Papirius Cursor, at least in Livy’s telling, held that the Dictator’s powers were superior and unrestricted with regard to his Magister Equitum, as well as all other magistrates. Let us examine this claim with a closer look at the status of the Consuls (and by extension the lower officers) when a Dictator had been named.
A curious detail involving Fabius Maximus in the Hannibalic War informs us that the Consuls were not to approach the Dictator with their lictors and insignia of office, such as toga praetexta and sella curulis, but essentially as privati (or so Plutarch puts it), and that he could enforce that rule by ordering them to comply—perhaps this was one of the rituals accompanying a Dictator’s “investiture,” such as the request for permission to mount a horse.1 Certainly (as Jeffrey Tatum reminds in a private communication) with sources that ultimately go back to the annalists, it is often unclear whether a given detail “was routine but not routinely included in narratives, or something singular.” But Livy—in a different context, the appointment of Iunius Pera as Dictator after Cannae, in 216—makes it clear that the request for permission to mount a horse was in fact routine (ut solet). In connection with Fabius’ appointment in 217, Livy mentions only the order that Servilius approach without his lictors: no word about other insignia or mounting the horse. Now Plutarch’s discussion of the taboo against mounting a horse occurs not on the occasion of Pera’s dictatorship (Fab. 9.4), but with Fabius Maximus here. Which makes it probable that the biographer was not simply elaborating on Livy’s account of Fabius’ accession but followed a common (Latin) source that spelled out the entire ceremonial, noting the Dictator’s twenty-four lictors (Fab. 4.3) and specifying the requirement that the Consul shed his lictors and other insignia, as well as the implications of their absence: without them the Consul was effectively rendered privatus. (The tenor of the entire passage is how Fabius strove to meet the appearance expected of a proper Dictator.) What precisely was the significance of a magistrate without his lictors?
In February of 49 bc, L. Domitius Ahenobarbus (cos. 54), newly appointed pro consule for Gallia Transalpina, surrendered to Caesar at Corfinium. A few days later, Atticus, then in Rome, in a letter to Cicero at Formiae had noted that he did not know whether Domitius (still) had the fasces with him; when he found out he would tell Cicero (ignoras Domitius cum fascibusne sit; quod cum scies, facies ut sciamus, Cic. Att. 8.15.1). Cicero in turn was eager to learn what Domitius’ next move would be, as the latter’s (and others’) example had a bearing on how to solve his own quandary: stay in Italy, or leave? (omnino ad id de quo dubito pertinet me scire quid Domitius acturus sit, quid noster Lentulus, 8.14.3.) As regards Domitius’ fasces, he seems to have already known, or assumed to know, the answer: in debating what to do, he points out that of those senators who were planning to leave Italy (or had left already), almost all except Ap. Claudius (cos. 54, cens. 50) were legally entitled to do so,2 because they either held imperium or served as Legates (memento praeter Appium neminem esse fere qui non ius habeat transeundi; nam aut cum imperio sunt, ut Pompeius, ut Scipio, Sufenas, Fannius, Voconius, Sestius, ipsi consules…aut legati sunt eorum, 8.15.3). Domitius is pointedly omitted from the list of those that hold imperium. As Caesar put it, Domitius at Corfinium, having thrown away his fasces and laid down his imperium, had come into another man’s power as a captive and privatus (BCiv 2.32.9, cum proiectis fascibus et deposito imperio privatus et captus ipse in alienam venisset potestatem). Cicero, it appears, took the same view.
Noting, however, “whether or not a magistrate was attended by lictors, his imperium persisted,” Tatum concludes that “something extraordinary happened to Domitius at Corfinium: in a letter to Atticus…Cicero refers to a general ignorance over whether or not Domitius still possesses his fasces (again, this had nothing to do with Domitius’ actual imperium).”3 On the contrary, it had everything to do with it. What possible interest could the question of Domitius’ fasces and their physical whereabouts, or his having them in his physical possession, hold for Atticus (and, presumably, others in Rome) if the answer had no bearing on Domitius’ official status? Why would Atticus consider the matter important enough to promise forwarding that information to Cicero as soon as he could obtain it (quod cum scies, facies ut sciamus)? In the letter, Cicero is wholly occupied with sorting out who was legally entitled—being cum imperio or someone’s legatus—to leave Italy or not, and what all of it implied for his own situation. To him and Atticus, wondering whether Domitius had his fasces was not a matter of idle curiosity.
As regards the persistence of a magistrate’s imperium with or without the presence of his lictors, Tatum may well be correct. Yet Domitius—and Cicero: the point is crucial—was not a magistrate, but pro magistratu, and, in that case, having lictors evidently made a difference (as will be seen further below). Nor can it be maintained that “Domitius remained a promagistrate, as even Caesar recognized when he allowed him to take away from Corfinium his public funds.” HS sexagies…Domitio reddit, etsi eam pecuniam publicam esse constabat (Caes. BCiv 1.23.4): Caesar let him have the money even though (etsi) it was comprised of public funds. The implication is clear: in Caesar’s view, Domitius had no claim to the money, being no longer a promagistrate. (In all likelihood, Caesar did not recognize Domitius’ appointment as Proconsul in the first place: audio enim eum ea senatus consulta improbare quae post discessum tribunorum facta sunt (Cic. Att. 11.7.1.).
The consularis disertus himself had a tale to tell when it came to lictors. Acclaimed imperator in Cilicia in 51 (Att. 5.20.3), Cicero returned to Italy late in the following year, his lictors carrying fasces duly wrapped in laurel. Yet as the Civil War broke out, he soon found his lictors an encumbrance, not to say embarrassment. Having reached the outskirts of the City on January 4, he saw the vote on his much-expected triumph postponed (Fam. 14.11.3); on the 18th, on news of Caesar’s rapid advance, he left for Campania before dawn, to avoid talk or spectacle, lictoribus praesertim laureatis (Att. 7.10): he—as others—was keenly aware of the incongruity of an imperator moving away from Rome with laureled fasces, especially when claiming a triumph. Within a few days, as it became clear that Pompeius might not put up a fight in Italy but go overseas, Cicero was wrestling with a difficult decision: how far should he commit himself? Some eminent friends had already drawn the line at leaving Italy, and his lictors posed an obstacle (scribe…si Pompeius Italia cedit, quid nobis agendum putes; M’. quidem Lepidus…eum finem statuit, L. Torquatus eundem; me cum multa tum etiam lictores impediunt, 7.12.4). Among the considerations that advocated staying were the inclement time of year, his lictors, and the incompetence of the republican leaders (ad manendum hiems, lictores, improvidi et neglegentes duces, 7.20.2). By mid-February, he could see the day when the City would be full of men of substance, including not a few consulars, that chose to stay behind: and he would join them, were it not for those awkward lictors (etsi prope diem video bonorum, id est lautorum et locupletum, urbem refertam fore…quo ego in numero essem, si hos lictores molestissimos non haberem, nec me M’. Lepidi, L. Vulcati, Ser. Sulpici comitum paeniteret, 8.1.3). Then again, to leave Italy with his laureled fasces, those foot-chains, would be just as awkward (age iam, has compedes, fascis, inquam, hos laureatos efferre ex Italia quam molestum est! 8.3.5). Better, perhaps, to stay and join Caesar—except that here, too, the fasces are a problem: what if Caesar were to offer him his triumph? To reject it could be dangerous, to accept would put him in bad odor with the boni (8.3.6). On March 6, finally, a decision: first from Formiae to Arpinum, then on to the Adriatic to join Pompeius—still at Brundisium—and the boni; the lictors he would put aside, or dismiss altogether (remotis sive omnino missis lictoribus, 9.1.3).
Yet two days later, Cicero had second thoughts (9.2a). By March 13, he thought it best to remain for now at Formiae and meet with Caesar, in hopes of being allowed to remain in Italy and out of public view, or come to Rome but not be forced to participate in measures against Pompeius; and he would readily let go of the triumph (de triumpho tibi adsentior, quem quidem totum facile et libenter abiecero, 9.7.2–5; cf. 9.9.1; 9.11A.3; 9.15.1). Another five days, and he was again determined to leave and join Pompeius (nunc si vel periculosum experiundum erit, experiar certe ut hinc avolem, 9.10.3; cf. 9.12.4), though not right away—he still wanted to talk with Caesar first. That interview, on the 28th, did not go well (9.18), and Cicero now pursued his plan to depart from Italy as soon as the season permitted and opportunity offered (9.19.3). By April 3, however, a new twist developed: he would go overseas, but seek a neutral location rather than join the optimates’ camp (10.1.2). Then again, perhaps it was more honorable to either follow the boni or come out openly against the scoundrels (10.1a). On April 14, C. Scribonius Curio paid a visit; Cicero explained that he wanted to leave Italy for a place of retirement and solitude—especially as he had his lictors (ego me recessum et solitudinem quaerere, maxime quod lictores haberem, 10.4.10), five weeks after indicating that he would get rid of them (9.1.3). On the 3rd or 4th of May, he told Caelius Rufus that he would happily stay in Italy, if it were not for that bothersome train of lictors and his title of imperator (Fam. 2.16.2, accedit etiam molesta haec pompa lictorum meorum nomenque imperi quo appellor); the laurel on his fasces gave rise to ill-willed looks and comments. About the same time, he had written in a similar vein to Antony: he wanted to leave Italy, being unwilling to hurry back and forth with his lictors (Att. 10.10.1, me autem, quia cum lictoribus invitus cursarem, abesse velle). On June 7, Cicero sailed for Greece (Fam. 14.7).
He returned to Italy, at Brundisium, well over a year later, in the fall of 48, still with his lictors, although on making his way through the crowd when approaching the town, he had them exchange their fasces for a single staff each,4 to avoid trouble with the troops stationed there; since then, he remained indoors (Att.11.6.2, quos [sc. lictores] ego †non paulisper† cum bacillis in turbam conieci ad oppidum accedens, ne quis impetus militum fieret; reliquo tempore me domo tenui). And how was he to come closer to Rome, as Atticus advised, without the lictors which the People had granted him, and which could not be taken away from him as long as he retained his status (Att.11.6.2, propius accedere, ut suades, quo modo sine lictoribus quos populus dedit possum? qui mihi incolumi adimi non possunt)? By mid-December, though, things were looking up: Balbus and Oppius, Caesar’s men in Rome, told Atticus that Cicero should continue to use the same lictors as he had now (ita faciam igitur ut scribis istis placere, isdem istis lictoribus me uti, 11.7.1).5 And so it happened. The man who was sole imperator in the empire of the Roman People permitted Cicero to be the other one, and keep his laureled fasces for as long as he should wish (Lig. 7, qui me, cum ipse imperator in toto imperio populi Romani unus esset, esse alterum passus est; a quo…concessos fascis laureatos tenui quoad tenendos putavi). For another year, Cicero remained at Brundisium. In October of 47, he returned to Rome, and finally dismissed his lictors before crossing the pomerium.6
The lictors with their fasces symbolize the magistrate’s powers, coercitio in the most immediate manner, and imperium more generally; that much is not in doubt.7 Yet the instances assembled above point to something more. Cicero had left Rome for Cilicia in the summer of 51, cum imperio pro consule. Under Sulla’s legislation, and by long-established practice, he retained his imperium—not his province!—until he returned to Rome and crossed the pomerium: that, too, is not in doubt.8 Once the outbreak of civil war had rendered a triumph doubtful and, if ever to be realized, a thing of the distant future, the lictors—laureled, no less—became an embarrassment, all the more in the case of a man who did everything to avoid being associated with a military command in those circumstances.9 One ought to think that, as symbols without substance, they could have been dismissed without much ado, and resumed in case a triumph did materialize. Yet Cicero clung to them for almost three years, long past the point at which he still entertained some hope of a triumph, and even when their public appearance posed a danger to his safety.10 The reason is not far to seek. Cicero explicitly connects his lictors with his nomen imperii (Fam. 2.16.2), with his status cum imperio pro consule, granted by the People (Att. 11.6.2): the lictors could not be taken from him without depriving him of this status as well. No lictors, no imperium. (He made the same connection with regard to Domitius, who relinquished his fasces at Corfinium.)
