de orat. I, 24Google Scholar: M. Antonius, homo et consiliorum in re p. socius et summa cum Crasso familiaritate coniunctus; Fam. XII, 15, 2Google Scholar (Lentulus writes): homo mihi cum familiaritate tum etiam sensibus in re p. coniunctissimus (Hellegouar'ch, 70, misinterprets both the last texts); Planc. 95, etc.

^{page 5 note 2} Att. I, 16, 11Google Scholar (cf. 17, 10); de domo 28; Fam. III, 10, 10Google Scholar; Att. VIII, 3, 2Google Scholar.

^{page 5 note 3} Fam. II, 12, 1Google Scholar; VIII, 3, 1; XII, 18, 2; IX, 15, 1–2.

^{page 5 note 4} For Trebatius cf. also Att. IX, 9, 4Google Scholar; X, 11, 4; Fam. VII, 19–20Google Scholar; XI, 27, 1; Topica 1–5.

^{page 5 note 5} Cf. Fam. XV, 14, 2Google Scholar.

^{page 5 note 6} Att. I, 17, 5–6Google Scholar; V, 18, 3; VIII, 6, 4; Fam. VII, 30, 2Google Scholar.

^{page 6 note 1} Att. III, 15, 4Google Scholar; cf. p. 2, n. 2.

^{page 5 note 2} Amic. 15; 79; 103–4.

^{page 5 note 3} Catullus 109; cf. 76, 3; 87, 3. Many parallels for such foedera can be found in TLL VI, 1004–6Google Scholar. Cf. Hellegouar'ch, 38 ff., for foedus being cognate with fides, on which amicitia rested (Amic. 65).

^{page 7 note 1} Cf. F. Schulz, Principles of Roman Law, ch. XI (of which Professor P. W. Duff reminded me); Gelzer, M., Kl. Schr. I, 70 ff.Google Scholar

^{page 7 note 2} Cf. de invent. II, 168Google Scholar: amicitiarum autem ratio, quoniam partim sunt religionibus iunctae, partim non sunt, et quia partim veteres sunt, partim novae, partim ab illorum, partim ab nostro beneficio profectae, partim utiliores, partim minus utiles, ex causarum dignitatibus, ex temporum opportunitatibus, ex officiis, ex religionibus, ex vetustatibus habebitur.

^{page 7 note 3} Att. I, 18, 1Google Scholar. The qualities of his friendship with Appius listed on p. 4 above were ‘domestica’ (Fam. III, 10, 9Google Scholar).

^{page 7 note 4} The New Eng. Dict. (1901), IV, 545–6Google Scholar, recognizes that ‘friend’ may be ‘applied to a mere acquaintance, or to a stranger, as a mark of goodwill or kindly condescension on the part of the speaker’, and that it may mean ‘one who is on good terms with another, not hostile or at variance’.

^{page 8 note 1} Fam. XIII, 71Google Scholar: etsi omnium causa quos commendo velle debeo, tamen cum omnibus non eadem mihi causa est.

^{page 8 note 2} Gelzer, M., Kl. Schr. I, 102 ff.Google Scholar

^{page 8 note 3} Ibid. 164 f.

^{page 9 note 1} Fam. XIV, 2, 2Google Scholar; Ibid. I, 9, 9.

^{page 9 note 2} Ibid. V, 8, 4; XIII, 16, 1; Qu. fr. II, 8, 2Google Scholar; Plut., Cr. 13Google Scholar.

^{page 10 note 1} Prov. Cons. 25 ff. (cf. Gelzer, , Kl. Schr. I, 165 ff.Google Scholar), esp. 40; Qu. fr. II, 14, 1Google Scholar; Fam. I, 9, 12Google Scholar.

^{page 10 note 2} Cat. IV, 9Google Scholar; Att. II, 1, 6Google Scholar; 18, 3; 19, 4–5; Prov. Cons. 41, cf. Lossmann's book and my review (p. 1, n. 1).

