Hostname: page-component-74656f9884-npj9w Total loading time: 0 Render date: 2026-11-07T13:54:26.434Z Has data issue: false hasContentIssue false
  • English
  • Français

TIMEKEEPING IN THE ROMAN ARMY*

Published online by Cambridge University Press:  09 October 2017

George Cupcea
George Cupcea*
Affiliation:
Babeș-Bolyai University, Cluj-Napoca
*
george.cupcea@gmail.com
Rights & Permissions [Opens in a new window]

Extract

The structure and organization of the Roman army is a complex subject for ancient historians. Of its multiple aspects, the schedule of the daily routine is one of the most interesting but, at the same time, is scarcely known. Of course, huge progress has been made with the publication of the daily rosters of one particular auxiliary unit in the East (cohors XX Palmyrenorum, at Dura, Syria), but the detail of the chronological organization of the unit's schedule is still to be revealed.

Information

Type
Research Article
Information
The Classical Quarterly , Volume 67 , Issue 2 , December 2017 , pp. 597 - 606
Check for updates
Copyright
Copyright © The Classical Association 2017 

1. INTRODUCTION

The structure and organization of the Roman army is a complex subject for ancient historians. Of its multiple aspects, the schedule of the daily routine is one of the most interesting but, at the same time, is scarcely known. Of course, huge progress has been made with the publication of the daily rosters of one particular auxiliary unit in the East (cohors XX Palmyrenorum, at Dura, Syria),Footnote 1 but the detail of the chronological organization of the unit's schedule is still to be revealed.

A rather simple inscription from Apulum offers important evidence for professional military ranks in the Roman army in Dacia and elsewhere, and also relates to the present investigation. CIL 3.1070 (= ILS 5625 = IDR 3/5.193) is a votive plaque presently lost; only some drawings remain.Footnote 2 The text is as follows:

I(oui) O(ptimo) M(aximo) et Iunoni Regin(ae)

pro sal(ute) Imp(eratoris) M(arci) Aur(eli) An

tonini Pii Aug(usti) et

Iuliae Aug(ustae) matris Aug(usti)

M(arcus) Vlp(ius) Mucianus mil(es) leg(ionis) XIII Gem(inae)

horologiar(ius?) templum a solo de suo ex uoto

fecit Falcone et Claro cons(ulibus)

To Jupiter, the Best and Greatest and to Juno the Queen, for the health of the Emperor, M. Aur. Antoninus Augustus, The Pious and of the Mother of the Emperor, Julia Augusta, M. Ulp. Mucianus, soldier of the legion XIII Gemina, ‘clockmaker?’, has erected the temple, alone, on his expense, following a vow, in the time of the consulship of Falco and Clarus.

We are dealing evidently with a votive monument erected by a legionary of XIII Gemina to Jupiter and Juno, for the health of Emperor Caracalla and his mother, mentioning the construction of a sanctuary involving a soldier with the role of clockmaker. However, the status of the character has not always generated agreementFootnote 3 with most scholars doubting the possibility of this soldier being a horologiarius.Footnote 4 This particular inscription crucially raises more general questions relating to the subject of timekeeping in the Roman army, an interesting subject for which there are, however, few sources.

2. TIMEKEEPING IN THE ROMAN WORLD

Time and timekeeping in Greek and Roman antiquity is a complex matter of debate, especially as ancient sources commenting on the subject in some cases engage with very precise technical details.Footnote 5

Throughout Roman history, even as late as the mid third century a.d., ancient writers complain about the scarcity of the sources on the subject of timekeeping. Three such writers approach the matter of the origins of timekeeping and horologia in the Roman world, all of them reaching the conclusion that the division of the day by hours was introduced in Rome in the third century b.c. and the adaptation of the first solar clock at the latitude of Rome took place sometime around 164 b.c. Footnote 6

The Romans used several separate pieces of equipment for measuring time, one of them related to the Sun and the other two to water. The sundial (horologium)Footnote 7 was used during sunlight hours, the klepsydra Footnote 8 was used for the measurement of fixed amounts of time, probably borrowed from Greek law courts,Footnote 9 while the water clock was used by night, in interiors or through cloudy days. The difference between the two water-operated mechanisms is appreciated since antiquity.Footnote 10

One might assume that the ancient clocks were at best approximate, and that exact timekeeping, at least in the civilian environment, was not a matter of too much concern.Footnote 11 In fact, detailed discussion on the matter of the daytime division into hours and the need to precisely evaluate these divisions points in a different direction. An inexact sundial from Sicily was used incorrectly in Rome for 99 years, without raising significant issues for the population.Footnote 12 On the other hand, the problem is noted repeatedly in Roman literature, and the establishment of a sundial adapted to Rome's latitude is acclaimed.Footnote 13 We can only assume that, by the third and second centuries b.c., the Romans were more likely to need general indications of the time of day rather than precise and detailed ones.Footnote 14