In 185 bc, the Consul M. Sempronius Tuditanus had drawn by lot the presidency over the elections; but his colleague, Ap. Claudius Pulcher, returned to Rome from their joint province, Liguria, first: his brother Publius was seeking the consulship, with scant prospects of success, given that year’s competition (Livy 39.32.5–9).11 Back in the City, Appius canvassed for his brother without lictors, and with abandon, which triggered vocal criticism from competitors and within the Senate: Claudius ought to remember that he was Consul first and foremost, not Publius’ brother (Claudius consul sine lictoribus cum fratre toto foro volitando, clamitantibus adversariis et maiore parte senatus meminisse eum debere se prius consulem populi Romani quam fratrem P. Claudi esse, 39.32.10–11).
When Appius campaigned for his brother in this fashion, he evidently meant to separate his canvassing activity from his official position as the Consul12—yet that precisely is the criticism directed at him: in his eagerness to support his brother, he had forgotten that he was Consul. And for the Consul, the only proper role during elections was to sit on his tribunal and either conduct them or be a silent spectator: quin ille sedens pro tribunali aut arbitrum aut tacitum spectatorem comitiorum se praeberet (Livy 39.32.11).13 Only as a privatus could Appius campaign for his brother without impropriety; whatever he did while appearing to exercise the consularis potestas could not be separated from official action as Consul. Hence no lictors. Without them, it would be inappropriate for him to exercise his powers, and thus, without them, he was free to act like any privatus. But by ridding himself of his lictors, he effectively ceased to function—for the duration of that condition—as Consul, thus opening himself to charges that he had put family interest ahead of his public duties, in effect suspending, for the time being, his office, while yet retaining it: for obviously, he still remained Consul. If symbolism without substance were at stake here, one should expect the attempt thus to separate his private persona from his public one to have drawn accusations of abusing his consular powers, rather than opprobrium for neglecting them.
In 62 bc, the Senate decreed that one of the Praetors, C. Iulius Caesar, should be removed from managing public affairs (donec ambo14 administratione rei publicae decreto patrum submoverentur, Suet. DIul 16.1). At first refusing to comply (nihilo minus permanere in magistratu et ius dicere ausus), Caesar quickly bowed to the threat of armed violence: he dismissed his lictors, shed the scarlet-bordered toga, and withdrew to his home (ut comperit paratos, qui vi ac per arma prohiberent, dimissis lictoribus abiectaque praetexta domum clam refugit). But a few days later, he defused an angry crowd demonstrating in his support outside his house, which prompted the Senate to repeal the decree and reinstate him (senatus…in integrum restituit inducto priore decreto, 16.2).
Caesar’s actions conform to the view developed here. The Senate had voted him removed from managing the res publica; but the Senate lacked the constitutional means to enforce such a removal: they must of necessity rely on the magistrate’s cooperation.15 Hence, when Caesar balked, the threat of using violence. Dismissing his lictors was not merely a symbolic action, necessary only to show that Caesar did not want to extend his conflict with the Senate: there was no other way he could remove himself, effectively and visibly, from public affairs.16 As long as he had lictors, he exercised the praetoria potestas in anything he did: by definition and default, he was involved in the administratio rei publicae.
In 58 bc, Cicero spent most of his exile in Macedonia, sheltered by his friend the Quaestor Cn. Plancius. While the governor, L. Appuleius (pr. by 59), maintained a neutral distance towards the exsul, Plancius met him at Dyrrachium, without lictors or insignia (lictoribus dimissis, insignibus abiectis), and escorted him to his official residence in Thessalonica (Thessalonicam me in quaestoriumque perduxit, Cic. Planc. 98–99). Plancius persisted in his solicitous attitude for the duration of Cicero’s stay, rejecting the role of Quaestor for that of a companion (abiecta quaestoris persona comitisque sumpta, 100).
Like Ap. Claudius, the Consul in 185, Plancius was at pains to separate his public persona from his private one, and, like Appius, his endeavor brought only limited success. In meeting Cicero without lictors and insignia, Plancius wanted to be seen as not exercising his official powers when welcoming a fugitive, but effectively acting as a private citizen. But setting aside his powers for the time being was not the same thing as renouncing his office altogether: he remained Quaestor just as Appius had remained Consul (or Caesar, Praetor). Thus he still resided in the quaestorium, and indeed could find no better place for his exiled guest to stay.
How did Plancius come by his lictors, though? As a rule, Quaestors were not entitled to them (Varro apud Gell. 13.12.6). If Appuleius the governor was not (yet) in provincia when Cicero arrived, Plancius could have been Quaestor pro praetore, acting governor, surely with lictors; we know nothing in that regard, of course. But it seems that, in the first century, Legates in the provinces were commonly granted lictors by their commanders; even senators travelling abroad on private business under a legatio libera could expect to be so endowed.17 It would not surprise if provincial Quaestors, too, by this time had acquired lictors as a matter of course; again, there seems to be no evidence apart from this passage. Now, if a Quaestor held his lictors—as did the legati—merely by way of a grant from the provincial governor, rather than lege publica, dismissing them could hardly affect Plancius’ official appearance: of necessity, he must be seen as acting pro magistratu even without them, having no legal claim to such instruments. Dropping his quaestorian insignia would have offered greater meaning but, naturally, less spectacle.
Under normal circumstances, the magistrate was virtually inseparable from his lictors: they accompany him wherever he goes or stays, be it the Forum or the bath, the battlefield or a dinner party.18 No one was allowed to step between the lictor proximus (the one immediately preceding the magistrate, or otherwise standing next to him) and the magistrate, except the latter’s filius praetextatus—wearing the same outfit (Val. Max. 2.2.4). The only known exceptions are the private home (the magistrate’s own, or that of anyone he visited), where the lictors remain in the vestibule,19 and—perhaps—the brothel: si praecedentibus fascibus praetor deducetur in lupanar, maiestatem laedet, we are told (Sen. Contr. 9.2.17). Inside a private home, the magistrate is effectively privatus; and in the bordello, that is what he ought to be, should a visit become unavoidable.20 Manumission before the Praetor (that is, vindicta) may be secured anywhere he can be found, in the theater or the bath, not only pro tribunali; but it requires the presence of his lictors—unless the transaction occurs in the privacy of his home.21
Lastly, there is the ancient and well-known principle of limiting the active management of the res publica to one of the two Consuls at a time, alternating monthly in the City, and daily in the field (if with the same army). Only the Consul “in charge”—penes quem sunt fasces—is preceded by his twelve lictors, carrying the rods. His colleague goes about with an accensus leading the way; that much is clear, but little else.22 Both Livy and Cicero seem to imply the physical alternation of a single set of lictors.23 Dionysios, though, envisaged two sets of lictors: one with axes (πελέκεις) for the Consul in charge; the other with rods only, as some of his sources told it, for his colleague, or with some kind of clubs or staffs as well (5.2.1, τοῦ δ’ ἑτέρου δώδεκα ὑπηρέτας ῥάβδους ἔχοντας μόνον, ὡς δέ τινες ἱστοροῦσι, καὶ κορύνας).24 This has been thought confused and conflated with the removal of the axes intra pomerium, but Dionysios subsequently reports the latter under the measures ascribed to Poplicola.25 Now, Caesar as Consul in 59 bc is said to have revived—at least in part—the ancient custom of ordering his lictors to walk behind him in the off-months, rather than in front.26 That clearly implies two sets of lictors, as in Dionysios. It also implies that from some point on prior to the first century—when exactly remains unknown—both Consuls appeared in public preceded by their lictors at all times; but the evidence amply demonstrates that they continued to take turns, month by month, at actually administering the res publica: even though both now appeared with lictors before them, only one at any given time was considered, constitutionally, to be holding the fasces, and capable of independent action.27 Did the lictors accompanying the Consul not in charge carry fasces?
In 28 bc, Imp. Caesar, Consul for the sixth time and observing all the practices prescribed by mos since ancient times, handed over his bundles of rods—in accordance with what was due—to M. Agrippa his colleague, and himself used the others (τοὺς φακέλους τῶν ῥάβδων τῷ Ἀγρίππᾳ συνάρχοντί οἱ κατὰ τὸ ἐπιβάλλον παρέδωκεν, αὐτός τε ταῖς ἑτέραις ἐχρήσατο, Dio 53.1.1). The “others” (αἱ ἕτεραι) can only refer to the rods (αἱ ῥάβδοι), not to the bundles (οἱ φάκελοι): Dio draws a contrast here between fasces and virgae. As so often, Staveley saw the correct explanation more clearly than most: only the lictors of the Consul penes quem fasces carried a full set of rods, tied up in a bundle; those of his colleague carried something else, and visibly different.28 Staveley called this “dummy rods”; but Dio’s contrast between bundle and rod suggests that each such lictor carried only a single rod, not multiple ones. Lictors, it seems, usually carried the fasces over their left shoulder, while holding a shorter staff in their right hand, presumably to keep others at an appropriate distance from the magistrate.29 The staffs are surely identical with the κορύναι known from Dionysios, hence still carried by the lictors of the Consul not in charge:30 were they “the others”? But Dionysios—who would have seen it all in action—clearly thought of those lictors as carrying two distinct implements, ῥάβδοι and κορύναι. Now the lictors in Roman colonies and municipia did not carry fasces in the form of an actual bundle of rods, but a single staff (bacillum) each; and when Cicero, on his return to Italy in 48 bc, wanted to enter Brundisium with a minimum of public visibility, he had his lictors appear in the same manner.31 Almost certainly, this explains Dio’s “other” set: the lictors of the Consul not in charge did not carry a bundle (φάκελος) each, but a single rod (ῥάβδος) over their left shoulder, presumably along with a staff (Dionysios’ κορύνη) in their right hand. “What matters is not the lictors but the emblems they carried.”32
The Consul not in charge retained his capacity to obstruct—intercessio and obnuntiatio—his colleague’s actions; he retained full command of his own army, including the right to auspicate.33 But he was unable to initiate acts of state, except in agreement with his colleague and, of course, whenever the latter was unable to do so himself due to illness, death, or absence from the scene.34
It is evident that the physical presence of the lictors was closely associated with the magistrate’s ability to exercise the powers of his office, as the Consul in charge, or being perceived as magistratus rather than privatus. Yet, as so often, the principle does not neatly translate into rigid constitutional rules. For an incumbent magistrate, dismissing his lictors and shedding the trappings of office evidently did not amount to ending the latter: Appius remained Consul, Caesar remained Praetor, and Plancius, Quaestor. Premature termination of an annual magistracy required a formal renunciation under oath.35
The temporary relegation of lictors to a place behind the Consul during the periods in which he did not hold the fasces signified an incapacity for independent action; but their physical—and visible—presence, along with the antecedence of the accensus, made it clear that he was not acting in any capacity other than his official one. But to appear in public without lictors or other insignia signalled that the magistrate, on this occasion, was not to be considered such: whatever he did while without them was to be seen as the action of a privatus, not an exercise of official powers. In effect, a magistrate without his lictors was deemed suspended from office, be it as part of a regular rotation or for some other reason.36 Even if in theory the absence itself of lictors did not render the magistrate unable to exercise his powers, it made little difference: as a practical matter, in a culture so steeped in ritual, it would be well-nigh unthinkable for him to do so except in the most routine kind of business.37
A man commanding pro consule, although holding imperium consulare and the outward trappings of curule office, and in that regard the equal of a Consul,38 clearly was not in possession of the full consularis potestas: he lacked, for instance, the ability to convene the People or the Senate even extra pomerium, or to engage in any governmental activity within the City. In fact, his imperium existed only outside the pomerium; to cross into the City meant the end of it, without further ado and ceremony. Hence he could not have lictors in the City, and must dismiss them before entering it: he crossed the pomerium as privatus, not pro magistratu.39 Yet something more is clear from Cicero: to a promagistrate, the lictors were inseparable from his appointment. To dismiss them anywhere (even under duress, like Domitius at Corfinium) meant the end of his imperium, hence of his tenure.40
Even so, Cicero at one point appears to contemplate their removal as distinct from complete dismissal (Att. 9.1.3, remotis sive omnino missis lictoribus: in early March, 49): could he have gone without them temporarily without giving up his imperium and status pro consule? (A magistrate, as we have seen, certainly could.) Yet except for this one brief instance, Cicero never mentions that possibility, nor did he adopt what would seem to us the obvious course of action that would have solved the problem of the lictors’ inconvenient and embarrassing presence. In fact, the distinction is imaginary; sive here, as commonly (and in Cicero especially), serves not to offer a true alternative, but to correct the first of the two elements it connects, or express it with greater precision: “with my lictors removed, or rather discharged altogether.”41
It is on the very occasion of Fabius’ dictatorship in 217 that we first encounter, in Polybios, the universal view of Greek writers (insofar as they comment on the matter) that all magistracies—other than the Tribunes of the Plebs—cease when a Dictator is appointed: οὗτος δ’ ἔστιν αὐτοκράτωρ στρατηγός, οὗ κατασταθέντος παραχρῆμα διαλύεσθαι συμβαίνει πάσας τὰς ἀρχὰς ἐν τῇ Ῥώμῃ πλὴν τῶν δημάρχων (3.87.8). Mommsen dismissed it as an error, albeit a puzzling one, coming from Polybios.42
It is clear enough from the Latin evidence that all magistrates stayed in office during a dictatorship. Yet it is difficult to believe that the Greeks, with such unanimity from Polybios (not noted for easily misunderstanding things Roman) on down, should have so thoroughly got this wrong, dependent as they all were, ultimately, on Roman sources of information. As so often, we may with greater benefit assume that they simplified the matter, but did not misrepresent it altogether.
