¹ All translations are my own. I am indebted to Bruce Gibson, Jason König, Christina Kraus, Myles Lavan, Christopher Whitton, Greg Woolf and the anonymous reader at CQ for their generous feedback on earlier drafts of this article.

² Syme, R., Tacitus (Oxford, 1958), 122Google Scholar, McGing, B.C., ‘Synkrisis in Tacitus' Agricola’, Hermathena 133 (1982), 15–25Google Scholar, at 18, and Rodgers, R.H., Frontinus. De Aquaeductu Urbis Romae (Cambridge, 2004)Google Scholar, 2 all note how ‘rapidly’ Tacitus passes over Frontinus' governorship of Britain; this has led them, and others, to do likewise. They are among the very few scholars to comment on it at all.

³ There was nothing new in this, of course; the same emphasis on discontinuity and transformation can be observed at the start of the Flavian ‘era’, Nero's ‘golden age’, and when Augustus came to power (Syme [n. 2], 217). On political periodization in Trajanic literature, see esp. Ramage, E.S., ‘Juvenal and the establishment: denigration of predecessor in the Satires’, ANRW 2.33.1 (1989), 640–707Google Scholar; also C.L. Whitton, ‘The rhetoric of accession: Tacitus' early historical works as Trajanic legitimation’ (Diss., Cambridge University, 2008), 19–28, whose particular focus is the Agricola; cf. Wilson, M., ‘After the silence: Tacitus, Suetonius and Juvenal’, in Boyle, A.J. and Dominik, W.J. (edd.), Flavian Rome: Culture, Image, Text (Leiden, 2003), 523–42Google Scholar.

⁴ On this point, see esp. Sailor, D., Writing and Empire in Tacitus (Cambridge, 2008), 52, 60–2, 110–12CrossRefGoogle Scholar; also Hedrick, C.W., History and Silence: Purge and Rehabilitation of Memory in Late Antiquity (Austin, 2000), 113–130, 153–69Google Scholar.

⁵ Sailor (n. 4), 53. Tacitus makes the suppression of writing and speech an important feature of his depiction of Domitian's reign (Agr. 1–2). His publication of a biography which (he claims) could not have been written under Domitian thus marks a striking difference between the two eras. On this, see esp. Sailor (n. 4), 67; Whitmarsh, T., ‘“This in-between book”: language, politics and genre in the Agricola’, in McGing, B.C. and Mossman, J. (edd.), The Limits of Ancient Biography (Swansea, 2006), 305–33, at 311Google Scholar; Haynes, H., ‘Survival and memory in the Agricola’, Arethusa 39 (2006), 149–70, at 153–4CrossRefGoogle Scholar; Gibson, B., ‘Contemporary contexts’, in Roche, P.A. (ed.), Pliny's Praise: The Panegyricus in the Roman World (Cambridge, 2011), 104–24, at 109CrossRefGoogle Scholar. Cf. Rutledge, S., ‘Reading the prince: textual politics in Tacitus and Pliny’, in Dominik, W.J., Garthwaite, J. and Roche, P.A. (edd.), Writing Politics in Imperial Rome (Leiden, 2009), 429–46, at 438–9CrossRefGoogle Scholar.

⁶ On the Agricola's celebration and promotion of the new Nervan/Trajanic ‘era’, see e.g. Syme (n. 2), 125; Sailor, D., ‘Becoming Tacitus: significance and inconsequentiality in the prologue of Agricola’, ClAnt 23 (2004), 139–77, at 140Google Scholar; and Whitton (n. 3), 42–3 (who suggests that it may even have served as a model for Pliny's Panegyricus). Of course, the literature of the period exaggerates the extent of the discontinuity between Domitian and his successors, though (as Saller, R., ‘Domitian and his successors: methodological traps in assessing emperors’, AJAH 15 [2000], 4–18Google Scholar, underlines) the limited extent of other evidence makes it difficult to know by how much. For an overview of the propaganda and policies of Nerva and Trajan – and the tricky balance they each struck between repudiation of and continuity with various aspects of Domitian's reign – see esp. Griffin, M., ‘Nerva to Hadrian’, in Bowman, A.K., Garnsey, P. and Rathbone, D. (edd.), The Cambridge Ancient History, vol. 11 (2000 ²), 84–131CrossRefGoogle Scholar, at 86–92 and 98–108.