As far as epigraphy is concerned, clocks are mentioned especially in the form of horologia,Footnote 15 at a first glimpse in around one hundred instances, in both Greek and Latin, throughout the Roman world. Most of them are to be found in Italy and the western provinces, with the vast majority, more than 90%, occurring in civilian environments. Often the clock is part of a larger monument or public building complex,Footnote 16 the most famous being Andronicus' Tower, nowadays the Tower of the Winds, from Athens, dating back to the first century b.c. Footnote 17 These clocks were considered important artefacts, as they needed to be decoratedFootnote 18 and they seem to have been operated by slaves.Footnote 19

The making of such clocks is another matter, as Vitruvius (De arch. 9.7) argues that even if many books were available for the construction of portable clocks, their construction was accessible only to those that knew the analēmma—the skeletal celestial sphere, therefore confirming that there were no practical methods to construct sundials or water clocks available to the public.

Moreover, epigraphic evidence suggests that there were special workshops and craftsmen concerned with their fabrication. Inscriptions of Asia Minor record both what seems to qualify as a workshop for sundialsFootnote 20 and special craftsmen who are in charge with their construction.Footnote 21 Magistrates, especially consuls, usually pay for their construction and they are intended to be installed in political and juridical places in the city, therefore bearing, at least initially, an exclusive political role.Footnote 22

An overview of the matter indicates that sometime in the sixth or fifth centuries b.c. the first concerns for the scientific approach of the cosmos and the calendar occur, and by the fourth century also the first sundials, klepsydrae and hydraulic clocks, borrowed and adapted by the second century b.c. also in Rome.Footnote 23

3. TIMEKEEPING IN THE MILITARY

Literary Accounts

In the military, timekeeping must have been at least as important. The division of the day by hours, both in civilian and military environments, is known since the seventh century b.c.,Footnote 24 from Egypt, but from the first century b.c. onwards the sources come exclusively from the Roman world,Footnote 25 which may derive from the establishment of a professional soldiery and semi-permanent provincial garrisons.

Ancient authors admire the unity of action of the Roman military machine. Josephus (BJ 3.85) is impressed by the fact that the entire camp wakes, eats and goes to work at the same time. The march is also something to be chronologically defined; Vegetius (Mil. 1.9) observes that an infantry unit must march twenty miles in five summer hours.Footnote 26 Even if the schedule must have suffered alterations, a unity of direction and a well-established program must be presumed.Footnote 27

We know of few accounts in ancient sources relating to precise timekeeping in the army. Even so, Josephus (BJ 6.58, 68, 79, 131, 147, 157, 244, 248, 290, 294) notes that action was planned by the hours, when describing Titus' siege of Jerusalem.Footnote 28 Polybius argues that an ideal commander must be able to tell the time of the day, to ensure his success in campaign.Footnote 29 As no actual mention of a sundial or water clock appears in these instances, we might assume that astronomical reckoning is used. A classic example of the relativity of actual timekeeping is the battle of the Colline Gate in 82 b.c. at the end of Sulla's Civil War, where ancient sources give approximate and sometimes contradictory time relations.Footnote 30 On the other hand, Caesar in 54 b.c. acknowledges that night-time hours are shorter in Britain than on the Continent, measuring with a water-operated timepiece.Footnote 31

At the same time, the duty rosters of Dura,Footnote 32 Vindolanda or Egypt present tasks set out by day, not by hour, like in the modern armies.

The best-known instance of temporal organization inside the Roman army is the night watch (uigiliae), divided into four equal parts of three hours each, with the help of the klepsydra. Footnote 33 This is one of the most important tasks of the army, as is also evident from the strict system of the watchword, written on tesserae and distributed only by specialized principales, the tesserarii.