Indeed, Polybios did not imagine that all other magistrates were actually terminated under a Dictator, and had to be elected de novo afterwards. He terms Servilius “the current Consul” (τὸν ὑπάρχοντα στρατηγόν43) when the Dictator deprives him of his command by land (3.88.8), and, after Fabius’ abdication, reintroduces Servilius and Atilius as “the previously serving Consuls” (οἱ προϋπάρχοντες ὕπατοι, 3.106.2), while noting that Atilius had been elected after the death of Flaminius—by implication, while Fabius was Dictator. He states expressly that the διάλυσις affected all magistrates except the Tribunes of the Plebs. The exception is significant. One is hard pressed to see how it could be the result of a misunderstanding: it rather shows that he was following good information that made the crucial distinction between (patricii) magistratus and tribuni plebis. He understood the matter as something other than removal from office.
Appian, commenting on the question in the same context as did Polybios, clearly imagined termination (ἕως ἀφικόμενος Φάβιος Μάξιμος ὁ δικτάτωρ Σερουίλιον μὲν ἐς Ῥώμην ἔπεμπεν ὡς οὔτε ὕπατον οὔτε στρατηγὸν ἔτι ὄντα δικτάτορος ᾑρημένου, Hannib. 12.50). Mommsen thought that this, like Plutarch Ant. 8.5 and QR 81, derived simply from Polybios, and was inclined to see in him the informant of Dionysios as well.44 The matter is perhaps more complicated.
Plutarch comments on constitutional aspects of the dictatorship on seven occasions, as follows:
(1) Marc. 24.11: ὁ γὰρ δικτάτωρ οὐκ ἔστιν ὑπὸ τοῦ πλήθους οὐδὲ τῆς βουλῆς αἱρετός, ἀλλὰ τῶν ὑπάτων τις ἢ τῶν στρατηγῶν προελθὼν εἰς τὸν δῆμον ὃν αὐτῷ δοκεῖ λέγει δικτάτορα.
“For the Dictator is not chosen by popular majority or by the Senate, but one of the Consuls or the Praetors, stepping in front of the People, names as Dictator the man whom he thinks best.”
The description of the mode of appointment stays close to the Latin terminology dictatorem dicere (thus also in the surrounding narrative: εἰπεῖν δικτάτορα, 24.10; λέγειν δικτάτορα and τοῦτον <ἀν>ειπεῖν, 25.1; ἀνεῖπε, 25.2). The biographer continues (24.12–13) with two explanations of the word dictator. One hails from dicere, in line with Varro’s and Cicero’s.45 The other derives the title from the Dictator’s not being subject to votes and elections, and his powers to issue ordinances (edicta) at his sole discretion: ἔνιοι δέ <φασι> τὸν δικτάτορα τῷ μὴ προτιθέναι ψῆφον ἢ χειροτονίαν, ἀλλ’ ἀφ’ αὑτοῦ τὰ δόξαντα προστάττειν καὶ λέγειν οὕτως ὠνομάσθαι· καὶ γὰρ τὰ διαγράμματα τῶν ἀρχόντων Ἕλληνες <μὲν> διατάγματα, Ῥωμαῖοι δ’ ἔδικτα προσαγορεύουσιν. With the possible exception of QR 81 (below, #8), this ought to be Plutarch’s earliest discussion of the office. It contains no hint of other magistrates’ ceasing to hold office during a dictatorship.
(2) Cam. 18.6: καίτοι πρότερόν γε καὶ πρὸς ἐλάττονας ἀγῶνας εἵλοντο πολλάκις μονάρχους, οὓς δικτάτορας καλοῦσιν, οὐκ ἀγνοοῦντες ὅσον ἐστὶν εἰς ἐπισφαλῆ καιρὸν ὄφελος μιᾷ χρωμένους γνώμῃ πρὸς ἀνυπεύθυνον ἀρχὴν ἐν χερσὶ τὴν δίκην ἔχουσαν εὐτακτεῖν.
“And yet they had previously, and for lesser contests, frequently chosen single commanders, whom they call Dictators—understanding full well how useful it is to be under the discipline, in a dangerous situation, of an unrestricted office holding the law in its hands, relying on a single judgment.”
(3) Fab. 3.6–7: πάντες δ’ εἰς μίαν γνώμην συνηνέχθησαν, ἀνυπευθύνου τε δεῖσθαι τὰ πράγματα μοναρχίας, ἣν δικτατορίαν καλοῦσι.
“All agreed unanimously that the circumstances required the unrestricted rule of a single man, which they call dictatorship.”
(4) Cam. 29.3: ἤδη γὰρ αὐτοῦ δικτάτορος ᾑρημένου καὶ μηδένος ἄρχοντος ἑτέρου νόμῳ, πρὸς οὐκ ἔχοντας ἐξουσίαν ὁμολογηθῆναι.
“For since he had already been chosen Dictator and there was no other magistrate under the law, the treaty had been concluded with those who had no power to do so.”
(5) Cam. 5.1: ἡ δὲ σύγκλητος εἰς τὸ δέκατον ἔτος τοῦ πολέμου καταλύσασα τὰς ἄλλας ἀρχὰς δικτάτορα Κάμιλλον ἀπέδειξεν.
“In the tenth year of the war, the Senate, having dissolved all the other magistracies, named Camillus Dictator.”
(6) Fab. 9.2: ὁ δὲ Μετίλιος ἔχων τὴν ἀπὸ τῆς δημαρχίας ἄδειαν—μόνη γὰρ αὕτη δικτάτορος αἱρεθέντος ἡ ἀρχὴ τὸ κράτος οὐκ ἀπόλλυσιν, ἀλλὰ μένει τῶν ἄλλων καταλυθεισῶν.
“Metilius enjoyed the safety granted by his being Tribune of the Plebs; for this office alone does not lose its power when a Dictator has been chosen, but continues even though all the other magistracies have been dissolved.”
(7) Ant. 8.4–5: Ἀντώνιον δ’ ἵππαρχον ἑλόμενος εἰς Ῥώμην ἔπεμψεν. ἔστι δ’ ἡ ἀρχὴ δευτέρα τοῦ δικτάτορος παρόντος· ἂν δὲ μὴ παρῇ, πρώτη καὶ μόνη σχεδόν· ἡ γὰρ δημαρχία διαμένει, τὰς δ’ ἄλλας καταλύουσι πάσας δικτάτορος αἱρεθέντος.
“Having picked Antonius as Master of the Horse, he sent him to Rome. This magistracy is the second highest when the Dictator is present; but when he is absent, it is the highest and almost the only one: for the tribunate of the Plebs remains, but the other magistracies are all dissolved once a Dictator has been chosen.”
(8) QR 81 (Mor. 283B): οὐδὲ παύονται (sc. οἱ δήμαρχοι) δικτάτωρος αἱρεθέντος ἀλλὰ πᾶσαν ἀρχὴν ἐκείνου μετατιθέντος εἰς ἑαυτὸν αὐτοὶ μόνοι διαμένουσιν, ὥσπερ οὐκ ὄντες ἄρχοντες ἀλλ’ ἑτέραν τινὰ τάξιν ἔχοντες.
“For the Tribunes of the Plebs do not cease to function when a Dictator has been chosen, but while he transfers any other office unto himself, they alone remain, precisely because they are not magistrates, but have some other status.”
In four (##5–8) out of eight discussions of the office, Plutarch explicitly states that all other magistrates cease to function under a dictatorship, and his description of the office as an “unaccountable monarchy” in two more (##2, 3) can accommodate such a view. A fifth passage (#4) implies that, under a Dictator, there are no other magistrates; yet the narrative here very closely follows Livy, who makes it clear that other magistrates still remained in office—though bereft of power: negat eam pactionem ratam esse quae postquam ipse dictator creatus esset iniussu suo ab inferioris iuris magistratu facta esset (5.49.2). In three of the instances stating cessation (##5–7), the verb employed (καταλύω) corresponds to Polybios’ διαλύομαι. Yet παύομαι (#8) may evoke more of an indefinite halt than complete termination, and the phrase μόνη αὕτη ἡ ἀρχὴ τὸ κράτος οὐκ ἀπόλλυσιν (#6), taken literally, implies that the other magistracies do not simply disappear, but—unlike the tribunate—lose their power. On three occasions (##6–8), Plutarch notes the same exception as did Polybios: the Tribunes of the Plebs. But his additional comment, in the QR (#8), that the exception applies because the Tribunes are not magistrates—again, coming from a source knowledgeable and accurate in such matters, if not revealing the biographer’s own understanding—along with the intriguing notion of the Dictator’s absorbing all other magistracies into himself (δικτάτωρος…πᾶσαν ἀρχὴν…μετατίθεντος εἰς ἑαυτόν) plainly cannot be read into what Polybios wrote.46 It is part of a lengthy and learned—indeed, highly percipient—disquisition on the constitutional character of the tribunate (Mor. 283B–D); he did not come upon the material for any of this in the extant work of Polybios. Nor could all of it derive from lost parts of the work: Curio’s quip about a Tribune being the public’s doormat (τὸν δὲ δήμαρχον…καταπατεῖσθαι δεῖ) was uttered a century after the historian wrote. The underlying thoughts, however, are fully compatible with what we have; if Polybios eventually did produce his promised treatment of the dictatorship in detail (οὐ μὴν ἀλλὰ περὶ μὲν τούτων ἐν ἄλλοις ἀκριβεστέραν ποιησόμεθα τὴν διαστολήν, 3.87.9), it will have come within a systematic description of Roman institutions after 6.18.47 Plutarch in that case is likely enough to have taken much of QR 81 from it. It is improbable that Polybios in that lost treatment would have expressed himself less clearly on the subject than he did in Book 3; unfortunately, we cannot assume a greater degree of clarity, either: it may never have occurred to him that readers could mistake his words for implying the actual abdication of all magistrates. (His narrative in Book 3 had implied the opposite.) Nor can we be certain that Plutarch thought of actual abdication in the instances he implies cessation (##5, 6, 7, and 8): the language, so close to Polybios’, supports a temporary loss of power as readily as termination.
