¹ QFr. and the collection of letters ad Brutum (henceforth, ad Brut.) are overshadowed by the two more prominent Ciceronian collections, Att. and the Epistulae ad familiares (henceforth, Fam.); neither QFr. nor ad Brut. is typically studied for its own sake. This ‘phenomenon’ is confirmed by the scaled-down bibliography (when compared to that on Att. and Fam.) on the two minor collections: from 2002 to 2021, there have been only 15 new entries in APh on QFr. and 5 on ad Brut., whereas there have been 121 new studies on Att. and 111 on Fam.

² Almost half of the 56 entries in APh on QFr. focus on a singular theme/letter or employ the collection as background for Att. and Fam. Ad Brut. presents a similar scenario: among the 44 entries in APh, the majority presents historical and prosopographical studies on Brutus; the others employ it as source for the discussion of Att. and Fam.

³ Beard, M., ‘Ciceronian correspondences: making a book out of letters’, in Wiseman, T.P. (ed.), Classics in Progress: Essays on Ancient Greece and Rome (Oxford, 2002), 103–44Google Scholar. See also Gibson, R.K., ‘On the nature of ancient letter collections’, JRS 102 (2012), 56–78Google Scholar and Gibson, R.K. and Morrison, A., ‘Patterns of arrangement in Greco-Roman letter collections: 400 bce–400 ce’, in Facione, S. (ed.), Concatenantur sibi epistulae nostrae. Reading Ancient Latin Letter Collections (Foggia, 2023), 11–57Google Scholar. For the analysis on the organizing patterns of single books, cf. Grillo, L., ‘Reading Cicero's Ad familiares 1 as a collection’, CQ 100 (2015), 1–14Google Scholar and Grillo, L., ‘The artistic architecture and closural devices of Cicero's Ad familiares 1 and 6’, Arethusa 49 (2016), 399–413CrossRefGoogle Scholar, Martelli, F., ‘The triumph of letters: rewriting Cicero in Ad fam. 15’, JRS 107 (2017), 90–115Google Scholar. Gibson, R.K., ‘The Pro Marcello without Caesar: grief, exile and death in Cicero, Ad familiares 4’, Hermathena 202/203 (2017), 105–46Google Scholar, S. Cammoranesi, ‘Cicero's Epistulae ad familiares narratives of the Civil War’ (Diss., Manchester University, 2022). On Att. and ad Brut., see R.K. Gibson (forthcoming).

⁴ The massive number of editions of Att. and Fam., from the fifteenth century onwards, also demonstrates the growing interest in the two major collections only. On Att.: Gibson (n. 3 [forthcoming]); on Fam. and on QFr.: Gibson (n. 3 [forthcoming]).

⁵ Henderson—in, J. ‘Cicero's letters to Cicero ad QFR: big brother's keeper’, Arethusa 49 (2016), 439–61CrossRefGoogle Scholar and ‘Continued: soon after’, Arethusa 49 (2016), 489–501—is the first modern scholar to have analysed QFr. as a corpus and to have raised questions about the books’ narrative.

⁶ A similar idea is explored, for Fam. Book 15, by Martelli (n. 3).

⁷ White, P., Cicero in Letters: Epistolary Relations of the Late Republic (Oxford and New York, 2010)CrossRefGoogle Scholar. For the scholarship that has followed White's approach: see n. 3 above.

⁸ See Gibson (n. 3 [forthcoming]) on QFr. 1.3.1 and Ov. Tr. 4.3.39–40 and QFr. 1.3.1 and Ov. Tr. 3.11.25–30. Possible allusions to QFr. have been also detected in Plin. Ep. 8.24.1, 8.

⁹ The earliest certain evidence for the circulation of Fam. is Seneca the Elder (Suas. 1.1.5); the first certain references to items from Att. are found in Seneca's letters (97.3–6, 118.1–2). On the evolution of epistolography in the period here discussed, see Cugusi, P., Evoluzione e forme dell'epistolografia latina nella tarda repubblica e nei primi due secoli dell'impero, con cenni sull'epistolografia preciceroniana (Rome, 1983)Google Scholar.