Archaeology

Archaeology also contributes to the issue, with a series of time markers being discovered in forts or locations related to them.Footnote 34 One of the most discussed such finds is a recently discovered bronze disk fragment from Vindolanda.Footnote 35 It was subsequently considered to be an anaphoric water clockFootnote 36 or an astrological calendar,Footnote 37 this not diminishing the significance of its discovery in an auxiliary fort on the northern frontier. It is a precious indication of the need for chronological unity in the empire's armies, and the usage of precise timekeeping in the military environment. Two other similar artefacts were discovered in Britain, at Housesteads and Richborough, both rudimentary but relatively usable. These being the only such discoveries coming from Roman Britain, one could argue that in this province the military brought the Roman clocks and timekeeping.Footnote 38

Epigraphy

A series of newly published ostraka from the Krokodilô fortlet in Egypt sheds some light also on the usage of daytime hours in the Roman army. These are daily registries for the arrivals and departures of soldiers and messengers in this military post station. This is a small-scale building, fit for ten to fifteen soldiers, led by a curator praesidii. For each day and each messenger, the time of arrival, the source and content of the package and the time of departure are registered. The sources present different hours in the day, telling us that timekeeping was an exact matter in this situation, and was probably the task of the curator, using a water clock.Footnote 39

In at least two cases we have a horologium mentioned in a military context, and in another four cases the soldier who can be assumed to be responsible for building and looking after the official clock—the horologiarius.

The first example qualifies as circumstantial evidence. It is a sundial discovered in Si‛â (Syria), with an inscription that mentions its owners/builders, two legionaries of VIII Augusta, one of them ranking as κούστως Σεία.Footnote 40

The second example is more closely related to the issue, CIL 13.7800 (= AE 1977.154), from Rigomagus in Lower Germany:

[--- Diadumeniano]

nobil[issimo Caesari]

sub C[lau(dio)] M[arcio Agrippa(?)]

leg(ato) Au[g(usti) p]r(o) pr(aetore) pr(ouinciae) agens Pe

tronius Athenodorus prae[f(ectus)]

coh(ortis) I Fl(auiae) horolegium [sic!] ab ho

ris intermissum et uetus

tate co(n)labsum [sic!] suis inpendi(i)s

restituit [[Imp(eratore) d(omino) n(ostro) Macrino Aug(usto) II co(n)s(ule)]]

‘… Diadumenianus, noble Caesar and Cl. Marcius Agrippa, governor of the province, with Petronius Athenodorus, prefect of the cohort I Flavia, acting as agent, have reconstructed, on their own expense, the (official) clock (of the fort), malfunctioning and collapsed owing to old age, in the year when our Lord, the Emperor Macrinus Augustus, was consul for the second time.’

This is an obvious case of a repair of an official fort clock, paid by the prefect of cohors I Flauia, after it deteriorated owing to its old age, dated a.d. 218. It is an official attestation of the presence and usage of such equipment in the military environment. The official character of the endeavour is evident from the dedication to the Caesar Diadumenianus and the patronage of the provincial governor, as well as its placement inside the principia.Footnote 41 On this evidence, the presence of an official sundial (Lageruhr) of the fortress has been presumed, which could have been placed on the front of the principia, facing the uia principalis. Footnote 42

4. THE HOROLOGIARIVS

As far as the horologiarius is concerned, the matter exercised the attention of scholars for a long time. Von Domaszewski acknowledges the existence of the rank/function in the military, in two different instances, first using as evidence CIL 3.1070, from Apulum,Footnote 43 second using as circumstantial evidence CIL 13.7800, from Rigomagus.Footnote 44 Connecting the evidence from Apulum to the religion of the Roman army, he argues that this dedication is related to the cult of the buildings inside the fort, in this case suggesting that the building has a sacred character, as the sanctuary of the official clock of the fortress.Footnote 45 Later, in Rangordnung, he would state that the horologiarius is a rank equivalent to the principales of the legion, attending to the official clock of the fortress,Footnote 46 sometimes presumably also signalling to the horn-blowers the time to announce the change of guard.Footnote 47

Even if evidence is scarce for this function, there are at least three epigraphic examples coming from Rome that can be related to it, all connected to the uigiles. Footnote 48

The first example is a construction plate dated to a.d. 113, inaugurating an aedicule, mentioning the commanding officers of the uigiles and a series of twelve principales, belonging to the same centuria. The men are part of the centuria of C. Iulius Rufus and occupy the following functions: beneficiarius subpraefecti, beneficiarius tribuni, uexillarius, two optiones, two secutores tribuni, tesserarius, two librarii, bucinator and the peculiar HOR LEG C. Peturcius Pudens. This last function has always been considered as an abbreviation for horrearius legionis,Footnote 49 even if this is not probable for two reasons: the role has no place within the uigiles, in a centuria, between two secutores; second, the function is not attested elsewhere in the army.Footnote 50 Moreover, the horrearii are attested exclusively in the civil environment and in the central administration in Rome, and are, in their vast majority, slavesFootnote 51 or freedmen. This is not to say that the army had no horrea, or no soldiers attending to them, simply that no military horrearii are attested.