Plutarch, though, comes close to suggesting abdication in the Fabius:
(9) Having been made Dictator, Fabius orders the surviving Consul to dismiss his lictors and meet him as a private citizen (καὶ τὰ παράσημα τῆς ἀρχῆς ἀποθέμενον ἰδιώτην ἀπαντᾶν, 4.3); and when the Dictator’s term is up, Consuls are elected again (14.1)—who turn out to be not those of 216, but an unnamed pair (in reality, Servilius and his suffect colleague), adhering closely to Fabius’ precepts.
That Plutarch imagined termination when he had Fabius dismiss Servilius as a private citizen is possible; something similar to Γνάιον μὲν τὸν ὑπάρχοντα στρατηγὸν ἀπολύσας τῆς κατὰ γῆν στρατείας (Polyb. 3.88.8) could have been misunderstood, in light of the Dictator’s ordering the Consul to approach without lictors. (Appian certainly thought the same.) But the “election” of new Consuls after the Dictator’s term was up may be more readily attributed to a sloppy reading of his source—probably not Polybios48 but the one that furnished him with additional details he could not have found in the latter or in Livy (Fabius requesting permission to mount a horse, his twenty-four fasces, and his ordering Servilius to approach him without lictors, 4.1–3). That source was well informed on minutiae of Roman ceremonial: it is unlikely to have presented Servilius’ dismissal as an actual abdication.
On the other hand, the comment in Antony (#7) fairly closely parallels Polybios in the information furnished, in particular as regards the Magister Equitum being not merely a second-in-command, but a deputy capable of—in the Dictator’s absence—independent action (οὗτος δὲ τέτακται μὲν ὑπὸ τὸν αὐτοκράτορα, γίνεται δ’ οἷον εἰ διάδοχος τῆς ἀρχῆς ἐν τοῖς ἐκείνου περισπασμοῖς, 3.87.9). One need not conclude, from its appearance only in a relatively late life, that the biographer had been unaware of it when he wrote the Fabius, although an explanation of the Master’s position might have been as useful in that life as in the Antony. (But Minucius was an incidental character, not the hero.)
Cessation—even termination—of other magistracies is also the case in Dionysios’ view. His long disquisition on the establishment of the dictatorship in 498 V repeatedly emphasizes it as characteristic of the office:
(10) The consular power is temporarily “removed,” for a maximum period of six months; afterwards, the Consuls again assume control of the government: ἔκρινε τὴν μὲν ὑπατικὴν ἐξουσίαν ἀνελεῖν κατὰ τὸ παρόν…χρόνου δ’ εἶναι μέτρον τῇ νέᾳ ἀρχῇ μῆνας ἕξ, μετὰ δὲ τὴν ἑξάμηνον αὖθις ἄρχειν τοὺς ὑπάτους (5.70.1–2).
(11) The Consuls and all others holding public office are to lay down their powers: τοὺς τότε ὑπατεύοντας ἀποθέσθαι τὴν ἐξουσίαν, καὶ εἴ τις ἄλλος ἀρχήν τινα εἶχεν ἢ πραγμάτων τινῶν κοινῶν ἐπιμέλειαν (5.70.4).
(12) The Consul Q. Cloelius, having named his colleague, T. Larcius, Dictator, promptly abdicates his office: ἀναστὰς ὁ Κλοίλιος ἀναγορεύει τε αὐτόν…καὶ τὴν ὑπατείαν αὐτὸς ἐξόμνυται (5.72.3).
(13) Having completed his task, Larcius the Dictator sees to the appointment of new Consuls, then abdicates his office; clearly, Dionysios does not envisage him and his (former) colleague resuming their consulship: ὑπάτους ἀποδείξας ἀπέθετο τὴν ἀρχήν (5.77.1).
(14) Frequently it has been necessary to eliminate the lawful offices and place the entire government under a single man: πολλάκις ἀναγκασθείσης τῆς πόλεως καταλῦσαι τὰς νομίμους ἀρχὰς καὶ πάντα ποιῆσαι τὰ πράγματα ὑφ’ ἑνί (5.77.2).
(15) The same view recurs in a much later context, on recalling Cincinnatus’ dictatorship in a speech directed against the Decemvirs: “Gathering in the Curia at midnight, you appointed a single office to exercise supreme command in war and peace, and in doing so you dissolved all the other offices” (περὶ μέσας νύκτας εἰς τὸ βουλευτήριον συνελθόντες ἀρχὴν ἀπεδείξατε μίαν αὐτοκράτορα πολέμου καὶ εἰρήνης, ἁπάσας τὰς ἄλλας καταλύσαντες ἀρχάς, 11.20.3). Yet in his actual Cincinnatus narrative, Dionysios had both Consuls remain in office (below, #16).
The first of these passages (#10) certainly, and two more (##14, 15) possibly, could be taken as implying a temporary cessation of Consuls and other magistrates, along the lines of Polybios. But another two (##12, 13) expressly speak of termination through abdication, and while #11 in isolation could be understood either way, depending on whether ἐξουσία is to mean “power” or “office,” its contextual proximity to ##12 and 13 renders abdication more likely by far. It would require a considerable stretch—much more so than Plutarch’s in ##7 and 8—to trace that notion to Polybios’ curt notice, or, for that matter, his longer treatise on the magistracies. Not that Dionysios was entirely consistent:
(16) Both Consuls stay in office under Cincinnatus as Dictator; the unfortunate L. Minucius is not forced to resign until the crisis has been resolved (ταῦτα πράξας καὶ τὸν Μηνύκιον ἀποθέσθαι τὴν ἀρχὴν ἀναγκάσας ἀνέστρεψεν εἰς τὴν Ῥώμην (10.25.2), and C. Nautius triumphs well after the Dictator has abdicated (10.25.4).
The notion that all magistrates are removed from office—as opposed to losing the ability to independently exercise their powers—under a Dictator is not Polybian, as we have seen; and although a careless use of his work might have led others to such a misconception, all the completely unambiguous expressions of that view other than Appian’s—which is to say, Dionysios at ##11, 12, and 13—occur in contexts that cannot be reduced to Polybios’ extant work. Another source, perhaps more directly concerned with the origins of the dictatorship, appears probable. That source, unfortunately, may still have been Polybios: his lost “archaeologia” of how Rome’s constitution developed, from the kings to the decemvirate,49 likely enough included the establishment of the dictatorship, and a fuller discussion of the office could have been found in his treatise of Roman institutions, if not directly in the “archaeologia.” Termination would still result from a misunderstanding of what Polybios meant; and perhaps of something else. Unless Dionysios made it up out of whole cloth, one of his sources must have reported the abdication of Q. Cloelius the Consul in 498 V after he named his colleague Dictator; but that source need not have presented it as a requirement of the dictatorship. That may have been the rhetorician’s own inference, “guided” by what he thought he had read in Polybios.
Nor—it must at last be noted—is the Polybian view entirely absent from the Latin evidence. In 390 V, Camillus stops payment of ransom, agreed to by Q. Sulpicius the Military Tribune, to Brennus and the Gauls: cum illi (sc. Galli) renitentes pactos dicerent sese, negat eam pactionem ratam esse quae postquam ipse dictator creatus esset iniussu suo ab inferioris iuris magistratu facta esset (Livy 5.49.2). If, with a Dictator in office, anything done by another magistrate—and in this case, one consulari potestate—can be deemed eo ipso invalid unless approved or confirmed by the Dictator,50 such other magistrate can obviously not be seen as being able to exercise his regular powers independently. If full weight is allowed to omne imperium, Cicero’s comment that, during a dictatorship, all public power was to be concentrated in the hands of a single man (Rep. 1.63, sine collega omne imperium nostri penes singulos esse voluerunt) supports precisely such a view. A closer look at the nature of the office held by that single man will come next.
The appointment of a Dictator threw the Consuls and all lower magistrates (save Tribunes of the Plebs) into a state of suspension. What were the terms that governed its appointment and duration, and what precisely were its powers compared to those of the Consuls?
Mommsen saw in the Dictator nothing but a collega maior of the Consuls, just as the Praetor was a collega minor. Corey Brennan’s discussion has rendered that view untenable; the Dictator, for one, was not elected—unlike the Praetor—under the same auspices as the Consul, and his complete absence from Messalla’s discussion of the auspicia maxima (those of Consuls, Praetors, and Censors) and their relation to, and interaction with, each other precludes any notion of collegiality.51 Instead, Brennan defined the Dictator’s imperium in terms diametrically opposite to Mommsen’s: rather than being of the same kind as the Consul’s, though maius, it was not technically maius but somehow “different”; indeed, the Dictator’s powers were in no way greater than the Consuls’.52
That latter contention, however, runs into difficulties of its own. It rests on two lines of reasoning. First, an axiomatic assertion: “The consul had the imperium and auspicia of the kings of Rome; logically, the dictator could have no more”—but where is it written that there can be no power greater than the King’s? Livy, for one, certainly thought that the Dictator’s power could be understood in precisely that fashion: the Dictator’s imperium is supreme, and the Consuls—the regal power!—do obey him (cum summum imperium dictatoris sit pareantque ei consules, regia potestas, 8.32.3).53
Next, Cicero’s provision in Leg. 3.9, oenus ne amplius sex menses…idem iuris quod duo consules teneto, isque ave sinistra dictus populi magister esto. Brennan, by implication of his argument, takes idem iuris…teneto for synonymous with aequum (sive par) imperium…habeto, and concludes that the Dictator had twenty-four lictors “to mark him clearly as the full equal of a consular pair, with the ability to give either consul orders”; yet, by the same token, he supposedly could not override both Consuls together. This contradicts everything we know about the Roman concept of imperium, which was never cumulative, but indivisible in the person of every holder: each Consul exercised the imperium consulare in full, not just one half of it, and two Praetors together possessed no more of imperium than did a third one by himself, alone. Nor did two Consuls have more power, in a quantitative sense, than one. Hence Cicero’s contrasting oenus (sc. populi magister) with duo consules cannot mean, as Brennan would have it, that the Dictator had “idem iuris, ‘just as much power’ as the two consuls, but no more”: either he held as much power as one Consul, in which case he could not “give orders” to any magistrate of that rank (being his equal), or he possessed more power than one Consul, in which case he could “give orders” to (which is to say, override, rescind, or forbid) any number of Consuls simultaneously—but such “more” power could not derive from a cumulation of the two Consuls’ individual powers, their sum being no greater than each of its components. It is not too difficult to see what Cicero had in mind: his magister populi had “the same legal status” as the two Consuls not in a quantitative sense, but in a qualitative one. In other words, he took their place.54
Almost certainly, Claudius Caesar furnished us with a clue (ILS 212) when he noted that in times of severe emergency, military or domestic, the ancients had found the Dictator’s imperium to be valentius—“stronger, more powerful, more effective”—than the Consuls’: dictaturae hoc ipso consulari imperium valentius repertum apud maiores nostros, quo in a[s]perioribus bellis aut in civili motu difficiliore uterentur. (Claudius’ phrasing is curiously reminiscent of Cicero’s, Leg. 3.9: ast quando duellum gravius discordiaeve civium escunt.) Brennan doubted that Claudius here alluded to a characteristic of the Dictator’s constitutional position, and preferred to ascribe valentius to the emperor’s “desire for literary variation.”55 There is no good reason to think so. The connotation of superior force in general inherent in valere, and of special ability to achieve a desired objective in particular, eminently fits the Dictator’s function as an extraordinary official appointed to deal with extraordinary situations. Nor does the word lack religious overtones; witness the faustum omen associated with the name “Valerius,” especially in a military context.56
Long ago, in a study that has drawn scant attention, Cohen conclusively demonstrated how the Dictator was more than simply an exalted version of the Consul: he possessed a religious, near-magical dimension altogether lacking in the latter.57 The Dictator’s edictum is said to be pro numine semper observatum, his imperium described as sua vi vehemens,58 and the office itself was thought to instill an elementary, visceral fear never associated with the consulship. Repeatedly the enemy withdraws, or sues for peace, or commits suicide, on the mere news that Rome has named a Dictator: tantus eius magistratus terror erat.59 That such ingens terror was also felt at home could be explained, perhaps, by the Dictator’s freedom from provocatio and tribunician veto; but it is out of the question that this constitutional peculiarity should be of concern to non-Romans, or that the Romans should expect it to cause fright and panic among others. It was a quality associated with the Dictator, and with him alone—not with any other magistrate.60
The title of the office may provide illumination. No other magistracy’s appellation (with the exception, perhaps, of consul) has defied understanding as tenaciously as the word dictator. The earliest surviving etymology has it derive from dicere: hinc dictator magister populi, quod is a consule debet dici (Varro LL 6.61), and dictator quidem ab eo appellatur quia dicitur (Cic. Rep. 1.63). Already Varro combines this with a second, albeit subordinate, explanation: dictator quod a consule dicebatur, cui dicto audientes omnes essent (LL 5.82). In a similar vein, and clearly following a Latin source, Dionysios and Plutarch connect the title to edicere/edictum, on the grounds that the Dictator could issue public ordinances (“edicts”) solely according to his own judgment; both offer the manner of appointment, by being “named,” as an alternative.61 Priscian simply has dictator…a dictando (Inst. 8.78 = GramLat 2: 432.25).