⁷ On the political crises and conspiracies which marked his short reign, see e.g. Syme (n. 2), 8–11, 220–1; Brennan, T.C., ‘Principes and plebs: Nerva's reign as turning-point?’, AJAH 15 (2000), 40–66Google Scholar; Berriman, A. and Todd, M., ‘A very Roman coup: the hidden war of imperial succession, ad 96–8’, Historia 50 (2001), 312–31, at 315Google Scholar; Grainger, J.D., Nerva and the Succession Crisis of ad 96–99 (London, 2003), 51, 67CrossRefGoogle Scholar. And on the uncertainty surrounding who or what was to succeed him: Griffin (n. 6), 91, 94–6; Berriman and Todd (op. cit.), 324–31; Eck, W., ‘An emperor is made: senatorial politics and Trajan's adoption by Nerva in 97’, in Clark, G. and Rajak, T. (edd.), Philosophy and Power in the Graeco-Roman World (Oxford, 2002), 211–26CrossRefGoogle Scholar; Grainger (op. cit.), 96–108.

⁸ Syme (n. 2), 17–18; Griffin (n. 6), 102; Grainger (n. 7), 111–17.

⁹ Whitmarsh (n. 5), 313: ‘Trajan's new age is liminal, inchoate. Uncertain, even.’ See also Rutledge (n. 5), 438–9.

¹⁰ On this passage, see esp. Liebeschuetz, W., ‘The theme of liberty in the Agricola of Tacitus’, CQ 16 (1966), 126–39, at 133CrossRefGoogle Scholar; Bews, J.P., ‘Language and style in Tacitus' Agricola’, G&R 34 (1987), 201–11, at 204Google Scholar; Sailor (n. 6), 153–4; Whitmarsh (n. 5), 311–12. Haynes (n. 5), 158–60 notes a parallel with Plin. Ep. 8.14.9, where Pliny also ‘uses the metaphor of the body to express the distressed state of the body politic under Domitian’, but seems more optimistic about a swift recovery. On this letter (and its engagement with the Agricola) see also Whitton, C.L., ‘Pliny, Epistles 8.14: senate, slavery and the Agricola’, JRS 100 (2010), 118–39Google Scholar.

¹¹ Tacitus' claim (Agr. 1.4) that he has to seek permission ‘now’ (nunc) to write his biography, ‘so savage is the age and so inhospitable to virtue’, has also been read as a sign that he is not entirely persuaded by the rhetoric of change: e.g. Liebeschuetz (n. 10), 133; Whitmarsh (n. 5), 312. Cf. Ogilvie, R.M. and Richmond, I. (edd.), Cornelii Taciti de vita Agricolae (Oxford, 1967)Google Scholar, ad loc.; also C. Kraus and A. Woodman, A commentary on Tacitus' Agricola (Cambridge, forthcoming), who will argue that the ‘now’ at 1.4 can only refer to the Domitianic past, not the Trajanic present. On the difficulty of deciding exactly when this nunc refers to, see esp. Whitton (n. 3), 31–4. For me, attempts to pin it down overlook the salient point: as Sailor (n. 4), 58–9 notes, the ambiguity is there; and whether it is intentional or not, it is telling.

¹² e.g. Turner, A.J., ‘Approaches to Tacitus' Agricola’, Latomus 56 (1997), 582–93Google Scholar, at 592; Wilson (n. 3), 533; Haynes (n. 5), 162; Whitmarsh (n. 5), 313.