It is perhaps preferable to presume either a misspelling by the stonemason, who meant to inscribe HORLOG, which, if acceptable, would lead us to the more obvious development of hor(o)log(iarius), the principalis associated with the building and maintenance of the official clock in the fort, or to hor(ologiarius) leg(ionis), but then we cannot explain his involvement with the uigiles. The first option is more plausible also because the same misspelling is noticeable in CIL 13.7800, mentioned earlier, with horolegium instead of horologium. From this association, we can assume that all the military camps had at least one official water clock, but perhaps not all units had a non-commissioned officer in charge of its construction and maintenance. The duty of reading and announcing key moments throughout the day may have been available to any member of the military.Footnote 52

The other two inscriptions that can be considered as evidence for the horologiarius are the lists of cohors V uigilum, from a.d. 205 and 210 containing all their officers and soldiers.Footnote 53 One particular individual is of concern to us, M. Ulpius Irenaeus, present in both of the lists, with a supplemental designation as HO in a.d. 205 and HC in a.d. 210. As in the case of C. Peturcius Pudens from CIL 6.221, HC has been understood several times as h(orrearius), this time c(ohortis),Footnote 54 but the actual succession of letters is, in both cases, H followed by a reversed 7 (the centurion sign).Footnote 55 Such an equation to horrearius cohortis is never heard of before; therefore, by analogy to CIL 6.221, we could consider Irenaeus also a h(orologiarius) 7(centuriae).Footnote 56

There is, however, more to be said regarding Irenaeus. In the five-year span between the two lists, his career does not change in any way. In a.d. 205 he is part of the second centuria, of Aelius Torquatus, the eighty-seventh on the list and by a.d. 210 he remains in the same second centuria, now having a different centurion, Rufinus, mentioned fourth on the list. The interesting fact is that only Irenaeus remains in the same position for at least five years. His centurion is promoted or transferred, and the vast majority of his colleagues are no longer with him by a.d. 210. In fact, only forty-seven soldiers are to be found in the same second centuria over the five years, and only one has been promoted to codicillarius in this period, from a grand total of 167 (a.d. 205) and 140 (a.d. 210). We can assume that the man was rather unique and useful in his function, that of clock-making and maintenance, such specialists being in fact difficult to find. Placed by these three soldier-lists in the ranks of principales, indicative of an increase in pay, we would assume naturally that he is considered a technical immunis. Perhaps the solution for the identification of his position is related to the complex matter of the military medici, architecti, mensores, libratores or other categories of civilians contracted in the army, considering also his Greek origin, common to technicians and specialists.Footnote 57 On the other hand, whereas in a military camp the medical and architectural technicians were more or less constantly engaged in matters that required their technical competence, new clocks would only occasionally be required, but building them needed an in-depth astronomical qualification.

One of the most significant pieces of epigraphic evidence in this regard is the previously mentioned inscription from Apulum. In the first place, it had a tumultuous history. On the one hand, M. Ulpius Mucianus was considered to be a horologiarius only by von Domaszewski and later by von Petrikovits and Le Bohec,Footnote 58 while Dessau remained undecided,Footnote 59 but consequently the character's fate changed. The inscription was discussed in several papers with different purposes, published across twenty years, all of them reaching the consensus that Mucianus was not a horologiarius, but that he erected a horologiarium templum, a sanctuary enclosing or serving as a clock.Footnote 60 As mentioned above, the horologia were mostly adjacent or part of larger monuments or buildings, but there is no mention of a temple with a clock in this manner of topic, only as horologium cum … .Footnote 61 Very recently, the inscription has been reconsidered by Bonnin, who seriously doubts the solution regarding a temple provided with a horologium, especially considering the ὡρολογιά[ριος] of Nikaia.Footnote 62

The two examples mentioned above, from Nikaia, Bithynia and Pontus,Footnote 63 especially the example transliterating the actual Latin term in Greek,Footnote 64 serve as precise analogies, even if they are discovered in civilian environment. They attest a function related to clock-building and maintenance—the horologiarius—that must have been, as in many other cases, contracted by the army.

Another matter is that of the dating, because the inscription is clearly dated to a.d. 212–217, but the consuls were in office in a.d. 193.Footnote 65 Probably the pledge was made in connection with the accession of Severus to emperor.Footnote 66 However, this is difficult to establish since the actual stone is gone and no assumptions can be made on the integrity of the texts and their belonging to the same inscription.