As Mommsen noted, nothing suggests that the Dictator was in any special way associated with issuing edicts—no more so than other magistrates with imperium.62 Varro’s derivation from dicere is generally rejected as absurd, seeing how the title, in that case, ought to be dict(at)us, not dictator. It has found endorsement, however, by no less an authority than Rix: being “named” rather than elected was indeed a feature unique to the Dictator, and a hypothetical appellation (praetor) dictatus could have been transformed to dictator by analogy with nomina agentis such as praetor, quaestor, or imperator.63 But Rix was unable to offer a Latin parallel for such an “Umgestaltung,” and his solution lacks conviction. Against the derivation from dictare (in the sense of cui dicto audientes omnes essent), finally, speaks the scant evidence that this verb expressed the notion of giving orders.
And yet dictare holds the answer. As Rix points out, the word in origin cannot have had the narrow meaning of “dictate, recite” in which it is mostly attested; it simply is an intensive variant of dicere, just as captare/capere, iactare/iacere, tractare/trahere. Now, students of Roman law and religion well know the magical properties of the spoken word, the elementary importance of the correctly recited formula or prayer. Varro (LL 6.30) sets out the fundamental concept:
dies nefasti, per quos dies nefas fari praetorem “do dico addico”: itaque non potest agi: necesse est aliquo uti verbo, cum lege quid peragitur. quod si tum imprudens id verbum emisit ac quem manumisit, ille nihilo minus est liber, sed vitio, ut magistratus vitio creatus nihilo setius magistratus.
If on a dies nefastus the Praetor, carelessly or by mistake, speaks the ritually prescribed words, the legal business thus transacted is valid nonetheless. The Consul is elected by the People, but what made a man a magistrate was not the People’s vote: only the formal announcement (renuntiatio) by the presiding officer lends legal force and effect to the assembly’s vote.64 This held true even if the election occurred under a ritual augural flaw (vitium). “The formula ‘do dico addico’ was more powerful than the sacral division of time”;65 the announcement L. Titius consul factus est overrode the fact that Iuppiter had not given his permission to hold elections on that day. In augury throughout, mere observation of a sign by a third party is of no consequence to its intended recipient: to have effect, it must be properly and formally reported; and even a false report, if properly made, becomes as valid and effective as if the sign had truly been observed.66 The spoken word creates reality.67
Here, surely, is where we must seek both the origin of the name dictator and the reason for his being “named.” He is “the One Who Speaks”: his word is more powerful than any obstacle or limit, more efficacious than any word spoken by anybody else; it overcomes all opposition and resistance, at home and abroad. His word is strong enough to shape reality in favor of the Roman People.
Yet the office was originally called magister populi. We know that appellation was used in early days, and always in the libri augurales (Cic. Rep. 1.63);68 we do not know for a fact that it was older than the term dictator, or that at any time it was used exclusively. We also know that, in a number of Latin communities, a dictator served as the annual chief magistrate, often associated with religious functions, but nowhere resembling the Roman emergency official; occasionally, the title occurs together with other (lower) officials, but never with a magister equitum.69 Licinius Macer believed that at Alba after the death of Numitor, an annual magistrate called dictator replaced the King while holding the same power as the latter, and that the Romans derived their dictatorship from Alba.70
There is also some—albeit tenuous—evidence that dictator at some point was the title of a Latin federal official. Priscian (Inst. 4.21 = GramLat 2: 129.11–15) preserves a fragment of Cato’s Origines (HRR F 58/FRHist 5 F 36): lucum Dianium in nemore Aricino Egerius Laevius Tusculanus dedicavit dictator Latinus; Festus probably refers to the same event (Manius Eger<ius lucum> Nemorensem Dianae consecravit, 128.15–16L). This has been taken to mean that the Latin League was headed by an (annual) magistrate called dictator, with civil, religious, and—especially—military authority who eventually may have stood model for the Roman Dictator.71 But all we know is that Egerius Laevius (or perhaps Baebius, as in some codices) dedicated the grove at Aricia, and that, although a citizen of Tusculum, he did so in his capacity as dictator Latinus. It follows that this was a federal office, but Cato’s note tells us nothing about its nature: the occasion reported manifestly does not imply military functions.
Nor is dictator entirely certain as Egerius’ title: one of Priscian’s codices (R = Parisinus 7496) has dicator. The same word occurs in an inscription from the third century bc, where at Spoletium we encounter a dicator empowered to levy fines on persons violating a sacred grove: honce loucom ne qui[s] violatod neque exvehito neque exferto quod louci siet…sei quis scies violasit dolo malo Iovei bovid piaclum datod et a(sses) CCC moltai suntod. eius piacli moltaique dicator[ei] moltai suntod.72 Mommsen wanted to see in this dicator not an official title but simply a reference to the dedicating magistrate and his successors in office. But in a general prohibition such as this, it would be highly unusual to focus on one specific dedicator, and if dicator indeed merely denotes the person dedicating the grove, then nothing in the wording implies that the power to levy fines devolves on anyone else. Surely the natural assumption should be that dicator is the title of a local magistrate.73 The question is: What was this magistracy?
Instinsky followed Mommsen in seeing in the dicator the dedicating magistrate and his successors, on which basis he preferred the reading in Priscian’s R as the correct one: the dicator (not dictator) Latinus Egerius Laevius was a Latin federal official, albeit not a regular one, but named specifically for the purpose of dedicating the grove at Aricia.74 Adducing against this view, however, multiple examples from both Greek and Latin glossaries that equate δικάτωρ/dicator with δικτάτωρ/dictator, Buecheler and Mazzarino have made a strong case that dicator represents a linguistic variant of dictator, not a separate institution.75
Whichever form one accepts for Egerius Laevius, it will be prudent to remain skeptical about the notion of a Dictator as the regular commander of Latin League forces. Without exception, the unambiguous references to the Latin federal military command describe it as a dual magistracy called praetores as far back as the days of King Tullus.76 On the other hand, there is no reason to avoid the clear implication of Cato’s testimony: there existed a dic(t)ator as a federal magistrate of the League, responsible for religious matters (and perhaps civil ones, if such there were), be it as a permanent office or one filled only for specific occasions.77
The apparent variation of dicator/dictator as an appellation for the same office would reinforce the view that the official thus named received his title from performing acts of speaking. In a largely preliterate society, such acts could encompass virtually anything to do with government and religion. Any directive or ritual would, of necessity, be spoken; but it is the act of speaking—and speaking it exactly in the correct form—that gives it effect.78 Therein lies the task, and the power, of the magistrate, and therefore he can be known simply as dic(t)ator, the Speaker. If the King—perhaps, yet by no means certainly—was endowed with the same ability, it was for life instead of just a year or a specific task.
We must allow, then, room for the possibility that dictator was, in fact, originally a title found both as that of a Latin federal institution and in many, though not necessarily all, Latin communities in which the King had disappeared. One cannot even rule out the possibility that in those towns which did have Kings (attested only, it seems, for Alba and Rome) a dictator existed, beside or under the King, to perform acts of speaking which the rex, for whatever reason, could not. But (with the possible exception of the federal office) all the attested Latin Dictators have in common that they are regular, annual, and—presumably, after whatever fashion—elected magistrates, which stands in sharp contrast to the Roman variant.79 The evidence, such as it is, thus points to the conclusion that dictator and magister populi initially denoted two distinct offices.
The Roman magister populi almost certainly existed already in the Regal Period, as a military commander to serve instead of the King whenever the latter was indisposed, incapable, or for some other reason prevented from taking the field, or conceivably even as the regular actual commander in the presence of the King.80 Such a position would have been held either permanently, or by appointment for each specific campaign, but hardly on an annual basis. If a similar institution existed in other Latin communities, it is likely that it went by the same title, and improbable that it provided the basis for the annual magistrate called dictator that eventually replaced the King. The title magister populi is descriptive of the institution’s principal function; presumably, so is the Latin dictator—the one who speaks and, in so doing, shapes reality.
At Rome, however, institutions underwent a development different from that in many Latin towns. The King was replaced not by a single dictator but by two annual magistrates, of equal or unequal collegiality, most likely known until at least the mid-fifth century as iudices or praetores.81 The magister populi survived, though not as the regular military commander (a function assumed, it would seem, by the new chief magistrates): he was called upon to assume command—not merely of the army but the entire State—in a time of crisis. The office thus acquired a function significantly different from the one it had held originally, a function that now included—unlimited—civil authority and the expectation to shape events in favor of the Roman People. The ability—previously attached to the official known as dictator—to “speak with effect” would be integral to its new purpose.82 To make it so, it would be essential that he himself be “spoken,” dictus, and that this be done under conditions ensuring the most stringent augural correctness, de nocte silentio, by night in ritual silence (Livy 8.23.15; 9.38.14); this manner of appointment presumably belonged to the magister populi already in regal times, and must continue to be observed unalterably, lest his efficacy be put in jeopardy. As the iudex could be called praetor, especially in a military context,83 so the magister populi could now be called dictator, without either losing his original appellation in ritual parlance—the iudices remained thus in the commentarii consulares, and the magister populi in the libri augurales.84 Nor does it necessarily follow that the term dictator overshadowed magister populi from its first republican manifestation; it may have entered common parlance as late as the fourth century, when Dictators came to be named with increasing frequency for special, non-military purposes: first clavi figendi causa, in 363 V and again in 331, on numerous occasions to hold elections (352, 351, 350, 349, 335, 327, 321, 306), and once each feriarum constituendarum causa (344) and ludorum faciendorum causa (32285).
One may object that our sources show no awareness of this “speaking-with-effect” aspect of the office. Not consciously, it is true; but neither do they exhibit much conscious awareness of most archaic concepts underlying Roman cult and government. When Varro wrote how the Praetor’s saying the magic words do dico addico on a dies nefastus gave legal force to the matter despite the ritual flaw attached, he surely did not think of it in terms of the spoken word creating reality: he knew the way things worked, not necessarily the reason why. Livy’s consistent picture of the Dictator as an office imbued with a degree of terror not quite of this world is credible precisely because it remains unexplained (and not consciously understood) by him: antiquarian speculation or invention would have told us everything.