¹³ e.g. Sailor (n. 4), 66 and (n. 6), 158–60. Whitton (n. 3), 34–6 also stresses Tacitus' critical focus on his senatorial readers, especially at Agr. 2.3, where ‘our’ ‘passivity’ under Domitian comes under scathing scrutiny (dedimus profecto grande patientiae documentum …); it is precisely this slavish mentality which Tacitus suggests ‘we’ are still struggling to shake off at 3.1. See also Lavan, M., ‘Slavishness in Britain and Rome in Tacitus' Agricola’, CQ 61 (2011), 205–16, at 211–12CrossRefGoogle Scholar.

¹⁴ As Wilson (n. 3), 533 argues, ‘In complimenting Nerva and Trajan for their restoration of libertas, [Tacitus] puts pressure on the new Emperor to live up to the high standards of tolerance for which he is being acclaimed’. See also Sailor (n. 6), 153, who notes that the connection which Tacitus establishes between hostility to texts and ‘bad’ or oppressive emperors ‘leaves the present principes little to do but positively to support the present text’ – or appear Domitianic themselves. Ramage (n. 3), 643–4 makes a similar point in connection with Pliny's Panegyricus; on the coercive, or at least protreptic, nature of Pliny's praise of Trajan there, see also e.g. Radice, B., ‘Pliny and the Panegyricus’, G&R 15 (1968), 166–72Google Scholar, at 168; Morford, M., ‘Iubes esse liberos: Pliny's Panegyricus and liberty’, AJPh 113 (1992), 575–93Google Scholar, at 578; S.M. Braund, ‘Praise and protreptic in early imperial panegyric: Cicero, Seneca, Pliny’, in Whitby, M. (ed.), The Propaganda of Power: The Role of Panegyric in Late Antiquity (Leiden, 1998), 53–76, at 66–8Google Scholar; Griffin, M., ‘Pliny and Tacitus’, SCI 18 (1999), 139–58, at 153Google Scholar.

¹⁵ Turner (n. 12), 592; Sailor (n. 6), 140, 154, 160; Sailor (n. 4), 70–2. Syme (n. 2), 125 also described the Agricola as a ‘manifesto … for the new imperial aristocracy’, as well as for the emperor Trajan.

¹⁶ Mart. 10.72; Plin. Pan. 2.3: non enim de tyranno sed de cive, non de domino sed de parente loquimur.

¹⁷ See esp. Plin. Pan. 62–77.

¹⁸ See above, n. 7; and below, n. 48.

¹⁹ On the autocratic reality behind the ‘republican façade’, see esp. Wallace-Hadrill, A., ‘Civilis princeps: between citizen and king’, JRS 72 (1997), 32–48Google Scholar, who notes that while ‘[d]oubtless there was a real contrast between the reign of a Domitian and a Trajan’ (39), the rhetoric of libertas and display of respect for the senate tended, above all, to confirm the supremacy of the emperor, without conceding any real power to his subjects (37). On gaps between Trajanic rhetoric and reality (and contemporary awareness of them), see also e.g. Syme (n. 2), 12, 131, 220–1; Waters, K.H., ‘Traianus Domitiani continuator’, AJPh 90 (1969), 385–405, at 394Google Scholar; Bartsch, S., Actors in the Audience: Theatricality and Doublespeak from Nero to Hadrian (Cambridge, MA, 1994), 166–7CrossRefGoogle Scholar; Bennett, J., Trajan Optimus Princeps: A Life and Times (London, 1997), 71–3, 106, 208Google Scholar; Griffin (n. 14), 154–5; Griffin (n. 6), 98–9, 103–8.