¹⁰ The surviving manuscripts descend from the (now lost) Verona manuscript of Cicero's letters which Petrarch found in 1345 in the Cathedral Library of Verona. It probably represents the late antique circulation of the three collections (Att., QFr., ad Brut.) together. Among extant manuscripts, the most important is the Florentine Laur. Med. 49.18 (14th cent. = M), which recreates the contents and order of Petrarch's manuscript (similarly in the Parisinus lat. 8538 [15th cent. = R]). On this, see Cook, B.L., ‘Petrarch's reading of Cicero's letters: a Ciceronian response’, C&M 63 (2012), 321–53Google Scholar and Gibson (n. 3 [forthcoming]).

¹¹ Ad Brut. and Att. were probably published between the end of the first century b.c.e. and the first century c.e.: Gibson (n. 3 [forthcoming]).

¹² See section 4 below.

¹³ The letters mostly follow a chronological sequence, with the only exception being 2.8 (early February 55), which has been moved forward to precede 2.9 (June 56). A similar situation obtains in the much larger Att. (and also in ad Brut.), where the letters follow a broadly chronological order (roughly seventeen per cent of the 426 free-standing letters of Att. appear to be positioned out of chronological order: Gibson (n. 3 [forthcoming]), while only 6 out of the 26 extant missives of ad Brut. fall out of this pattern. Contrast Fam., where the letters follow different criteria, including, but not limited to, chronology: see n. 3 above.

¹⁴ QFr. is the only extant collection that respects this principle strictly. The collection that comes closest is Att., since it contains only a small number of letters (six per cent: 27 out of a total of 453: Gibson [n. 3 (forthcoming)]) sent to/from someone else. Ad Brut. is next, with thirty-five per cent (9 out of 17) of letters that fall outside this principle of a single addressee. Fam. presents a different structure, since it includes letters from a range of Cicero's different correspondents: Gibson (n. 3 [forthcoming]).

¹⁵ Quintus was proconsul of Asia from 61 to May 58. From late 57 to June 56, he acted as Pompey's legate in Sardinia and from 54 to 52 as Caesar's legate in Gallia: see Broughton, R.T., The Magistrates of the Roman Republic (New York, 1952), 2.181, 213, 226Google Scholar.

¹⁶ The last missive included in the collection is QFr. 3.7 (December 54). Cicero surely addressed other letters to Quintus during the years of his consulate and from exile or after December 54. This intended selection could be a signal of editorial design, since it is possible that Cicero and Quintus exchanged other letters, as might be demonstrated by some letters included in Att.: see section 3 below.

¹⁷ Quintus must have written to Cicero especially when abroad (on this, see n. 52 below). Writing to those who were in Rome, to secure their protection, was crucial for those who were abroad: White (n. 7), 10 n. 7. Selected internal evidence: te puto saepe habere qui num quid Romam uelis quaerant. quod ad me <de> Lentuli et Sesti nomine scripsisti (QFr. 2.2.1); a te post illam Vlbiensem epistulam (QFr. 2.3.7).

¹⁸ The contents of the collection have also been discussed by Harvey, P.B., ‘Cicero Epistulae ad Quintum fratrem et ad Brutum: content and comment’, Athenaeum 79 (1991), 17–29Google Scholar.

¹⁹ Gibson (n. 3 [forthcoming]), who has a spreadsheet on QFr. The line counts in Gibson are based on W.S. Watt (ed.), Marcus Tullius Cicero: M. Tulli Ciceronis epistulae (Oxford, 1958).

²⁰ MS M registers nine letters for Book 3, instead of the seven Shackleton Bailey includes in his edition (2002). Copyists of M failed to recognize 3.5 as one letter alone, but believed that it contained three different letters: Gibson (n. 3 [forthcoming]).

²¹ A similar progressive degradation of Cicero’ social prestige is also shown in the similarly short compass of Fam. 5.1–21, even if the ancient editor's/editors’ intervention here is more invasive and probably counter-historically orientated. For the counter-historical narrative on Cicero that emerges from Fam. 15, see Martelli (n. 3). Whitton has traced a similar movement in ad Brut.: Whitton, C., ‘Last but not least: Ad M. Brutum’, Hermathena 202/203 (2017), 185–224Google Scholar, at 194.