M. Ulpius Mucianus bears an obvious Thracian name and the origins of his citizenship go back to the reign of Emperor Trajan. He is a soldier of the legion XIII Gemina, in service. His status is clearly and continuously stated: mil(es) leg(ionis) XIII Gem(inae) / horologiar(ius). We have no reason to assume that he was not a soldier with this rank, simply because a regular soldier of the legion could not have erected (a solo de suo ex uoto fecit) a temple of Jupiter and Juno.Footnote 67 In addition, this is the only abbreviation HOROLOGIAR known until now in Latin inscriptions, and can most feasibly be understood as horologiarius. Taking into account the Greek transliteration ὡρολογιά[ριος],Footnote 68 we have thus the perfect analogy. Not least, the Oxford Latin Dictionary Footnote 69 considers horologiarius an adjective, determining the presence of a horologium on/in a building, but the argument is supported by the misread of CIL 3.1070. The precise problem of this reading of the inscription is the association of this adjective to the noun templum, and for this we have no other epigraphic analogy, therefore leading to an understandable vicious circle. Moreover, the rank of a soldier is described with a noun, in many instances mentioned after miles legionis.Footnote 70 The inscription seems to have an official character, mentioning the emperor and dated by consuls, meaning that it must have referred to a hydraulic clock, rather than to a sundial, which would have been easier to maintain.Footnote 71

5. CONCLUSION

In the military, the need to keep good time has always been important. The Roman army needed such timekeeping devices in order to organize the working day, preparations for battles and sieges and most importantly the night watch. The only way in which this was possible was to have one or several official clocks in the fort, so that everyone could tell the time and conform to the schedule. Since all ancient clocks depend directly on geographic coordinates, latitude and longitude, it was necessary for each unit to have its own timekeeping instruments. Even if Caesar makes specific mention only of the use of a water clock in the army, we may assume the presence of such a facility in all permanent military installations. Its attestation, amongst other instances, in the legionary fortress of Apulum, in the vicinity of a provincial capital, simply points out to the importance of telling precise time in a military environment and also to the importance of the individual, probably the only such specialist with such expertise in the province, for which he was contracted and enlisted in the legion. However, telling the time was not a particular skill, therefore no special personnel were needed for this. For their construction, on the other hand, the specialist horologiarii were likely to be needed.

The inscription of Apulum, backed by the other epigraphic examples discussed above, seems to be the best evidence we possess on this particular function in the army. The horologiarius was probably not a simple clock attendant and reader, but his importance and specialization can also be connected with clock-building and maintenance. On the one hand, this skill was available only to those who had been initiated in astronomy,Footnote 72 people who were sought by civil or military communities in order to equip themselves with timekeeping devices. On the other hand, as we can notice from the example of Narbonensis,Footnote 73 the operation and administration of such a public work was entrusted to a lower category of people, perhaps to slaves.

Footnotes

*

This work was supported by a grant of the Romanian National Authority for Scientific Research, CNCS-UEFISCDI, project number PN-II-RU-TE-2011-2-0273. In the completion of the present paper, the feedback and advice of Prof. Bruce Gibson was of paramount importance, and for this I wish to express my gratitude to him. For any errors I am solely responsible.

References

1 Most of them gathered by Fink, R.O., Roman Military Records on Papyrus (Ann Arbor, 1971)Google Scholar.

2 That of Bongarsius is the most accurate and reliable acc. IDR 3/5, 148. See also the partial drawing of Ariosti, which has the same text. He took the monument to Vienna, where it was consequently lost. See Buonopane, A. and Monaca, V. La, ‘Le iscrizioni della Transilvania nel codice Veronese di Giuseppe Ariosti (Biblioteca Capitolare, cod. CCLXVII)’, in Marchi, G.P. and Pál, J. (edd.), Epigrafi romane di Transilvania raccolte da Giuseppe Ariosti e postillate da Scipione Maffei. Bibliotheca Capitolare di Verona, Manoscritto CCLXVII. Studi e ricerche (Verona and Szeged, 2010), 245–374 Google Scholar, at 272 and Ariosti, G., Inscrizioni antiche della Transilvania (Vienna, 1723), I.XIXGoogle Scholar.

3 Piso, I., Inscriptions d'Apulum (Inscriptions de la Dacie romaine – III 5) (Paris, 2001), 147–8Google Scholar = IDR 3/5.

4 As it has been considered first by Domaszewski, A. von, ‘Die Religion des römischen Heeres’, WZ 14 (1895), 1–129 Google Scholar, at 103; cf. id., Die Rangordnung des römischen Heeres (Bonn, 19081), 46. Contra, see n. 62 below.

5 The most complex description is that in Book 9 of Vitruvius (De arch. 9.7.1–8.14); see also Plin. HN 7.212–15; Gell. NA 3.2.1–16; Varro, Ling. 4.4 and Censorinus, De die natali 23.7.