²⁰ On the Agricola as a personal or collective apologia for those who kept their heads down under Domitian, see e.g. Syme (n. 2), 25; Ogilvie and Richmond (n. 11), 17; Hedrick (n. 4), 166–9. The fact that Agricola is not a clear-cut hero complicates this reading (Bastomsky, S.J., ‘The not-so-perfect man: some ambiguities in Tacitus' picture of Agricola’, Latomus 44 [1985], 388–93Google Scholar; Whitmarsh [n. 5], 306; Haynes [n. 5], 163–9; Lavan [n. 13], 215–16). In fact Whitmarsh suggests that the tensions in the text (especially between the different models championed in the British and Roman sections) and the ambiguity surrounding its central character are so marked that it may amount to ‘a self-subverting critique [rather than a defence] of Tacitus and Agricola's complaisance’ (310).

²¹ See Clarke, K., ‘An island nation: re-reading Tacitus' Agricola’, JRS 91 (2001), 94–112Google Scholar, at 105–6 on Calgacus as a representative of ‘Old Rome’ and ‘Old Roman virtues’; 106 n. 34 on Calgacus as an embodiment of ‘what Agricola himself might have been like if he had not been a Roman general at the time of Domitian’; and 108 on Agricola as a ‘a hero of the old style’.

²² e.g. 6.3, where Tacitus characterizes Agricola's behaviour under Nero (Bastomsky [n. 20], 389–90; Whitmarsh [n. 5], 319–20); also 40.3–4, for an example of Agricola's unobtrusiveness under Domitian. Bews (n. 10), 208–10 discusses the ‘verbal pattern’ which Tacitus weaves around Agricola, and notes a contrast between the passivity which characterizes him when in the environs of the imperial court and his ‘energy, competence and incisiveness’ when campaigning in Britain.

²³ e.g. 5.1, 6.2, 6.5, 9.2–3.

²⁴ See esp. 19.2–3, where we learn that he led by example, transacting no public business via freedmen or slaves, making appointments not on the basis of personal bias but on merit, and finding out everything he needed to know.

²⁵ See e.g. 7.3, 9.2–3, 19.3–20.1 for examples of his combination of justice and humanity, his work against corruption and his imposition of order on chaos.

²⁶ There is a strong anti-corruption thread running through e.g. Plin. Ep. 10, but reminders too that the new regime is not excessively severe (e.g. 10.32, 38, 48, 82, 97, 111; Stadter, P., ‘Pliny and the ideology of empire: the correspondence with Trajan’, Prometheus 32 [2006], 61–76, at 61Google Scholar); see also Frontin. Aq., esp. 130 (discussed below).

²⁷ Again, see esp. Plin. Ep. 10 (e.g., 10.20, 38, 62, 99, where Trajan appeals to or commends Pliny's diligentia); and Frontin. Aq., passim, esp. 1, 64.1, 103–13, 119–23.

²⁸ One virtue which Tacitus particularly associates with Agricola is moderatio (5.1, 7.3, 42.3; Liebeschuetz [n. 10], 126–7; Clarke [n. 21], 108); Wallace-Hadrill (n. 19), 41–2 notes that this is the quality which Pliny attributes most often to Trajan in the Panegyricus.

²⁹ On the role the text (self-consciously) plays in ensuring Agricola's immortality, see esp. Hedrick (n. 4), 115, 153; Harrison, S., ‘From man to book: the close of Tacitus' Agricola’, in Heyworth, S.J., Fowler, P.G. and Harrison, S. (edd.), Classical Constructions: Papers in memory of Don Fowler (Oxford, 2007), 310–19CrossRefGoogle Scholar; Sailor (n. 4), 103–10.

³⁰ On Tacitus' implication of everybody (except Agricola) in the desecration of Rome, see esp. Syme (n. 2), 25; Liebeschuetz (n. 10), 133–4; Bews (n. 10), 204; Whitton (n. 3), 35–6; and Whitton (n. 10), 126.

³¹ McGing (n. 2), 16 notes that Tacitus' synopsis is both brief and inaccurate, ‘and dwells as far as possible on the unsuccessful features of the Roman administration’, with a view to underlining Agricola's strengths by comparison; see also Sailor (n. 4), 83.