^{page 10 note 3} Correspondence with Caesar, , Att. VII, 22, 3Google Scholar; VIII, 2, 1; 3, 2; II, 5; IX, 11 A; 11 A; 16, 2; their meeting, IX, 18. Relations with Balbus and Oppius, VII, 3, 11; VIII, 15 A; IX, 7A and B; 13A; X, 18, 2; with Trebatius (a quo me unice diligi scio), VII, 17, 3–4; Fam. IV, 1Google Scholar; Att. IX, 9, 4Google Scholar; 12; 15, 4 and 6; 17, 1; X, 1, 3; 11, 4 (still in early May vir plane et civis bonus); with Matius, IX, 11, 2 (and often linked with Trebatius); Fam. XI, 27, 3Google Scholar; with Caelius, , Fam. VIII, 15–16Google Scholar; Att. X, 9AGoogle Scholar; with Antony who professed the warmest affection, X, 8A; 10, 1; XI, 7, 2, and visited him, X, 11, 4, as did Curio, X, 4, 8, cf. 16, 3. In Pompey's camp he received a letter from his son-in-law, Dolabella, written in Caesar, 's, Fam. IX, 9Google Scholar.

^{page 10 note 4} Fam. XI, 27, 4 (Matius)Google Scholar; 29, 2 (Oppius); XV, 21, 2 (Trebonius); for Antony, Balbus, Oppius, Hirtius, Pansa, Trebatius, Vatinius cf. Att. XI, 5, 4Google Scholar; 6, 3; 7, 2; 8, 1; 9, 2; 14, 1; 18.

^{page 10 note 5} Fam. XI, 27–8 (Matius)Google Scholar; 29, cf. Att. XVI, 12, 1 (Oppius)Google Scholar; Fam. VII, 19, 1Google Scholar; XI, 27, 1; Top. 1 ff. (Trebatius). Fam. VI, 12, 2 (46 B.C.)Google Scholar: omnis Caesaris familiaris satis opportune habeo implicatos consuetudine et benevolentia sic ut, cum ab illo discesserint, me habeant proximum. hoc Pansa, Hirtius, Balbus, Oppius, Postumius plane ita faciunt ut me unice diligant, cf. IX, 16, 2 (their amor is genuine); VI, 10, 2; 14, 3. References to his teaching rhetoric to Hirtius and Pansa are particularly common. In 44 he hoped to turn Hirtius into an optimate, Att. XIV, 20, 4Google Scholar; 21, 4; but for distrust of Hirtius, Balbus and Matius at this time see XIV, 1, 1; 21, 2; 22, 1; XV, 2, 3.

^{page 11 note 1} Att. X, 8 BGoogle Scholar, written after the interview from which Cicero concluded hunc me non amare (IX, 18, 1).

^{page 11 note 2} Cf. Att. VIII, 9, 4Google Scholar.

^{page 11 note 3} Fam. XI, 3Google Scholar, esp. at 4: nos in hac sententia sumus, ut te cupiamus in libera re p. magnum atque honestum esse, vocemus te ad nullas inimicitias, sed tamen pluris nostram libertatem quam tuam amicitiam aestimemus.

^{page 12 note 1} Prov. Cons. 18 ff.; cf. Flacc. 2; Post Red. Quir. 23; and see Fam. V, 4, 2Google Scholar and Post Red. Sen. 25 for Metellus Nepos' quarrel and reconciliation with Cicero.

^{page 12 note 2} E.g. Att. I, 19, 6Google Scholar; 20, 3; II, 1, 7; 16, 2; III, 7, 2; IV, 1, 8; 5, 1; Fam. I, 9, 5Google Scholar. Prov. Cons. 19 distinguishes alieni from inimici, Fam. I, 9, 17Google Scholar from amici.

^{page 12 note 3} Dom. 29; Verr. II, 5, 182Google Scholar; Mur. 45; Fam. III, 10, 6Google Scholar. In 56 Pompey spoke to Cicero of Curio, Bibulus and ceteris suis obtrectatoribus and of nobilitate inimica (Qu. fr. II, 3, 4Google Scholar); but their hostility was covert, cf. Qu. fr. II, 1, 1Google Scholar; 5, 3. Note Caelius in Fam. VIII, 14, 2Google Scholar: sic illi amores et invidiosa coniunctio (of Caesar and Pompey) non ad occultam recidit obtrectationem (as might have been expected), sed ad bellum se erupit.