6 Bonnin, J., La mesure du temps dans l'Antiquité (Paris, 2015), 60–3Google Scholar. Confusion in the precise terminology is noticeable throughout these texts. Varro (Ling. 4.4) debates the terminology of the solarium, the sundial, which shows the hours of the day from the Sun, and was first placed by Scipio Nasica near the basilica Aemilia and the basilica Fulvia. Pliny (HN 7.212–15) places the introduction of hours in Rome after the Twelve Tables, but gives more precise dates for the adoption of the Greek sundial from Catania, after the First Punic War (263 b.c.), and also for the building of the first sundial adapted to the latitude of Rome (164 b.c.). On the matter of the clock placed by Scipio Nasica, he argues that it actually is a complicated water clock, made for timekeeping also during the night (159 b.c.). Censorinus (De die natali 23) confirms this chronology and adds that the Romans were the first to measure the day from midnight to midnight and that the Twelve Tables divided the day into only four parts, like the military uigiliae at night.

7 For general information and terminology, see RE VIII.2, cols. 2416–28; DNP 12/1, cols. 971–3; Gibbs, S., Greek and Roman Sundials (New Haven and London, 1976); Bonnin (n. 6), 85–6Google Scholar; and, for Marcu, Dacia F., ‘The sundial from Florești’, in Pop, H. et al. (edd.), Identități culturale locale și regionale în context european. Studii de arheologie și antropologie istorică. In memoriam Al. V. Matei (Zalău, 2012), 533–8Google Scholar.

8 See RE VIII.2, cols. 2428–33; DNP 12/1, cols. 973–6. For the clarification of the term, see Bonnin, J., ‘Wasseruhr und Klepsydra. Zeitmesser der Antike’, in Sonne Zeit, Rundschreiben der Arbeitsgruppe Sonnenuhren im Österreichischen Astronomischen Verein 45 (June 2013), 11–13 Google Scholar and id., (n. 6), 56–60 and 85–6.

9 Bonnin (n. 8), 11–12 records several references to it in Greek literature (Aristophanes, fifth century b.c.), mentioning it as a couple of pots for the controlled draining of water, used as a stopwatch in courts, and noting that archaeological finds are few, but relevant, especially from Athens in the fourth century b.c.

10 Bonnin (n. 8), 12–13. Also worked by water, it can measure significantly larger amounts of time and can record the hours' succession (horologion, hydrion horoskopeion, hydrologion, aqua horologium, hor. hibernium).

11 Carcopino, J., Daily Life in Ancient Rome: The People and the City at the Height of the Empire (New Haven, 1940)Google Scholar; Wright, J.K. and Lobeck, A.K., ‘Man and time in ancient Rome: notes on a recent publication’, Geographical Review 31 (1941), 659–62CrossRefGoogle Scholar.

12 Bonnin (n. 6), 63–5.

13 Bonnin (n. 6), 63–4.

14 When considering that by 190 b.c. the month of March fell in October, the matter of the hours of the day seems not so significant: see Derow, P., Rome, Polybius, and the East. Edited by A. Erskine and J. Crawley Quinn (Oxford and New York, 2015)Google Scholar, especially 212, 214–15, 222–35 and Bonnin (n. 6), 64–5.

15 The first mention of a horologium in the Roman world comes from Alatrium, datable between the years 134 and 90 b.c.—CIL 12.1529: see Bonnin (n. 6), 70.

16 AE 1975.232 (Italy): h. praetorii; CIL 2.4316 (Taracco): h. collegii fabrum; CIL 2.93 (Baetica): in a public place; CIL 8.978 (Africa) and 10.831 (Italy): scholas item h.; CIL 8.25533 (Africa): h. with columns and portico; CIL 9.2334 (Italy): h. with a table; CIL 13.11978a (Upper Germany): h. et aedes cum ornamentis suis omnibus et signis; CIL 5.2035 (Italy) and IRC 3.38 (Spain): h. cum sedibus; CIL 6.10237 (Rome): h. with a marble basin; CIL 10.5807 (Italy): h. in a building complex; CIL 12.3100 (Nîmes): h. cum II cerulas argenteas; AE 2005.454 (Italy): h. distylis signisque (with two metal rods); CIL 12.2522 (Narbonensis): h. cum suo aedificio et signis. The earliest discoveries of sundials come from Umbria and Pompeii, both datable to the second century b.c., according to Bonnin (n. 6), 69.