³² More political periodization, of course. Indeed, not only does Vespasian's arrival herald a sudden return to order and good government after the upheavals of a.d. 69; those governors whom Tacitus describes as less than satisfactory had all, significantly, been appointed by Nero (with the exception of the last, Vettius Bolanus, who was sent out by Vitellius [Tac. Hist. 2.65.2], and whose achievements – in Britain and elsewhere – had recently been lauded by Statius [Silv. 5.2.30–60]).

³³ Ogilvie and Richmond (n. 11), ad loc. note a parallel at Ann. 14.47.1; see also Sailor (n. 4), 52, 79. Surprisingly, most of those who have discussed Tacitus' description of Frontinus have overlooked this qualification, focussing only on the phrase magnus vir: e.g. R. Mellor, ‘The new aristocracy of power’, in Flavian Rome (n. 3), 69–101, at 89; Martin, R.H., ‘Tacitus on Agricola: truth and stereotype’, in Bird, J. (ed.), Form and Fabric: Studies in Rome's Material Past in honour of B.R. Hartley (Oxford, 1998), 9–12, at 9Google Scholar; DeLaine, J., ‘“De aquis suis”? The “commentarius” of Frontinus’, in Nicolet, C. (ed.) Les littératures techniques dans l'antiquité romaine: Statut, public et destination, tradition, Entretiens Hardt 42 (Vandoevres–Geneva, 1995), 117–45, at 137Google Scholar; and McGing (n. 2), 18.

³⁴ His choice of an impersonal verb, licebat, is telling, for it makes Frontinus the passive object of some external, institutional force, not (just) Vespasian personally. That said, Tacitus' claim at Agr. 3 that Nerva ‘has brought together things previously incompatible, namely the principate and freedom’ reminds us that imperial rule has always involved the loss of freedom, until now; and his mention of Helvidius Priscus at 2.1 even connects Vespasian (under whom he was executed) with the oppressive kind of tyranny (against both men and books) embodied by Nero and Domitian, who are not named but evoked in Tacitus' list of imperial victims.

³⁵ Tacitus' emotive statement thus puts a particular, and rather misleading, spin on Frontinus' career, which flourished under the Flavians, especially Vespasian (indeed, it may even have been Vespasian who brought him into the Senate: Syme [n. 2], 592, 790; Eck, W., ‘Die Gestalt Frontins in ihrer politischen und sozialen Umwelt’, in Wasserversorgung im Anitken Rom, vol. 1 [ed. Frontinus-Gesellschaft, Munich, 1982], 47–62, at 50–1Google Scholar; Mellor [n. 33], 88–92).

³⁶ See also e.g. 8.3, 9.4, 22.4.

³⁷ Magnus is used again of the deeds about which Agricola had maintained a modest silence just a few lines later (… tam magna tacuisset).

³⁸ Ogilvie and Richmond (n. 11), 285; Sailor (n. 4), 93–4.

³⁹ On Tacitus' ‘controversial’ argument, and his attitude to the martyrs he seems to denigrate here, see esp. Syme (n. 2), 24–5; Liebeschuetz (n. 10), 127–32; McGing (n. 2), 22–3; Turner (n. 12), 590–2; Whitmarsh (n. 5), 308–10.

⁴⁰ As Sailor (n. 4), 114–15 points out, Tacitus has a tricky balance to strike here: ‘On the one hand, he needs Agricola to have been killed by the princeps, to prove that relations between the two were not too cosy … Yet Tacitus also needs Agricola not to have been killed, in order to make a distinction between the paths of prestige, that is, in order to make clear that there is a glory that does not consist only in the moment of death at the hands of the princeps.’

⁴¹ Sailor (n. 4), 107 and Martin (n. 33), 11–12 note Tacitus' repeated use of magnus for Agricola. When Tacitus first employs it (18.5) it is to tell us that others thought him great; Tacitus then calls Agricola ‘great’ himself, but in the context of others not fully appreciating his greatness; by the end, he seems confident that his readers will have been convinced and be themselves happy to use the adjective for him.