^{page 12 note 4} Val. Max. IV, 1, 12; Plin., NH VII, 144Google Scholar; Plut., Mor. 202 AGoogle Scholar.

^{page 13 note 1} Font. 23 (inimicitiarum suspicio); 27 (obtrectatorem); Val. Max. VIII, 5, 1.

^{page 13 note 2} Rep. I, 31Google Scholar; Brut. 81; cf. Plut., Ti. Gr. 14, 2Google Scholar; Cic., Phil. VIII, 14Google Scholar.

^{page 13 note 3} On inimicitiae (hostile acts or declarations) cf. Hellegouar'ch, 186 ff. Rogers, R. S., TAPA XC (1959), 224 ff.Google Scholar, discusses the formalities, mostly from imperial evidence; Fam. V, 2, 5Google Scholar shows that there might be doubt in Cicero's time whether hostilities existed, requiring formal reconciliation. Hereditary: e.g. Ascon. 62–3c; Offic. II, 50Google Scholar; Ac. II, 1. But there were no such secular feuds as in Medieval or Renaissance Florence.

^{page 13 note 4} Offic. II, 50Google Scholar; Sull. 6–7; Prov. Cons. 24; Fam. I, 9, 10Google Scholar.

^{page 13 note 5} Div. in Caec. 55–8 (but cf. Verr. II, 1, 15Google Scholar); Verr. II, 3, 7Google Scholar.

^{page 13 note 6} Font. 23 ff.; Sall., Cat. 49, 2 (Piso)Google Scholar; Ascon. 60c. For what follows cf. generally Offic. II, 49 ff.Google ScholarPliny, , ep. IX, 13, 2Google Scholar (materiam insectandi nocentes, miseros vindicandi, se proferendi) corresponds to Republican practice.

^{page 13 note 7} Principes once appeared pro sociis, now only imperiti adulescentes, Div. in Caec. 63 ff.; cf. Verr. II, 3, 6Google Scholar; Cael. 73 ff.; Offic. loc. cit. (last note); Quint., Inst. XII, 6, 1Google Scholar; 7, 1 ff.

^{page 14 note 1} Div. in Caec. 1–4; Verr. II, 1, 98Google Scholar; 2, 10 and 179; 5, 189; Rab. perd. 1; Vat. 5 (cum in hac civitate oppugnatio soleat,…defensio numquam vituperari); Mur. 45; de orat. II, 200Google Scholar.

^{page 14 note 2} Reg. Deiot. 30; Sull. 81.

^{page 14 note 3} RE VIII, 2470 ff.Google Scholar (v. der Mühll).

^{page 14 note 4} Mur. 9. De orat. I, 184Google Scholar, depicts the jurisconsult praesidium clientibus atque opem amicis et prope cunctis civibus lucem ingeni et consili porrigentem. Cf. Cicero on his own practice: tantum enitor, ut neque amicis neque etiam alienioribus opera, consilio, labore desim (Fam. I, 9, 17Google Scholar).

^{page 14 note 5} Att. IV, 15, 9Google Scholar; for his relations to the candidates, Qu. fr. III, 1, 16Google Scholar; he preferred Messala, III, 6 (8), 3.

^{page 14 note 6} Party Politics, 7.

^{page 15 note 7} Ascon. 18–20c. Cf. perhaps the trial of Catilina in 65, ibid. 87 with Sull. 49; 81.

^{page 15 note 2} See (i) Font. 36; (ii) Ascon. 60 f.; Brut. 271 (pro Cornelio); (iii) Cluent. 10; 118; Brut. 271; (iv) Mur. 3; 7; 10; (v) Sull. 2; 47; (vi) Flacc. 2; (vii) Cael. 7; 25; (viii) Planc. 2 ff.; 58; (ix) Scaur. 31; (x) Rab. Post. 32; (xi) Lig. 21. The bitter attacks on Hortensius in the Verrines are exceptional and did not preclude a hollow friendship later. L. Torquatus' taunts did indeed make him threaten to renounce friendship, Sull. 21 f.; 47, but his moderation on this occasion was not unique, ibid. 49.