17 Also mentioned by Vitruvius (De arch. 1.6). Kienast, H.J., Der Turm der Winde in Athen (DAI Archäologische Forschungen Band 30) (Wiesbaden, 2014), 120–8Google Scholar. In fact, this complex installation held an armillary sphere inside, giving it an astronomical and astrological role, to present the mechanisms of the universe.

18 CIL 12.535, Italy.

19 CIL 12.2522, Narbonensis: … ad id horologium administrandum seruum.

20 AE 1992.1620 (Bacakale, Asia): … officina horologi caesura. See also Christol, M. and Drew-Bear, T., ‘Les carrieres de Dokimeion a l'epoque Severienne’, Epigraphica 53 (1991), 113–74Google Scholar, at 135–7: the officina could have served for the fabrication or most likely the restoration of a monumental horologium.

21 SEG 36.1153: Αἰλιανὸς Ἀσκληπιόδοτος γνωμονηκός—dedication to Nemesis by a clock-maker; and IGRR 3.1397 (a.d. 288–289): an unknown ὡρολογιά[ριος] τῆς τετρακονίας—attesting the clock-makers of the metropolitan rural community around Nikaia; both of Nikaia, Bithynia and Pontus. See also Bonnin (n. 6), 85–6.

22 Bonnin (n. 6), 68.

23 Bonnin (n. 6), 71.

24 Whitrow, G.J., Time in History. Views of Time from Prehistory to the Present Day (Oxford and New York, 1988), 28Google Scholar.

25 Bonnin (n. 6), 267–8.

26 militari ergo gradu XX milia passuum horis quinque dumtaxat aestiuis conficienda sunt.

27 Phang, S.E., Roman Military Service. Ideologies of Discipline in the Late Republic and Early Principate (Cambridge, 2008), 213CrossRefGoogle Scholar.

28 See also Remijsen, S., ‘The postal service and the hour as a unit of time in antiquity’, Historia 65 (2007), 127–40Google Scholar, at 140.

29 Polyb. 9.14–15.

30 App. B Civ. 1.93—late in the afternoon; Plut. Sull. 29.3–6—at the fourth hour/daybreak.

31 Caes. BGall. 5.13: nos nihil de eo percontationibus reperiebamus, nisi certis ex aqua mensuris breuiores esse quam in continenti noctes uidebamus: see Bonnin, J., ‘Time keepers in Britain 43–780 a.d. Origins, the Roman contribution, and Anglo-Saxon continuity’, British Sundial Society Bulletin 22 (2010), 34–7Google Scholar, at 36, and id., (n. 6), 268.

32 See Fink (n. 1), nos. 1, 2, 9, 10 and 47.

33 Polyb. 6.35.6–36.9; Veg. Mil. 3.8.16–18; Dig. 49.16.12.2. Phang (n. 27), 213.

34 Bonnin (n. 6), 270 reports eight such finds, from which six are related to forts. See also Marcu (n. 7).

35 Lewis, M., ‘A Roman clock at Vindolanda’, Current Archaeology 228 (2009), 12–17 Google Scholar; Bonnin (n. 31), 36; Meyer, A., ‘Notes on the Vindolanda “calendar”: related artefacts and the purpose of the Vindolanda fragment’, in Collins, R. and McIntosh, F. (edd.), Life in the Limes. Studies of the People and Objects of the Roman Frontiers (Oxford, 2014), 109–15CrossRefGoogle Scholar.

36 Lewis (n. 35), 16–17; Bonnin (n. 31), 35 and id., (n. 6), 270 (A_338).

37 Meyer (n. 35), 109–10.

38 Bonnin (n. 31), 35–7.

39 Remijsen (n. 28), 135–6.

40 Dunand, M., Mission archéologique au Djebel Druze. Le Musée de Soueïda. Inscriptions et monuments figurés (Bibliothèque Archéologique et Historique 20) (Paris, 1934), 77 n. 162Google Scholar (fig. XXXIII); AE 1936.147; Bonnin (n. 6), 270.

41 See Horster, M., Bauinschriften römischer Kaiser. Untersuchungen zu Inschriftenpraxis und Bautätigkeit in Städten des westlichen Imperium Romanum in der Zeit des Prinzipats (Stuttgart, 2001), 188–207 Google Scholar.

42 Petrikovits, H. von, Die Innenbauten römischer Legionslager während der Prinzipatszeit (Opladen, 1975), 75CrossRefGoogle Scholar and n. 79; Johnson, A., Römische Kastelle: des 1. und 2. Jahrhunderts n. Chr. in Britannien und in den germanischen Provinzen des Römerreiches (Mainz, 1987), 124–5Google Scholar.

43 von Domaszewski (n. 4 [1895]), 103.