⁴² Sailor (n. 4), 79 suggests that Tacitus' ‘passing observation’ on Frontinus ‘conjures up Nero's relationship to Agricola's early career’; I hope I have shown that it does much more than that.

⁴³ For Tacitus' career, see Syme (n. 2), 59–74; Birley, A.R., ‘The life and death of Cornelius Tacitus’, Historia 49 (2000), 230–47Google Scholar.

⁴⁴ On Frontinus' career, see esp. Eck (n. 35), 47–52; Rodgers (n. 2), 1–5.

⁴⁵ CIL 16.42; Inscr. Ital. 13.1.

⁴⁶ CIL 6.2222, 8.7066; Inscr. Ital. 13.1, 195.

⁴⁷ Rodgers (n. 2), 3–5; Eck (n. 7), 225–6; Grainger (n. 7), 124; Birley, A.R., The Fasti of Roman Britain (Oxford, 1981), 72Google Scholar; Bennett (n. 19), 76.

⁴⁸ On Frontinus' role in Nerva's adoption of Trajan as his heir, see e.g. Syme (n. 2), 16–17; Eck (n. 7), 219–26. Grainger (n. 7), 14 and 100 wonders if Frontinus had also been involved in choosing Nerva as the new emperor to take over from Domitian.

⁴⁹ Rodgers (n. 2), 8; Grainger (n. 7), 118; DeLaine (n. 33), 132. Prior to that (in 97) Frontinus had also been part of a senatorial commission set up by Nerva to identify possible economies in the management of the state (Rodgers [n. 2], 3–4; Grainger [n. 7], 56). On Frontinus' more general influence with Trajan, see also Syme (n. 2), 49–50.

⁵⁰ See Rodgers (n. 2), 5–8 on the likely publication date.

⁵¹ See esp. Aq. 2.2–3, where Frontinus goes out of his way to present the treatise as a self-instruction manual.

⁵² See esp. Evans, H.B., Water Distribution in Ancient Rome: The Evidence of Frontinus (Ann Arbor, 1994), 53–64CrossRefGoogle Scholar; DeLaine (n. 33); Peachin, M., Frontinus and the Curae of the Curator Aquarum (Stuttgart, 2004)Google Scholar. Rodgers (n. 2), 12–14 provides a useful survey of the most recent interpretations of the text.

⁵³ Aq. 1: cum omnis res ab imperatore delegata intentiorem exigat curam, et me seu naturalis sollicitudo seu fides sedula non ad diligentiam modo verum ad amorem quoque commissae rei instigent sitque nunc mihi ab Nerva Augusto, nescio diligentiore an amantiore rei publicae imperatore, aquarum iniunctum officium ad usum … primum ac potissimum existimo, sicut in ceteris negotiis institueram, nosse quod suscepi.

⁵⁴ On this possibility, see Grimal, P., Frontin: Les aqueducs de la ville de Rome (Paris, 1944), xvGoogle Scholar; Hodge, A.T., Roman Aqueducts and Water Supply (London, 2002 ²), 16–17Google Scholar; Peachin (n. 52), 116 n. 95; Rodgers (n. 2), 16; Gibson (n. 5), 110. Cf. Bruun, C., The Water Supply of Ancient Rome: A Study of Roman Imperial Administration (Helsinki, 1991), 15 n. 13Google Scholar.