According to Brennan, the Dictator’s imperium, rather than being of the same nature as the Consul’s, albeit maius,86 was not technically maius but of an altogether different kind that caused the Consuls (and all other magistrates) to lose their powers in the presence of the Dictator; in that sense, it could be termed valentius. Wherever the Dictator went, he was the sole magistrate penes quem fasces, “and so always had the capacity for independent action”; in consequence, the Consuls’ imperium, in his presence, “lay dormant,” with both of them reduced to the same state simultaneously as each experienced, under normal circumstances, every other month when his colleague held the fasces.87 In principle, this is surely correct. Yet nothing in Claudius’ comment compels the notion that the Dictator’s imperium as such was defined differently from the Consul’s: imperium, even in the Late Republic, simply was imperium; qualifiers such as maius, minus, consulare, or praetorium did not describe distinct types or kinds of imperium, but merely served to regulate precedence among magistrates of different or equal rank. Surely valentius falls into the same category; we ought to proceed under the assumption that it describes effect rather than nature. The Dictator’s imperium was more powerful than the Consul’s not because it belonged to a different kind; indeed, no source suggests that the Romans had any concept of such a difference, let alone a word for it.88 (Even the term valentius occurs just once, in Claudius’ address.) What rendered the Dictator’s imperium more “effective” was something else, abundantly attested in the sources: the peculiar nature of the dictatorship.89
Alone among magistrates, the Dictator could be expected to cause a turn of events in favor of the Roman People merely by being dictator; alone among magistrates, he not only lacked a colleague but effected the cessation of all other offices (save the Tribunes of the Plebs, of course) as independent agents. Ordering the Consul to appear before the Dictator without lictors and insignia did not cause this cessation: it simply gave visible expression to it. As we have seen, an incumbent magistrate who appeared in public without his lictors or insignia remained a magistrate; but the gesture signalled that he was not going to exercise his powers for the time being, and that his actions while in such a state of “undress” were to be seen as those of a privatus, not pro magistratu. The Consul could forbid a Praetor to exercise the powers of his office (vetari quicquam agere pro magistratu) and thus effectively suspend him, or deprive him of his insignia by having his curule chair smashed, or toga praetexta ripped: but all attested instances of this kind represent disciplinary measures against Praetors who had treated the Consul with disrespect, or otherwise defied higher authority.90 Yet we have no evidence that the Consul could order the Praetor to approach him without lictors, let alone did so as a matter of routine ceremonial, and when Fabius ordered Servilius to appear before him in such fashion in 217, it is out of the question that he intended it to signal punishment or humiliation for some infraction.91 Hence the relation of the Consul to the Dictator was different in character from that of the Praetor to the Consul.
These observations confirm—paradoxically, at first glance—Brennan’s view that the Dictator did not possess an imperium maius vis-à-vis the Consul in the same sense as the latter’s imperium was maius compared to that of the Praetor: otherwise, there would be no need to make the Consuls dismiss their lictors. Even with their fasces, they would not be able to oppose the Dictator any more than the Praetors were able to oppose the Consul. Brennan likened the relationship between Consul and Dictator to that between the Consul “not in charge” and his colleague penes quem fasces, while emphasizing—correctly—that the Dictator was not a colleague, maior or otherwise, of the Consul: hence the latter was unable to obstruct him.92 Yet the Consul holding the fasces could not order his colleague to appear before him without his lictors (or other insignia): the Dictator could. Nor could the Consul “in charge” remove the other Consul from command of an army, and assign him a different military task instead: the Dictator could, and did.93 Not only was the relation between Consul and Dictator different from that between Praetor and Consul; it also differed from that between the Consul holding the fasces and his colleague. From all we have seen concerning the presence of lictors, it follows that when facing the Dictator without lictors, the Consuls stood on the same footing, essentially, with regard to him as a private citizen stood with regard to the Consul. (Plutarch’s καὶ τὰ παράσημα τῆς ἀρχῆς ἀποθέμενον ἰδιώτην ἀπαντᾶν, Fab. 4.3, is not far off the mark.) A Consul could not issue a direct order to the Praetor: the latter, although subordinate to the Consul, held his powers, including his auspices, directly from Iuppiter and the People, and was answerable to them independently for all his actions. In consequence, the Consul could make the Praetor do something against his will only by forbidding him, vi maioris imperii, to do the thing he wanted to.94 But against a Dictator, the Consuls were not in a position to freely exercise the powers of their office, and consequently could not be held responsible for carrying out his orders: he even had no need to couch those in the language of prohibition.
As Brennan notes, Servilius in 217 did not lose his lictors the moment Fabius assumed office as Dictator: that happened only later, when they met in person, and only at the Dictator’s express orders.95 It is the Dictator’s actual presence, not merely his existence, that matters: wherever he goes, all other magistrates’ imperia are in abeyance. In this context, it is unfortunate that Messalla’s treatment of magistrates’ auspicia, as preserved in Gellius (13.15.4), omits the Dictator. The reason is probably not, as Brennan entertained, that Dictator and Magister Equitum were not considered “proper magistrates” in augural parlance; it is difficult to see how an official called magister (be it populi or equitum) could not be magistratus—an obvious abstraction of the former.96 Gellius quoted the passage to explain the distinction between magistratus maiores and minores, but this is incidental to Messalla’s disquisition, which concerns itself with the auspices “of the Patricians”: patriciorum auspicia in duas sunt divisa potestates; maxima sunt consulum, praetorum, censorum…reliquorum magistratuum minora sunt auspicia. Presumably, Messalla—if he thought it necessary to discuss them at all—treated the auspices of the Dictator in a passage nearby,97 as well as those of another official who undoubtedly possessed the patriciorum auspicia maxima—the Interrex. Now Messalla explains at length how the auspices of Consuls and Praetors interact with each other (all of them being colleagues), prevailing or overriding as the case may be, but not with those of the Censors, the latter not being colleagues of the former (neque consules aut praetores censoribus neque censores consulibus aut praetoribus turbant aut retinent auspicia). Hence, within the group of auspicia maxima, the relative strength of auspices affects only those who are colleagues within the same office: the Consuls’ take precedence over those of the Praetors, but have no effect on those of the Censors, and vice versa (at censores inter se, rursus praetores consulesque inter se et vitiant et obtinent). The separate nature of the Censors’ auspices, though maxima, is further indicated by the fact that, unlike all other patrician magistrates, Censors upon election are confirmed by a lex centuriata instead of a lex curiata (Cic. LegAgr 2.26). Like Consul and Praetor, but unlike the Censors, the Dictator requires a lex curiata (as attested explicitly by Livy 9.38.15); he is no one’s colleague, but named directly by the Consul with the Consul’s own auspices. It stands to reason that the Dictator’s auspices cannot fall into a category separate from those of the Consuls, as opposed to the Censors’ auspices.98 In other words, the auspices of the Dictator are of the same nature as the auspicia consulum.99 But, in his presence, there are no Consuls able to exercise their powers, hence no auspicia consulum other than his own—and no possibility of collision.
Yet Consuls, and Praetors, evidently did not cease to function altogether. The crucial qualifier was, as Brennan saw, the presence of the Dictator on the scene, be it in the City or militiae. The qualifier is not without parallel: one may compare the Tribune’s sacrosanctity—and, consequently, veto power—affecting everyone and everything in his presence, but extending no further. In 43 bc, Cicero proposed, unsuccessfully, that C. Cassius Longinus, pro consule in Syria, be granted a maius imperium with regard to the commanders in the Eastern provinces, whenever Cassius entered any of these provinces (Phil. 11.30: utique, quamcumque in provinciam eius belli gerendi causa advenerit, ibi maius imperium C. Cassi pro consule sit quam eius erit qui eam provinciam tum obtinebit cum C. Cassius pro consule in eam provinciam venerit): his “personal presence” was made “a condition of the operation of his imperium maius.”100 Away from the Dictator, Consuls and other magistrates continued to tend to the duties of their office, including those that entailed military command. There are no grounds for thinking that they did so without the usual trappings, lictors and all;101 if Servilius the Consul was entitled to retain his lictors until coming face to face with the Dictator, he ought to be entitled to resume them after departing. We know for certain that the Consul, in the Dictator’s absence, retained his auspices: in 216, Varro returned to Rome in order to name a second Dictator while the other was with the army (Livy 23.22.10–11); which nomination could only happen auspicato. In theory, thus, a Consul might attempt to act contrary to the Dictator’s orders once the two had parted; in reality, the chance of such was nil. The Dictator could, at any moment, re-establish his presence: at which point, the Consul would be rendered effectively privatus, and subject to be dealt with like any other recalcitrant citizen.
As with the auspices, so with imperium. The Dictator’s was technically no different from the Consul’s; what rendered it valentius was the appointee’s near-magical ability to shape events in favor of the Roman People—an ability that must not be jeopardized by allowing any other powers to exist beside him (let alone any power of the same kind). Hence the incapacity of the Consuls to exercise their imperium in his presence. From which it does not follow, though, that the Dictator had no more power than the Consul. This unique aspect of his office made the Dictator eo ipso superior not merely to one but both Consuls at the same time, and allowed him to issue direct orders to either or both of them, just as to any private citizen. In practical terms, it gave the Dictator exclusive control over the res publica; insofar, Livy naturally and justifiably describes his imperium as maius or summum.102 Or as the Greeks would put it, with a Dictator in office, all other magistrates, save Tribunes, lose their powers. The Consuls indeed are not “lower” magistrates compared to the Dictator: they simply cease to function as independent magistrates during a dictatorship. Which is not to be confused with ceasing to function as magistrates at all: they (as do the other magistrates) clearly carry on with their regular tasks and duties, including military operations—but all under the direction and at the discretion of the Dictator. The apparent contradiction between the Consul’s naming the Dictator and the fundamental rule that no magistrate can create another that holds an imperium greater than his own103 is thus resolved. The Consul does not appoint a magistrate with maius imperium; he names one in whose presence no other magistrate can freely exercise his powers, and whose imperium, by virtue of that exclusive character of his office, remains the only one in force whenever other magistrates are present. That the ancients should find it to be valentius, stronger and more effective than the Consuls’, need hardly surprise.104
What was it that, practically speaking, put an end to a Dictator’s tenure? The office had no fixed term like the one-year duration of the permanent magistracies; in principle, its tenure was indefinite, until the completion of the Dictator’s task, up to a maximum of six months. Yet even on reaching that six-month limit the office did not terminate automatically—unlike the annual magistracies, which expired with the last day of their term. For most Dictators, Livy notes their abdication: this cannot be an empty gesture, but must reflect a ritual necessity. Every Dictator (whether “special” or rei gerundae causa) was expected to abdicate as soon as the situation that had prompted his appointment was resolved; but his office ended only with that abdication.105 Moreover, it appears that, until the late third century, the six-month limit applied only to Dictators named rei gerundae causa.
In 216 bc, after M. Iunius Pera (cos. 230; cens. 225) had assumed office as Dictator rei gerundae causa, M. Fabius Buteo (cos. 245; cens. 241) was named Dictator senatus legendi causa—for six months, but without a Magister Equitum. In a contio immediately upon assuming office, Fabius found much—indeed, everything—wrong with his own appointment: two Dictators at the same time (which had never happened before); no Master of the Horse; censorial authority allowed to a single man, and to the same man twice; and imperium given to a Dictator for six months, for reasons other than rei gerundae.106 That last objection implies both that for “special” Dictators the six-month appointment was an innovation, and that for Dictators rei gerundae causa it was well established, or at least had precedent. But how exactly was the limit imposed?
The Dictator received his powers through the dictio: being named (dictus), with auspices ave sinistra, by the Consul. As Mommsen saw long ago, the auspical formula pronounced by the Consul when asking Iuppiter for permission to name a Dictator contained the limits of the appointment: Iuppiter pater, si est fas me hic hodie magistrum populi dicere in sex menses, or something to that effect. All that was necessary to create a Dictator not subject to that limit was to omit the phrase in sex menses at the auspication for the dictio.107 But the terms of the question put to Iuppiter were a matter of the ius augurale, and could not as such determine the length of office. Their force lay elsewhere: the Dictator held his auspices for as long as they had been requested, by the Consul naming him, from Iuppiter. When that time expired, so did his auspices. He remained Dictator, though—but lacking valid auspices, vitium now attached to anything he undertook; hence he must abdicate and thus terminate his office. If no particular duration was specified in the dictio, he retained his auspices until he abdicated.