^{page 15 note 3} Planc. 56; Ascon. 60 f.

^{page 15 note 4} Tull. 3; Rab. perd. 25.

^{page 15 note 5} Hortensius could even profess to assume that this was the one legitimate reason, Verr. II, 3, 6Google Scholar.

^{page 15 note 6} St. in Greek and Rom. Hist. 46 f.

^{page 15 note 7} Cf. Div. in Caec. 61; Planc. 28.

^{page 15 note 8} Cluent. 57; Mur. 2; 59; 86; Sull. 2; 4, etc. Of course prosecutors were vexed by the success of defending counsel, and it could be said that in getting off C. Cornelius tr. pl. 67 Cicero had displeased the boni, Vat. 5; but cf. Sull. 49.

^{page 16 note 1} Kl. Schr. I, 208Google Scholar.

^{page 16 note 2} Fam. I, 1 ff.Google Scholar, esp. 1, 3.

^{page 16 note 3} Rom. Rev. ch. 3. He admits that they were not ‘all, or consistently, allies of Pompeius’ (44, n. 2). There is surely little reason to expect that they would act together. Lentulus Clodianus, censor 70, expelled Sura, Lentulus, cos. 71Google Scholar, from the senate. Syme does not mention L. Lentulus Niger, accused with his son by Vettius of plotting Pompey's death (Att. II, 24, 2Google Scholar; Vat. 25). Several Lentuli served under Pompey in his wars, but it seems to me dubious if a general's officers were necessarily his close associates. In 88 Sulla was deserted by his chief officers (App., BC I, 57Google Scholar); L. Piso's legates are said to have become hostile to him (Pis. 53–4); Cicero's, apart from his brother, had no close connexions with him earlier. Philus, L. Furius, cos. 136Google Scholar, took out with him to Spain two legates, Q. Metellus and Q. Pompeius, who were on bad terms with him and each other (Val. Max. III, 7, 5; Dio, fr. 82).

^{page 16 note 4} E.g. Att. III, 22, 2Google Scholar; Fam. I, 9, 4Google Scholar; Caes., BC I, 22Google Scholar.

^{page 16 note 5} Fam. I, 9 (everywhere implied)Google Scholar.

^{page 17 note 1} Contrast Qu. fr. II, 1, with Fam. I, 2, 2Google Scholar; Qu. fr. II, 5, 3Google Scholar; Dio, XXXIX, 28 and 30.

^{page 17 note 2} Plut., Cr. 6, 4Google Scholar; 7; 12; Pomp. 22–3. Pompey expected collegam minorem et sui cultorem but Crassus was obtrectans potius collegae quam boni aut mali publici gnayus aestimator (Sall., Hist. IV, 48; 51)Google Scholar.

^{page 17 note 3} Plut., Pomp. 43Google Scholar; Cic., Flacc. 32Google Scholar.

^{page 17 note 4} Gelzer, Caesar ⁸, ch. II. Dio, XXXVI, 43; XXXVII, 22 has, as in his interpretation of some later events (XXXVII, 54–5; XXXIX, 24–6), allowed his knowledge of the civil war to colour and falsify the earlier relations of Caesar and Pompey.

^{page 17 note 5} Holmes, Rice, Rom. Rep. I, 475Google Scholar.

^{page 17 note 6} Gelzer, , Caesar⁶, 110Google Scholar.