⁵⁵ Frontinus returns time and again to the theme of past incompetence and corruption: e.g. 31–4, 65–7, 72–3, 75–6, 91 (Evans [n. 52], 57–8; Cuomo, S., ‘Divide and rule: Frontinus and Roman land surveying’, SHPS 31 [2000], 189–202, at 193–4Google Scholar; Peachin [n. 52], 109–13 and Appendix 7). But he reminds us repeatedly that he and Nerva are busy combating it (e.g. 9.4–7, 64 and 87.1–3); and the order which he systematically imposes on all of his data supports the impression that there is a new level of order and accountability in the running of the water supply (Cuomo, op. cit. 193–4). See also 118.3, where – in a neat parallel of what they have been doing with the water itself – we see Nerva and Frontinus redirecting income from water rentals away from the emperor's private ‘coffers’ (which is where, Frontinus claims, it had ended up under Domitian) and back into public funds (on this passage, see Griffin, M., ‘The Flavians’, in Bowman, A.K., Garnsey, P. and Rathbone, D. [edd.], The Cambridge Ancient History, vol. 11 [2000 ²], 1–83, at 74Google Scholar; Turner, A., ‘Frontinus and Domitian: laus principis in the Strategemata’, HSPh 103 [2007], 423–49, at 438–9Google Scholar; Gibson [n. 5], 111).

⁵⁶ On this chapter, see esp. Peachin (n. 52), who underlines (among other things) the importance the text seems to place on re-establishing ‘proper’ relations between the emperor and the elite (138, 140).

⁵⁷ Ramage (n. 3), 658; Whitton (n. 3), 21.

⁵⁸ Frontinus uses the same vocabulary – diligentiam, amorem; diligentiore, amantiore – to characterize his and Nerva's approach (see n. 53). DeLaine (n. 33), 129–30 argues that the comparatives place Nerva slightly above Frontinus. Even so, the sentence works so that Frontinus' diligence and devotion come first, and are then matched by his emperor's, almost as if Nerva is taking the lead from Frontinus (as well as the other way around). Thus the syntax, which alternates between the two men, shows them bound up in a mutually instructive relationship. See also 64.1 and 118.3, where further references to Nerva's ‘diligence’ and ‘justice’ are similarly preceded or followed by mention of Frontinus' own cura and sedulitas; and Chicca, F. Del, Frontino. De aquae ductu urbis Romae (Rome, 2004)Google Scholar, ad loc.

⁵⁹ Frontinus' language here is revealing: the ‘officiously plural verbs’ (Rodgers [n. 2], ad loc.) in this chapter may nod to the regime which Frontinus represents, but they are authorial plurals all the same; and they are framed by singular verbs, which stress Frontinus' agency. Note also 101.4, where Frontinus' refusal of the lictors which the senate traditionally granted to Rome's curatores spurns mere senatorial authority, asserting instead his own integrity and the authority which he derives from the emperor (which his syntax places on an equal footing with his own: fides nostra et auctoritas a principe data pro lictoribus erit).

⁶⁰ Evans (n. 52), 59–61; DeLaine (n. 33), 135; Peachin (n. 52), 75–7; Rodgers (n. 2), 17.

⁶¹ I discuss this more fully in König, A., ‘Knowledge and power in Frontinus' On Aqueducts’, in König, J. and Whitmarsh, T. (edd.), Ordering Knowledge in the Roman Empire (Cambridge, 2007), 177–205CrossRefGoogle Scholar; on the authority which his specialist knowledge confers, see also DeLaine (n. 33), 127–8, 139; Peachin (n. 52), 64.

⁶² Frontinus' repetition of phrases like ad caput inveni/invenerim becomes almost formulaic at 64–72; see also e.g. 94–6, where invenio/inveni repeatedly introduce more research; and 73–4, where his first-person singulars help him to stand out from everyone else in having discovered such discrepancies that will amaze the reader. The repeated authorial plurals (diximus, posuimus, deprehendimus and variations) at 65–72 have a similar effect, testifying again to his command of his subject.

⁶³ Bruun (n. 54), 15, 18; DeLaine (n. 33), 124–5; Rodgers (n. 2), 23–4.

⁶⁴ DeLaine (n. 33), 129: ‘By presenting the aqueducts as one of the wonders of the world, Frontinus shows that the post of curator aquarum is one of paramount importance … Some of the glory goes, of course, to the emperor, but the overall effect of the de aquis is to show that most of it goes to the curator … In his role as curator aquarum, Frontinus is, then, in a way, acting for, if not as, the emperor himself.’