²² As shown below (pages 6–15), the presence/absence of meaningful keywords in the letters and the length of the letters are crucial for understanding the narrative on Cicero's life presented by the collection.

²³ The idea that the decline of Cicero and his influence mirrors the decline of the Republic is also found in numerous post reditum speeches: J.H. Nicholson, Cicero's Return from Exile: The Orations Post reditum (Bern, 1992) and Riggsby, A.M., ‘The «post reditum» speeches’, in May, J.M. (ed.), Brill's Companion to Cicero: Oratory and Rhetoric (Leiden / Boston / Cologne, 2002), 159–95CrossRefGoogle Scholar.

²⁴ Instead, individual books of Att. and Fam. present micro-narratives within a much larger macro-narrative involving the events of Cicero's life. These micro-narratives (differently from QFr.) are not also recurrent at the level of the whole collection but are characteristically confined to one or more books: cf. the narrative on Cicero's exile in Greece, which is confined only to Att. Book 3; see also the narrative on Cilicia that emerges only from Att. Books 5–6: Gibson (n. 3 [forthcoming]).

²⁵ Similarly in Fam. Book 5, where the first set of letters presents a rather enthusiastic/politically engaged image of Cicero (1–12), which appears in contrast with that emerging from the concluding set of letters (13–21).

²⁶ Although at the beginning of QFr. 1.1 Cicero states that his intention is not to advise Quintus on how to behave (§8 atque haec nunc non ut facias … scribo), he eventually becomes Quintus’ mentor (§18 sed nescio quo pacto ad praecipiendi rationem delapsa est oratio mea, cum id mihi propositum initio non fuisset). To offer advice was a delicate business, especially if it was unwanted or unsought. Quintus, who at the time held a proconsulship, might feel offended by Cicero's attempt to advise him on a matter in which he had a certain expertise, but Cicero did not (Cicero will hold a proconsulship only between 51–50). Strategies of redressive politeness (for which, see J. Hall, Politeness and Politics in Cicero's Letters [Oxford and New York, 2009], 107–34), such as the praise of Quintus’ leading abilities (§§9, 25) and the stress put on Cicero's affection (§§41, 45), are employed by Cicero to mitigate Quintus’ possible irritation. Quintus was probably familiar with the greater part of Cicero's advice (see Cicero's advice on Quintus’ staff at §§10–15 or selectivity in friendships at §16), as Cicero points out later in §36 (sed quid ego te haec hortor quae tu non modo facere potes tua sponte sine cuiusquam praeceptis). On Cicero's language for offering advice: see n. 32 below.

²⁷ QFr. 1.2, written a year later than QFr. 1.1, repeats its themes. Moreover, the gap of a year between 1.1 and 1.2 makes it possible to believe that there must have been other letters exchanged between the two brothers: see nn. 16–17.

²⁸ Conversely, QFr. 1.3 and 1.4 are written from Thessalonica during Cicero's exile.

²⁹ In contrast to the 623 lines of Book 2 and to the 617 lines of Book 3: Gibson (n. 3 [forthcoming]). The decreasing number of lines in the letters included in each book might be intentional: it certainly helps to bolster the narrative on Cicero that I believe is in place in QFr. For a broader study on Cicero's book-length, see Stover, J., ‘The Ciceronian book and its influence: a statistical approach’, Ciceroniana On Line 5 (2021), 263–83Google Scholar. Length appears also to be a criterion for the organization of ancient letter collections: Gibson and Morrison (n. 3), 11–57.

³⁰ Letter 1.1 is the longest of the entire surviving Ciceronian corpus; the next longest is Fam. 1.9 (443 lines), a letter on Lentulus’ proconsulship in Cilicia.

³¹ Letter 1.2 is second only to 3.1 (255 lines): Gibson (n. 3 [forthcoming]). However, this does not invalidate my reading of the collection. Letter 3.1 is a different case: it inaugurates the dark narrative on Cicero that gradually emerges from the last book of the collection.