Mommsen further thought that, for Dictators named to complete a special task, the mention of that reason alone—say, clavi figendi causa—in the auspical formula constituted the limit of their term originally. Reasonable and attractive as that conclusion appears at first glance, it runs into obstacles. For one, while the case of Fabius Buteo shows that a temporal clause such as in sex menses could be added in the dictio (and usually was added for those named rei gerundae causa),we do not know for certain if the reason for the Dictator’s appointment was specified in the dictio. (That of Fabius Buteo, as reported by Livy 23.22.11, did not contain the words senatus legendi causa.) The fact that Sulla in 82 deemed it necessary to have his extraordinary commission—legibus scribundis et rei publicae constituendae—spelled out by comitial legislation108 rather suggests that its mere inclusion in the dictio would not have been sufficient to expand the powers of the office beyond those normally associated with it. Insofar as the dictio established the Dictator’s auspices—and, if in sex menses, also limited their duration—inclusion of a particular task would have restricted, presumably, their validity to actions related to that task, rather than create a requirement to abdicate upon completion of the task. Yet whether such a restriction of auspices to a particular task was augurally possible is doubtful; no examples appear to be on record. Magistrates had auspices, plain and simple; any constraints in their use were imposed by the powers of the office held: a magistrate could not auspicate for a task that he was legally not capable of performing.109 The Censors, for instance, held auspicia maxima, just like Consuls and Praetors, and, like the latter, they held them by virtue of their office (Messalla apud Gell. 13.15.4). They were not auspices custom-tailored to the needs and limits of that office; it was the office that limited their applicable use to its functions.
It seems certain that any Dictator, regardless of designation, possessed the undiminished imperium and unrestricted auspices of the office, and could attend to any business that came his way.110 The only limitation imposed on a special-purpose Dictator (sometimes termed by moderns, improperly, imminuto iure) was one of duration, the completion of the task for which he was named, upon which he was supposed to abdicate. The formulation of this reason—normally by way of a senatus consultum—did, of course, serve to prescribe the moment when his abdication would be expected, but it appears that this limit was not enshrined in comitial legislation. No doubt it was considered fas and ius, but for enforcement one must rely on each Dictator’s willingness to obey the rule—and on public pressure, if necessary.111 When L. Manlius Imperiosus, named clavi figendi causa in 363 V, proceeded to call up the levy for an impending war against the Hernici after completing the nailing ceremony, public opposition forced him to step down (perinde ac rei gerendae ac non solvendae religionis gratia creatus esset, bellum Hernicum adfectans dilectu acerbo iuventum agitavit; tandemque omnibus in eum tribunis plebis coortis seu vi seu verecundia victus dictatura abiit, Livy 7.3.9): not because as Dictator clavi figendi causa he could not call up the army, but because, having dealt with the situation that had occasioned the reason for his appointment, he was not supposed to remain in office. In Livy’s account, there is no hint at a violation of ius, be it publicum or augurale. But Cicero’s comment on the matter (Off. 3.112, L. Manlio A. f., cum dictator fuisset, M. Pomponius tr. pl. diem dixit, quod is paucos sibi dies ad dictaturam gerendam addidisset) shows that the Dictator’s failure to abdicate when expected could result in prosecution, once he was privatus again. At the same time, Cicero’s characterization of the offense—Manlius had added a few days to exercising his dictatorship—is instructive: the office itself is not described as having already expired (for instance, quod is dictatura functus per paucos dies magistratum continuisset).112 One doubts that Manlius would have prepared for a military campaign if his auspices were understood to have lapsed the moment he drove the nail into the temple wall.
In fact, the manner of Buteo’s appointment rather suggests that no reason for it was mentioned in the dictio. Evidently Varro the Consul had included the phrase in sex menses when he named him. Now, Livy rather makes it sound as if that limit, along with the peculiar lack of a Magister Equitum, was part of the decree ordering Buteo’s appointment (M. Fabium Buteonem ex senatus consulto sine magistro equitum dictatorem in sex menses dixit). Naming a Dictator without permitting him a Master of the Horse (more on which shortly) marked this dictatorship out as irregular—most likely in an attempt to mitigate the anomaly of having two Dictators in office simultaneously.113 (The year before, when Minucius the Magister Equitum had his imperium made equal to the Dictator’s, neither he nor Fabius Maximus appointed another Master of the Horse.) Conceivably, the measure could have been designed to prevent Buteo from attending to military matters, as Manlius Imperiosus had attempted a century and a half before. But why such an extreme measure should have been deemed necessary in this case is hard to imagine: no special-purpose Dictator after old Manlius had sought to exceed his commission. Moreover, the Senate decree, although it did not specify Fabius Buteo by name, had left no room for doubt as to who would be appointed: the Dictator was to be a former Censor, and the most senior among the ones still living (Livy 23.22.10, dictatorem, qui censor antea fuisset vetustissimusque ex iis qui viverent censoriis esset, creari placuit qui senatum legeret). Hence Buteo, like every senator, had known what was coming, and we may safely assume that he voiced his objections to the procedure not only in the contio upon taking office but also previously in the Senate meeting in which the decision was made. That anyone should suspect him of aiming to abuse his position passes belief. The same consideration applies to his six months. No one, surely, worried that Buteo might draw out the lectio senatus for more than a few days, and thus had better be constrained in the same manner as a Dictator rei gerundae causa.
Indeed, Fabius’ criticism clearly implies that he saw this less as a limitation on his office than as an undue extension: the Dictator rei gerundae causa was given up to six months because the situation he was to deal with might require that much time; for a special commission, this had hitherto been unthinkable. It follows that, in consequence of his having been named in sex menses, Buteo could remain in office for up to six months, without violating fas or ius, or even mos (otherwise, why grouse about it?), even if he accomplished the task in a matter of days. His appointment in sex menses thus runs counter to any limitation that may have been intended in not allowing him to name a Magister Equitum.
To turn to that other anomaly in Buteo’s appointment: how could the Consul name a Dictator sine magistro equitum? It would have to be included in the request of auspices for the dictio (Iuppiter pater, si est fas hic hodie me magistrum populi dicere in sex menses, ita uti is sine magistro equitum sit vel ita uti is magistrum equitum dicat neminem), there being no other conceivable way of preventing the Dictator from carrying out what ordinarily would be his first official act. A further thought arises. Did the dictio, perhaps, at all times contain a reference to the naming of a Master of the Horse: si est fas me magistrum populi dicere (in sex menses), ita uti is magistrum equitum dicat (in sex menses)? Such a ritual would readily lead to the loose parlance, in Livy and—perhaps—even the Augur Cicero, that has the nominator name both the Dictator and his deputy; and if indeed the Consul was able to name a Magister Equitum directly,114 that too would have to be contained within the auspical formula (si est fas me magistrum populi et magistrum equitum dicere), for otherwise, once the Dictator had been named, the Consul might lack the power to do so. If this be correct, the mere omission of ita uti is magistrum equitum dicat would suffice to keep the Dictator thus named from appointing a Master of the Horse. The auspical formula would also offer a ready procedure in those cases where comitial legislation prescribed the identity of both men to be named: the Consul would be instructed to ask Iuppiter, for example, si est fas me dicere magistrum populi L. Titium, ita uti L. Titius magister populi dicat C. Seium magistrum equitum.
Buteo’s is the fourth of five occasions known in which a Dictator served without a Master of the Horse. In the first such case, in 249, the unhappy M. Claudius Glicia, qui scriba fuerat, was forced to abdicate before he had a chance to name his deputy.115 In the last, in 49, Caesar did not name a Master of the Horse (evident from the Fasti, but not recorded in any literary source). The reason is unknown. He had gone to the trouble of procuring comitial legislation that empowered the Praetor to name a Dictator, and, on the precedents set in 217 (?), 210, and 82, we might expect the law to prescribe the names both of the Dictator and the Master of the Horse—unless in this instance it directed the Praetor to name Caesar Dictator sine magistro equitum. In any case, Caesar was at Massilia when he learned of his appointment as Dictator (Caes. BCiv 2.21.5). Being away from Italy, in Gaul, he could not name a Magister Equitum there without violating augural law. He was Dictator from the moment Lepidus named him; when Dio notes that he entered office after he arrived in Rome (41.36.2), it can only mean that he waited to obtain his initial auspices until then. He clearly intended to complete his business as quickly as possible, and abdicated after eleven days (Caes. BCiv 3.2.1); he may simply not have bothered with a Master of the Horse—the absence of which, in turn, might serve as a concession to those who held that this dictatorship, like Fabius Buteo’s in 216, was irregular. With the Augural College presumably on record as opposing nomination by the Praetor, it may have seemed prudent—the Civil War was far from decided, and support or, at least, neutrality could depend on small gestures.116
As Varro had named Iunius Pera Dictator without anomalies earlier in the year, the appointment of Fabius Buteo without a Magister Equitum cannot be ascribed to inadvertence on the Consul’s part; he clearly acted as directed by the Senate. The six-month limit is more difficult to understand; no plausible explanation offers why the Senate should have wanted it thus. Did Varro include in sex menses by mistake, because he had used that formula earlier when naming Iunius Pera? (Once the words had been spoken, their effect could not be recalled.)
The remaining two instances of a Dictator serving without a Magister Equitum require a separate discussion. The appointment in 216 of Fabius Buteo as Dictator senatus legendi causa, to hold office side by side with the Dictator rei gerundae causa, Iunius Pera, marked a constitutional anomaly; yet an even more spectacular one (quod nunquam ante eam diem factum erat, Livy 22.8.6) had occurred in the year before: the grant in 217, by plebiscite, of imperium equal to that of the Dictator, Q. Fabius Maximus, to his Master of the Horse, M. Minucius Rufus.
Polybios leaves his readers in no doubt as to the significance of that step, calling attention to it twice in as many sentences:
τὸν δὲ Μάρκον ἐπὶ τοσοῦτον ηὖξον διὰ τὸ συμβεβηκὸς ὥστε τότε γενέσθαι τὸ μηδέποτε γεγονός· αὐτοκράτορα γὰρ κἀκεῖνον κατέστησαν…καὶ δὴ δύο δικτάτορες ἐγεγόνεισαν ἐπὶ τὰς αὐτὰς πράξεις, ὃ πρότερον οὐδέποτε συνεβεβήκει παρὰ Ῥωμαίοις (3.103.4).
Minucius, too, was made Dictator, in an unprecedented move; and now indeed there were two Dictators in office for the same task, something that had never happened before. Did Polybios know of two Dictators in office together on an earlier occasion, but for different tasks (such as one rei gerundae, the other clavi figendi causa)? Perhaps, but it seems unlikely:117 the emphasis on the same task was probably prompted by the next instance of two simultaneous Dictators, just a year later—one rei gerundae causa, the other senatus legendi. Certainly Plutarch made that connection: the Romans voted that Minucius should pursue the war as an equal in command, and with the same powers as the Dictator—something that had never occurred previously at Rome, but was to happen again soon afterwards, in the wake of Cannae, when M. Iunius was Dictator in charge of the army and in the City Fabius Buteo was chosen, as a second Dictator, to replenish the ranks of the Senate.
τὸν δὲ Μινούκιον ἐψηφίσαντο τῆς στρατηγίας ὁμότιμον ὄντα διέπειν τὸν πόλεμον ἀπὸ τῆς αὐτῆς ἐξουσίας τῷ δικτάτορι, πρᾶγμα μὴ πρότερον ἐν Ῥώμῃ γεγονός, ὀλίγῳ δ’ ὕστερον αὖθις γενόμενον μετὰ τὴν ἐν Κάνναις ἀτυχίαν. καὶ γὰρ τότ’ ἐπὶ τῶν στρατοπέδων Μᾶρκος Ἰούνιος ἦν δικτάτωρ, καὶ κατὰ πόλιν τὸ βουλευτικὸν ἀναπληρῶσαι δεῆσαν…ἕτερον εἵλοντο δικτάτορα Φάβιον Βουτεῶνα (Fab. 9.3–4).