^{page 17 note 7} Fam. V, 2, 8Google Scholar for common friends of enemies; V, 17, 2: inimici non solum tui verum etiam amicorum tuorum; Att. XIV, 13 B, 3Google Scholar: semper ita statui, non esse insectandos inimicorum amicos, praesertim humiliores; Caes., BC I, 3Google Scholar: necessarii Pompei atque eorum qui veteres inimicitias cum Caesare gerebant. In 62 the enmity between Cicero and Metellus Nepos produced only a marked coolness in his relations with Nepos' brother, Celer (Fam. V, 1–2Google Scholar, esp. 2, 9–10). Reconciled with Cicero, (Sest. 130Google Scholar; Fam. V, 3Google Scholar), Nepos still backed his brodier-in-law, Clodius, P. (Sest. 89)Google Scholar. It was (I think) exceptional if Antony's hatred for Cicero extended to all his friends (Nepos, , Att. 10, 4Google Scholar); cf., however, Caes., BC I, 4, 4Google Scholar, and texts cited on p. 18, n. 6.

^{page 18 note 1} Pis. 35, but cf. Fam. III, 10, 8Google Scholar.

^{page 18 note 2} Scaur. 31 ff. Pompey was openly backing, though he later abandoned, Scaurus as consular candidate (Att. IV, 15, 7Google Scholar; Qu. fr. III, 6, 3Google Scholar); he had a secret grudge against him (cf. Ascon. 19–200), of which Appius might perhaps have known.

^{page 18 note 3} Sest. passim, cf. Fam. I, 9, 6 ff.Google Scholar (Against Balsdon, J. P. V. D., JRS XLVII, 18 ff.Google Scholar see Stockton, D. L., TAPA XCII, 471 ff.Google Scholar)

^{page 18 note 4} Fam. I, 8, 2–4Google Scholar.

^{page 18 note 5} Att. IV, 18, 2Google Scholar; Qu. fr. III, 4, 1Google Scholar; 5, 4; Fam. II, 4, 1Google Scholar.

^{page 18 note 6} Fam. I, 9, 19–20Google Scholar on Vatinius and Crassus; on his quarrel with Crassus the ‘boni’ gaudere se dicebant mihi et ilium inimicum et eos, qui in eadem causa essent, numquam amicos futuros; cf. 1, 9, 10. Gabinius, : Qu. fr. III, 1, 15Google Scholar; 2, 2; 4, 2–3 show the insincerity of Rab. Post. 32 ff. Piso: Cicero alleged that friendship with Caesar restrained him from prosecuting, Pis. 81 ff.

^{page 19 note 1} Fam. VIII, 14, 3Google Scholar; cf. 16 and 17, 1.

^{page 19 note 2} Fam. X, 31, 2Google Scholar: cum vero non liceret mihi nullius partis esse, quia utrubique magnos inimicos habebam, ea castra fugi in quibus plane tutum me ab insidiis sciebam non futurum.

^{page 19 note 3} JRS XXVIII, 113 ff.Google Scholar But Cicero thought that Labienus seemed damnasse sceleris hominem amicum rei publicae causa (Att. VII, 12, 5Google Scholar; cf. Fam. XIV, 14, 2Google Scholar; XVI, 12, 4).

^{page 19 note 4} Att. VIII, 15A, 2Google Scholar; IX, 7B, 2.

^{page 20 note 1} E.g. Att. VII, 12, 3Google Scholar (nec solum civis sed etiam amici officio revocor); VIII, 1, 4; 3, 2, etc.; IX, 13, 3 on exaggeration of Pompey's services.

^{page 20 note 2} pro Marcello 30; cf. Lig. 19.

^{page 20 note 3} Fam. XI, 27–8Google Scholar, on which see Heuss, A., Historia V, 53 ff.Google Scholar and B. Kytzler, ibid. IX, 96 ff.; I do not wholly agree with either, but any interpretation of Matius' views must probably remain subjective.

^{page 20 note 4} Attempts were often made to strengthen them by marriages, but they might actually introduce new sources of discord (see for one instance Badian, E., Studies in Gk. and Rom. Hist. 39 ff.Google Scholar), or at least fail in effect, cf. Cluent. 190: novis inter propinquos susceptis inimicitiis saepe fieri divortia atque adfinitatum discidia videmus…ceteri novis adfinitatibus adducti veteres inimicitias saepe deponunt.