³² The tone and the language here employed are appropriate to Cicero's role as adviser (on which, see n. 26 above): cf. the use of rogo: 1.1.4, 1.2.11, 1.2.14; admoneo: 1.2.14; praescribo: 1.1.12; hortor: 1.1.36, 1.1.46. On this letter as an essay on good rule and modelled on the philosopher's letter of advice to a ruler, see Rawson, E., ‘Roman rulers and the philosophic advisor’, in Griffin, M.T. and Barnes, J. (edd.), Philosophia Togata: Essays on Philosophy and Roman Society (Oxford, 1989), 233–57Google Scholar. On the contrary, Cicero is the one listening to Quintus’ advice in Book 3.

³³ Henderson (n. 5), 442 n. 6.

³⁴ Peter Heslin's data generated through a machine learning algorithm has been crucial for the analysis of the language of the letters. The data were made available for the Durham–Tübingen ‘Latin Text Collections’ event in September 2021.

³⁵ … rem publicam funditus amisimus, adeo ut <C.> Cato … in contionem ascendit et Pompeium priuatum dictatorem appellauit (1.2.15).

³⁶ On Cicero's language of amicitia: Hellegouarc'h, J., Le vocabulaire latin des relations et des partis politiques sous la République (Paris, 1963)Google Scholar. See also F. Prost, ‘«Amor» et «amicitia» dans la correspondance d'exil de Cicéron’, VL 191–192 (2015), 7–35.

³⁷ Letter 1.2.16 predates the lex de capite ciuis Romani by which Clodius condemned Cicero for his decision to sentence the Catilinarians to death without the prouocatio ad populum to exile.

³⁸ nostram gloriam tua uirtute augeri expeto (1.1.2); qua re quoniam in eam rationem uitae nos non tam cupiditas quaedam gloriae quam res ipsa ac Fortuna deduxit ut sempiternus sermo hominum de nobis futurus sit (1.1.38); simul et illud cogita, nos non de reliqua et sperata gloria iam laborare sed de parta dimicare, quae quidem non tam expetenda nobis fuit quam tuenda est (1.1.43). Glory will also come from his friends’ support in the quarrel with Clodius: si diem nobis dixerit, tota Italia concurret, ut multiplicata gloria discedamus (1.2.16).

³⁹ Whereas 1.1 and 1.2 were written from Rome.

⁴⁰ Similarly in Fam. Book 5, where the voluntary conciseness of some letters reveals the writer's emotions, even before their words do. On the length of the letters, see nn. 29, 30 and 31 above.

⁴¹ spero, si quid mea spes habet auctoritatis (1.3.5).

⁴² Cf. 1.3.2: atque utinam me mortuum prius uidisses aut audisses, utinam te non solum uitae sed etiam dignitatis meae superstitem reliquissem! sed testor omnis deos me hac una uoce a morte esse reuocatum, quod omnes in mea uita partem aliquam tuae uitae repositam esse dicebant … nam si occidissem, mors ipsa meam pietatem amoremque in te facile defenderet.

⁴³ amici partim deseruerint me, partim etiam prodiderint (1.3.5). Cf. 1.4.3 (cited above).

⁴⁴ On this, see page 7 above.

⁴⁵ See pages 6–7 above.

⁴⁶ Similarly in 1.3.1 and 1.4.1.

⁴⁷ gloria and its derivatives are absent in 1.3; gloria is found only once in 1.4.4 (sed saepe triduo summa cum gloria dicebar esse rediturus) but with reference to a promise of glory, after his return from exile, that does not fit anymore into Cicero's present.

⁴⁸ As discussed at page 5 above, letters that display an image of the influence and prestige of Cicero and his family open each book of the collection (1.1–2, 2.1–3, 3.1–4). Conversely, the remaining letters (1.3–4, 2.4–16, 3.5–7) convey an atmosphere of angst and despair, marked by Cicero and his relatives’ loss of power.

⁴⁹ In 2.1–3, the reader observes Cicero taking once again part in the Senate's meetings and advocating in favour of clients, as also shown by the language of activity employed in the letters.

⁵⁰ The language of inactivity of 2.4–16 highlights this image of Cicero.