That Polybios knew of Minucius as Dictator plain and simple, not just as a Magister Equitum with aequatum imperium, is clear enough (cf. also 3.106.2, οἱ μὲν δικτάτορες ἀπέθεντο τὴν ἀρχήν, “the Dictators laid down their office”). But Plutarch’s narrative in all other elements seems to follow Livy step by step; even the process of Minucius’ “promotion” is couched in terms (ἐψηφίσαντο…δικτάτορι) that parallel Livy’s rogatio de aequando magistri equitum et dictatoris iure (22.25.10), and the end of the affair is quite the same, with Minucius referring to Fabius as “the Dictator” (ὑμῖν δὲ τῶν μὲν ἄλλων ἐστὶν ἄρχων ὁ δικτάτωρ, Fab. 13.4)—by implication the only one—and returning under his command (Fab. 13).118 Only in the digression on the two Dictators of 216 does the biographer think of Minucius as Dictator; and while he could have found the information about Fabius Buteo in Livy,119 he would also have found there the claim that this, not 217, was the first such case of a dual dictatorship (neque duos dictatores tempore uno, quod nunquam antea factum est, probare se dixit, 23.23.2). Hence the comparison with the events of 217 is his own.
Or is it? Nothing in Livy could have given Plutarch the idea that Minucius was no longer Magister Equitum but Dictator. Another source,120 surely, alerted him to that distinction, and to the repetition of the anomaly in the next year. Polybios remains a possibility, either in his lost discussion of Roman institutions in Book 6, or in his narrative of events after Cannae in Book 7. (That he commented on Iunius Pera and Fabius Buteo is likely enough.) Indeed, Plutarch returns to the main story with the words τὸν δὲ Μινούκιον ἐπὶ τὰς αὐτὰς τῷ δικτάτορι πράξεις ἀποδείξαντες, “having appointed Minucius for the same purpose as the Dictator” (Fab. 10.1): an obvious echo of Polybios. But one rather suspects a source common to the biographer and Polybios—hence belonging to the earliest stratum of the tradition—that included the comparison with 216 at this point in its narrative, prompting Plutarch to adopt it here as well, and perhaps expand it with the brief story of Buteo’s one-day dictatorship (Fab. 9.5): a story carefully omitted, along with that of Iunius Pera, from his long account of Cannae’s aftermath (Fab. 17–18).
Livy does not portray Minucius as Dictator, calling him magister equitum (and Fabius dictator) throughout the period of his independent command.121 The magnitude of the innovation is noted, though only in Minucius’ voice (quod nulla memoria habeat annalium, 22.27.3), not the historian’s own. Yet, in an unguarded moment, he lets the Polybian view slip into his narrative: when the Consuls resume command toward the end of the year, Atilius takes over Fabius’ army, Servilius that of Minucius (22.32.1, consules Atilius Fabiano, Geminus Servilius Minuciano exercitu accepto). After all the talk about Minucius with his army returning under the Dictator’s orders (sub imperium auspiciumque tuum redeo et signa haec legionesque restituo…tu, quaeso, placatus me magisterium equitum, hos ordines suos quemque tenere iubeas, 22.30.4), it transpires that Minucius retained command of his own army until the end of his term.122
The other sources essentially adopt the Livian view—an upgrade of imperium without the title of Dictator, and resignation or return to his subordinate position after his rescue at the hands of Fabius.123 Zonaras’ account is of particular interest (8.26, οὐδ’ ἀνέμεινε τὸν δῆμον ἀναψηφίσασθαι, ἀλλ’ ἐθελοντὴς τὴν ἡγεμονίαν…ἀφῆκε): in Dio, it appears, the story took Minucius’ humiliation a step further, by reporting a threat that the People might revoke his imperium after his luck against Hannibal ran out, or even (as Vervaet argues) an actual plebiscite repealing the Metilian Law subsequent to Minucius’ self-subordination.124
Livy’s version has found champions among moderns. It will not do. The famous inscription of Minucius’ altar agrees with Polybios (and Plutarch): Hercolei sacrum M. Minuci C. f. dictator vovit (ILS 11).125 Nor do we have reason to think that Polybios, in a matter whose constitutional significance he emphasized not once, but twice (3.103.3–4), carelessly glossed over or mistook the true nature of Minucius’ enhanced position. The oldest tradition treated Minucius as Dictator. Was he dictus?
The lex Metilia gave the Magister Equitum imperium equal to that of the Dictator; that much is not in doubt. To permit him the title dictator as well, the legislation could have instructed the Consul M. Atilius Regulus, suffected just days before the passage of the plebiscite (Livy 22.25.16), to name Minucius Dictator. The insistence of Metilius the Tribune that Fabius see to the election of a Consul suffect before the vote on his bill (Livy 22.25.11) may point to that solution. But the Dictator, if uncooperative in the diminution of his power, could simply forbid the Consul to perform such a dictio: hence Metilius’ threat that he was prepared to have Fabius’ imperium abrogated altogether (Livy 22.25.10). Now Polybios and Livy agree on a pertinent point: Minucius’ term in office ended on the same day as did that of Fabius.126 Had he been dictator dictus a M. Atilio consule, his six-month term would have continued for several months beyond that of Fabius.
We have a further sign that the lex Metilia did not simply create a second Dictator in the normal sense of the office: nothing indicates that Minucius appointed a Magister Equitum of his own, or that Fabius replaced him with another deputy. When Fabius Buteo was named as a second Dictator the next year, it was with the express proviso sine magistro equitum (to which he objected, Livy 23.22.11–23.2): irregular, and presumably intended to mark this dictatorship as such.127 Did the plebiscite contain the same proviso, and extend it to Fabius as well: two Dictators rei gerundae causa, commanding at the same time, but each without a Master of the Horse?
It appears certain that Minucius’ dictatorship was not one established de novo—as would have been the inevitable effect of a dictio—but an extension of his magisterium equitum. If Minucius was dictus by the Dictator at his initial appointment, it may not have been augurally necessary to repeat the dictio—not even proper, in all likelihood, without his abdicating first. (If Fabius himself was not, indeed, a dictator dictus,128 the question would be moot in any case.) Hence the tradition of an aequatio imperii, without dictio by the Consul, is probably correct insofar as it describes the process that gave Minucius his enhanced position. Whether in consequence he could call himself “Dictator” would depend on the exact language used in the law.
At a minimum, we should expect something like velitis iubeatis ut M. Minucius magister equitum aequo (sive pari) sit imperio (sive iure) ac Q. Fabius dictator. If it continued, for instance, ita uti rem uterque gerat sine magistro equitum, Minucius (and contemporaries) could reasonably conclude that the prohibition to name a Master of the Horse made him, by implication, dictator, there being no other official with the power to make such an appointment. Others, of course, could conclude that the identical prohibition addressed to Fabius, along with the absence of the word dictator in direct reference to Minucius, implied that the latter was still Magister Equitum. In the light of Minucius’ inscription, Polybios, and Plutarch, there can be little doubt that the first interpretation prevailed at the time.129 Nor can we be certain that the law did not apply the word dictator to Minucius: velitis iubeatis ut dictator sit M. Minucius magister equitum aequo (sive pari) imperio (sive iure) ac Q. Fabius dictator, or ita uti dictator rem uterque gerat sine magistro equitum. With the passing of time, Minucius’ lasting humiliation in virtually the entire surviving tradition, and Fabius’ ever-increasing stature as an exemplary leader, would ensure ready acceptance of a version, once put forth, that denied him the title of Dictator altogether.
Whether the six-month limit had existed from primordial days we cannot say for certain; in the first century bc, Livy and perhaps most Romans thought so, despite traces of an earlier tradition that some Dictators held office for considerably longer than six months.130 The first such case is Camillus in 396 V: attested as Dictator on October 31 at the instauration of the Latin Festival, he proceeded to capture Veii, and after his triumph dedicated the temple of Mater Matuta, almost certainly on June 11. Those dates amount to a tenure of about eight months.131 The most explicit instance involves again Camillus, in 390 V: eaque causa fuit non abdicandae post triumphum dictaturae, senatu obsecrante ne rem publicam in incerto relinqueret statu (Livy 5.49.9); neque eum abdicare se dictatura nisi anno circumacto passi sunt (Livy 6.1.4). Mommsen, following Weissenborn and Müller, points out that anno circumacto, far from implying a year-long dictatorship, means that Camillus remained in office until the end of the consular year.132 But the consular year had started on July 1 (Livy 5.32.1), and the Gauls sacked Rome in the last third of that month. Livy does not give a date for Camillus’ appointment, but clearly envisages it at a point in time well before the Gauls’ departure seven months later.133 Whoever devised that version of the story was thinking of a dictatorship that lasted longer than six months, though not an entire year; and it may be significant that Livy makes no mention of a six-month limit here.134
Sulla has been thought to have stood model for Camillus’ dictatorship in 390 V, on grounds that “the whole notion of a dictator re-establishing the constitution is unparalleled from the fifth and fourth centuries.”135 Certainly; but Livy says—and implies—nothing about the Dictator’s re-establishing the constitution, and Camillus’ reputation as a Second Founder of Rome “was already current” by the late second century.136 It goes without saying (or so it should) that the silence of Polybios has no bearing on the issue. He evidently knew of—or believed in—no Roman victory over the Gauls (2.18.2–3, 22.4–5); but he was not concerned, in this or any part of his narrative, with constitutional details such as whether or not the Romans had named a Dictator on that occasion, or whether Camillus had any part in it. The silence of Diodoros might seem to carry more significance, since he does report the (third) dictatorship the other sources give to Camillus for the following year (14.117): but his account combines the defeats of Volsci, Aequi, and Sutrium with a victory over the retreating Gauls, complete with Camillus’ recovery of the Roman ransom; nor does he indicate that those events happened in the consular year after the sack. In other words, one cannot assert that Diodoros did not know of a dictatorship at all in connection with the Gallic catastrophe.
In 316 V (L. Aemilius Mamercinus, 9.21.1, 22.1) and 315 V (Q. Fabius Rullianus, 9.22.1, 24.1), Livy’s narrative clearly—though, perhaps, unawares—implies that those Dictators held office for approximately a year. To dismiss the inconvenient datum as “ein staatsrechtliches Unding”137 assumes as proven fact—without the slightest contemporary, that is, fifth- and fourth-century evidence—the very thing being called into question by that earlier source we glimpse through Livy: the notion that the six-month limit had always applied. To explain the discrepancy as “concocted by late historians ignorant about the nature of the long-disused dictatorship”138 takes no account of the fact that all but one of the “late historians” that have survived to comment on the matter—Dionysios, Livy, Plutarch, Appian, Dio, even Lydus—know the limit as original to the office.139 (The exception being, perhaps, Polybios, who in describing how the Dictator differed from the Consul says nothing about a six-month term: 3.87.7–9.) Are we to believe that the annalists of the second century, much closer in time to the real thing, were ignorant of how it worked, but those of the first somehow discovered the truth?
Lastly, the dictator years—however fictitious—indicate that in the eyes of some antiquarians a dictatorship exceeding six months was seen as at least a theoretical possibility for the more distant past.140 All in all, it is very little to go on. Perhaps the fact that the six-month limit was not applied to special-purpose Dictators until 216 bc provides a clue. The first such Dictators appear in the fourth century (beginning, in fact, with L. Manlius Imperiosus, clavi figendi causa in 363).141 It seems odd that the phrase in sex menses should be omitted in their dictio if it already constituted a fixed element in the naming of every Dictator; after all, the absence of such a limit would not increase the pressure to abdicate once the task was completed—if anything, it would invite temptation to prolong a man’s time in the office. By the late third century, the six months seem to have become established for the Dictator rei gerundae causa, perhaps even prescribed by legislation—which, however, would not alter the fact that only abdication could terminate his tenure.142
The Challenge to the Auspices: Studies on Magisterial Power in the Middle Roman Republic. C. F. Konrad, Oxford University Press. © C. F. Konrad 2022. DOI: 10.1093/oso/9780192855527.003.0003