⁶⁵ Rodgers (n. 2), 125 does not see anything suggestive in Frontinus' use of the term principes here; but DeLaine (n. 33) is more open to the possibility, noting (136) that throughout the text Frontinus seems to cast himself in the role of princeps senatus, ‘under an emperor who was no longer dominus but princeps’.

⁶⁶ Evans (n. 52), 61.

⁶⁷ On the De aquaeductu as an expression of Frontinus' (and perhaps wider) senatorial aspirations, see also DeLaine (n. 33), 131–3, 136; Cuomo (n. 55), 195; and Rodgers (n. 2), 14–18.

⁶⁸ In this respect, the De aquaeductu has something in common with Plin. Ep. 10, which also presents an idealizing (and perhaps even protreptic) image of a model senator-cum-administrator working in close partnership with a model emperor (on this aspect of Ep. 10, see esp. Woolf, G., ‘Pliny's province’, in Bekker-Nielsen, T. [ed.], Rome and the Black Sea Region: Domination, Romanisation, Resistance [Aarhus, 2006], 93–108Google Scholar; Stadter [n. 26]; Noreña, C.F., ‘The social economy of Pliny's correspondence with Trajan’, AJPh 128 [2007], 239–77Google Scholar).

⁶⁹ Although it is possible that sections of it may have been circulated prior to publication (perhaps at the kinds of literary gatherings we encounter in e.g. Plin. Ep. 1.13; 7.17; 8.21). DeLaine (n. 33), 137 suggests that Frontinus' political prominence would have made the De aquaeductu ‘a must for the senatorial bookshelf’. Peachin (n. 52) also assumes a large (elite) readership.

⁷⁰ On Martial's distancing of himself from Trajanic Rome throughout Epigrams 10, see Fearnley, H., ‘Reading the imperial revolution: Martial, Epigrams 10’, in Flavian Rome (n. 3), 613–35, esp. 626–35Google Scholar. Baldwin, B., ‘Notes on the De aquis of Frontinus’, in Deroux, C. (ed.), Studies in Latin Literature and Roman History, vol. 7 (Brussels, 1994), 484–506Google Scholar, at 485 suggests that Martial ‘missed a golden opportunity’ in 10.58 to allude more directly to Frontinus' role as curator aquarum, or to any of his other public appointments. None the less, though they are not mentioned explicitly, Frontinus' industrious reputation and diligent approach to public office must add an extra piquancy to Martial's rejection here of Roman public life.

⁷¹ See also Ep. 5.1.6, where Pliny leans on Frontinus' reputation (as one of the ‘two most respected men of his era’ – duos quos tunc civitas nostra spectatissimos habuit) to guarantee his own integrity and authority in a lawsuit concerning a legacy.

⁷² On this letter, see Baldwin (n. 70), 486; DeLaine (n. 33), 137; Peachin (n. 52), 87–91; Rodgers (n. 2), 5; Del Chicca (n. 58), viii–ix. However disingenuous his sentiments may have been, the words attributed to Frontinus here show him emphasizing duty and public service over personal ambition, which tallies with his (self-promoting) self-presentation in the De aquaeductu.

⁷³ Pan. 61–2: ‘Indeed, I thought I was gazing on the great Senate of old when I saw a consul for the third time seated beside you’.

⁷⁴ Indeed, Whitmarsh (n. 5), 310 distinguishes between Tacitus himself and the persona he adopts in the text, and wonders whether ‘Tacitus the “hidden author” subvert[s] the surface-level position of Tacitus the narrator’.

⁷⁵ On the other hand, if the policy of quietism is presented as ambiguously in the Agricola as Whitmarsh (n. 5), suggests, Tacitus' characterization of Frontinus at 17.2 may not be quite so flattering after all.