⁵¹ Att. has a similar chronological progression in the organization of its books: from November 68 (start date of Book 1) to November 44 (end date of Book 16). For this and other similarities to Att., see section 4.

⁵² Relying on family members, to be informed on the events occurred in Rome, was a common practice when aristocrats were (for duty or exile) abroad: White (n. 7), 12–18. Cicero's plea to his brother Quintus to write back to him on all matters that occurred in Rome (1.4.5) cannot have remained unheeded. Quintus, in fact, must have written letters to the exiled brother, since he had been in Rome for some time—after his return from Asia in May 58 and before his departure to Sardinia in the autumn of 57 (see also n. 17 above).

⁵³ A good number of letters from exile, between August 58 and November 57, have been included in Att. 3.13–27, 4.1–3 and Fam. 14.1–4.

⁵⁴ See the relevant sub-section.

⁵⁵ 2.1.1 on the Campanian land law; 2.1.2 on the process de ui against Clodius; 2.2.3 on the debate on who had to restore King Ptolemy XII Auletes; 2.3.1 on Clodius’ charge de ui against Milo.

⁵⁶ Gibson (n. 3 [forthcoming]).

⁵⁷ See pages 6–9.

⁵⁸ On QFr. 2.4–5, see R.G. Böhm, ‘Cicero, Ad Quintum fratrem II, 4–5’, QUCC 23 (1986), 93–107.

⁵⁹ His reports on the Senate's agenda are less detailed than the ones in 2.1–3: a distinctive feature of Cicero's changed mood.

⁶⁰ See pages 6–9.

⁶¹ The triumvirs appear also in the first set of letters (e.g. 2.3.4, where Cicero envisages that nothing good will come for their incessant research of power). However, differently from 2.4–16, Cicero keeps here a rather positive attitude (similarly in 2.3.7).

⁶² A reduction of over sixty per cent is esteemed.

⁶³ Leaving Rome to spend time in a villa could be considered a betrayal of one's duties; however, when the situation started to appear more desperate (especially in the period that immediately precedes/follows the congress of Lucca), Cicero takes refuge in his country estates: O. Rossi, ‘Letters from far away: ancient epistolary travel writing and the case of Cicero's correspondence’ (Diss., Yale University, 2010).

⁶⁴ me … otiantem (2.9.1) and … summum otium forense (2.14.5), where Cicero's political disengagement reflects the forum's inactivity owing to the increasing power of the triumvirs. Cicero's literary activity (a symbol of his political dissatisfaction) is mentioned also in 2.12.4 and 2.13.1–2.

⁶⁵ See also 2.7.1, 2.11.1, 2.13.3, 2.16.5.

⁶⁶ Cicero seems to revert to his literary works to find behavioural guidelines: White (n. 7), 109–12. Similarly in 2.8.1, where Cicero's invitation to grow apart from politics and his refuge in literary studies are (probably) concealed under an obscure reference to Jupiter's speech in the lost De temporibus meis: Di Spigno, C. (ed.), Epistole al fratello Quinto e altri epistolari minori (Turin, 2002), 166Google Scholar.

⁶⁷ The awareness of the hazards of the delivery might have urged Cicero to convert information into literary allusions that Quintus would understand, since they shared the same literary background: Schröder, B.-J., ‘Couriers and conventions in Cicero's epistolary network’, in Ceccarelli, P., Doering, L., Fögen, T. and Gildenhard, I. (edd.), Letters and Communities: Studies in the Socio-Political Dimensions of Ancient Epistolography (Oxford, 2018), 81–100Google Scholar. On the use of literature in Cicero's epistles, see Čulík-Baird, H., Cicero and the Early Latin Poets (Cambridge, 2022)CrossRefGoogle Scholar. Further examples are found in 2.9.3, where Quintus is depicted as meliorem ciuem esse quam Philoctetam, and in 2.11.3, where a quotation from an unknown Latin play is employed to highlight Pompey's role in Cicero's and Appius’ reconciliation. Similarly in 2.14.5, where Eur. Supp. 119 emphasizes the dreadful condition of the Republic, and in 2.16.3, where Cicero refers to Quintus’ (probable) interpretation of some episodes from Sophocles’ Banqueters to allude to events which happened in Caesar's camp in Gaul. See also 3.1.18, where Cicero alludes to Eur. Hipp. 436.

⁶⁸ See 2.14.5 and 3.1.10, 3.2.1, 3.3.3, 3.4.1–2.

⁶⁹ Similarly in 3.2.2 (nolo cum Pompeio pugnare); 3.3.3 (animum … moderatum); 3.4.3 (mediocritate delector); 3.5.2 (ne … offenderem quempiam); 3.7.1, 3 (lenitate … lenissime; ne … cuiusquam animum … offendant). Cicero's reference to moderatio might be key to understand why the collection concludes in 54: see section 3 below.

⁷⁰ On the number of letters included in Book 3, see n. 20 above.

⁷¹ In a lost letter, Quintus probably informed Cicero on Pompey's decision to appoint the latter as his legate; a reference is in … quod inferior epistula scribis me Id. Sept. Pompeio legatum iri (3.1.18). However, neither Cicero nor Pompey went to the province. Pompey remained in Italy in charge of the grain supply—also to control Caesar's movements. Cicero probably did not want to upset Caesar (3.1.18).

⁷² Differently from QFr. 2.1, this letter was written one year and four months after 1.4.

⁷³ A similar design is found in Att., where Books 5–12 (May 51–June 45) and 14–15 (April 44–October 44) follow the narrative arc of the previous book.

⁷⁴ A narrative on the Civil War(s) of the 40s is also displayed by Fam.; see Cammoranesi (n. 3).

⁷⁵ ‘Moderation’ might be one of the reasons for the absence of letters dated after 54 in the collection. Another reason might be connected to how Cicero's and Quintus’ relationship changed after 54.

⁷⁶ See the relevant sub-sections.

⁷⁷ E.g. Cicero's aid given to Curtius for the election to the tribunate (3.1.10) or his pledge of support to the candidates for the consulship in 53 (3.1.16).

⁷⁸ E.g. the support offered to Scaurus and Plancius (3.1.11).

⁷⁹ Cf. page 10.

⁸⁰ On Gabinius’ case, see 3.1.15, 3.2.1–2, 3.3.3, 3.4.1–2. Gabinius is also mentioned in 1.2.15, 2.7.1, 2.12.2.

⁸¹ Cf. 3.2.3 on the charges of bribery to the candidates for the consulship and 3.3.2 on the cancellation of the election days for contrary omens.

⁸² Letters 1.1 and 1.2 have an average of 711 lines (c.356 lines per letter), whereas letters 1.3 and 1.4 count 160 lines (c.80 per letter). A similar design is found in Book 2: 2.1–3 have a total of 200 lines (c.50 per letter), while 2.4–16 present 423 lines (c.35 per letter).

⁸³ For the occurrence of these keywords in Book 1, see pages 6–9. For the reason why they have a low weight in Book 3, see below.

⁸⁴ The employment of the language of activity in 3.1–4 (although with a different weight compared to 1.1–2 and 2.1–3) might be part of the ancient editor's/editors’ agenda.

⁸⁵ Similarly in 3.1.15, 3.2.2, 3.4.2, 3.5.4.

⁸⁶ White (n. 7), 11–21.

⁸⁷ See also the repetition of sit/sint in 1.1.13.

⁸⁸ The only exceptions are in 3.6.2 and 3.7.3, where Cicero suggests that his brother should carefully select what to include in the letters. This cannot be compared to the advice Cicero used to give on higher matters to Quintus in 1.1–2: see n. 26 above.

⁸⁹ Similarly to ad Brut., which covers an even shorter timespan (April–July 43), when compared to the wider timespan of Att. (covering twenty-four years of Cicero's life, from 68 to 44) and Fam. (nineteen years, from 62 to 43). But what we have of ad Brut. are the remnants of a much longer nine-book collection: this is a crucial difference.

⁹⁰ From QFr. 2.10 to 3.7.

⁹¹ Similarly to QFr., the narration in Att. stops in 54 (Att. 4.19) and resumes only in 51 (Att. 5.1). For the (probably) intentional omission of years 53–52, see below.

⁹² Largely discussed in 2.14–16 and in Book 3.

⁹³ Cf. 2.9.1–2, where Cicero seems to rebuke Quintus because he might have used Quintus Junior's illness as excuse not to see his brother. Letter 2.13.2 might display a similar case: Cicero might have emphasized his closeness to Quintus Junior (probably) to criticize Quintus’ failure to be a present father. On this, see Henderson (n. 5), 455–7; Ash, R., ‘Un-parallel lives? The younger Quintus and Marcus Cicero in Cicero's letters’, Hermathena 202/203 (2023), 71–104Google Scholar.

⁹⁴ The only exception is Fam. 16.16.1, written by Quintus to Cicero in late May/early June 53; it recounts Quintus’ joy for Tiro's emancipation and his affection for his brother. In Att. 11.13.2 (mid March 47), 15.2 (May 47), 16.4 (June 47), Cicero refers to an exchange of letters between himself and his brother, but none of those has survived. See also nn. 16–17 above.

⁹⁵ This is the earliest attestation on the deterioration of their relationship: see Bailey, D.R. Shackleton, Cicero (London, 1971), 179–85Google Scholar.

⁹⁶ Cicero was advised by Caesar not to leave Italy: Att. 10.8.10, 8a, 8b, 9a.

⁹⁷ scio post Hirtium conuentum arcessitum a Caesare, cum eo de meo animo a suis rationibus alienissimo et consilio relinquendi Italiam (Att. 10.4.6). Cf. Ash (n. 93), 25.

⁹⁸ Att. 10.6.2, 10.7.3, 10.11.3.

⁹⁹ Quintus Senior was Cicero's legate in Cilicia in 51–50: Broughton (n. 15), 245–53. His son Quintus went with them.

¹⁰⁰ See pages 12–13.

¹⁰¹ Cicero seems not to be in contact with them. In Att., Cicero highlights that he was informed of their activity by other sources: 11.6.7, 7.7, 8.2, 9.2, 10.1, 12.2, 14.3, 16.4. For Cicero's suffering in the face of Quintus father and his son's hostility: Att. 11.5.4, 10.1, 13.2, 15.2.

¹⁰² A reconciliation (or, more probably, a temporary truce) between the two brothers might have occurred between July/August 47: Att. 11.20.1, 23.2; however, rifts resumed in late August 47 (Att. 11.21.1, 22.1) and carried on in 45: see Ash (n. 93), 27.

¹⁰³ A probable allusion is in Att. 11.7 (from Brindisi); cf. also Fam. 7.3.3.

¹⁰⁴ Quintus father and Quintus son seem to have denigrated Cicero to gain Caesar's favour: Ash (n. 93), 24–34 and n. 74.

¹⁰⁵ The absence of letters from 51–50 might be justified by their vicinity since they were all in Cilicia.

¹⁰⁶ On allusions to Cicero's and Quintus’ lost letters, see nn. 16–17 above.

¹⁰⁷ Differently from Fam., which contains thirteen letters of 53 (2.16, 3.1, 7.11–15, 7.18, 16.10, 16.13–16) and two of 52 (5.18, 13.75).

¹⁰⁸ On this, see section 1.

¹⁰⁹ The majority of the manuscripts of QFr. contains also Att., with QFr. often in second position after Att.: see n. 10 above.

¹¹⁰ The only Ciceronian letter collection which adheres strictly to this rule is QFr., with the second closest collection being Att.: see n. 14 above.

¹¹¹ QFr. similarly to Att. and ad Brut.—but differently from Fam.—follow a largely chronological order: see n. 13 above.

¹¹² I follow Shackleton Bailey's 1987 edition for Att. and his 2002 edition for QFr.

¹¹³ The presence of twenty-six embedded dates in Att. Books 2–4 and eleven in QFr. Books 1–3 increases the plausibility of my reading.

¹¹⁴ Whereas many books of Att. feature letters from others enclosed by Cicero for the attention of Atticus, Att. Books 2–4 contain only Cicero's letters to Atticus.

¹¹⁵ Gibson (n. 19); Gibson (n. 3 [forthcoming]).

¹¹⁶ See n. 10 above.