44 von Domaszewski (n. 4 [1908]), 46.

45 von Domaszewski (n. 4 [1895]), 103. Worshipped in the same way as the aedes principiorum, the clock and the standards could moreover have shared the same space.

46 von Domaszewski (n. 4 [1908]), 46.

47 Bohec, Y. Le, L'armée romaine sous le Haut-Empire (Paris, 1989), 52Google Scholar. However, there is no ancient evidence for this.

48 CIL 6.221, 1057 and 1058.

49 von Domaszewski (n. 4 [1908]), 46, 14; Baillie-Reynolds, P.K., The Vigiles of Imperial Rome (London, 1926), 88Google Scholar. Dessau, however, doubts this assumption: ILS 2160 (dubitans proposuit Henzen).

50 Sablayrolles, R., Libertinus miles. Les cohortes de vigiles (Collections de l’École Française de Rome 224) (Rome, 1996), 232–3Google Scholar. Only a genius horrei is mentioned in connection with the military horrea.

51 Called precisely seruus in CIL 6.682, 8682; AE 1992.3723 or actor in CIL 6.9108.

52 Perhaps not at every hour, but at least for the change of the night guards, the horn-blowers would be asked to mark the ending and the beginning of each uigiliae. See Veg. Mil. 3.8.16–18.

53 CIL 6.1057 and 1058.

54 von Domaszewski  (n. 4 [1908]), 14; Baillie-Reynolds (n. 49), 88.

55 Sablayrolles (n. 50), 232.

56 Sablayrolles (n. 50), 233.

57 See Davies, R.W., ‘The medici of the Roman armed forces’, Epigraphische Studien 8 (1969), 83–99 Google ScholarPubMed; id., ‘The Roman military medical service’, Saalbuch Jahrbuch 27 (1970), 84–104 Google Scholar; Stoll, O., ‘ Ordinatus architectus. Römische Militärarchitekten und ihre Bedeutung für den Technologietransfer’, MAVORS 13 (2001), 300–68Google Scholar; id., ‘ Medicus centurio (PSI 1063). Ein Sanitätsoffizier mit taktischem Kommando? Probleme, Hypothesen, Lösungen’, Jahrbuch des römisch-germanischen Zentralmuseums Mainz 50 (2003), 329–54Google Scholar and Baker, P.A., Medical Care for the Roman Army on the Rhine, Danube and British Frontiers in the First, Second and Early Third Centuries a.d. (BAR IntS 1286) (Oxford, 2004)CrossRefGoogle Scholar.

58 See above, von Domaszewski (n. 4 [1908]), 46; von Petrikovits (n. 42), 75 and n. 79; Le Bohec (n. 47), 52, and id., Die römische Armee (Stuttgart, 1993), 55.

59 In ILS III.2 (page 732) he agrees that horologiarius is inter officia militaria, and refers to page 489, where the text of the inscription of Apulum is given, with no solution for the term horologiar.

60 Mommsen in CIL; Tudor, D., ‘Les constructions publiques de la Dacie romaine d'après les inscriptions’, Latomus 23 (1964), 271–301 Google Scholar, at 294; Fitz, J., Honorific Titles of Roman Military Units in the 3rd Century (Budapest and Bonn, 1983), 62Google Scholar; Moga, V., Din istoria militară a Daciei romane. Legiunea XIII Gemina (Cluj-Napoca, 1985), 42Google Scholar and Piso (n. 3), 147–8. Forisek, P., ‘Inscriptions of the Roman Dacia in the works of Tauriunus and Reicherstorffer’, in Németh, G., Piso, I. (edd.), Epigraphica II. Mensa rotunda epigraphiae Dacicae Pannonicaeque. Papers of the 4th Hungarian Epigraphic Roundtable. 1st Rumanian-Hungarian Epigraphic Roundtable, Sarmizegetusa 2003 (Debrecen, 2004), 237–53Google Scholar, at 246–8 is undecided.

61 See n. 15 above.

62 Bonnin (n. 6), 268–9.

63 See n. 22 above.

64 IGRR III.1397.

65 Falco was consul ordinarius in a.d. 193, allegedly plotted against Pertinax and was spared (SHA Pert. 10.1–7); therefore, he was likely to have left the consulship before the accession of Didius Iulianus and subsequently Severus. See further Champlin, J., ‘Notes on the heirs of Commodus’, AJPh 100 (1979), 288–306 Google Scholar, at 300–5.

66 IDR 3/5, 148.

67 Acc. IDR 3/5, 148.

68 See nn. 22 and 65 above.

69 OLD s.v. horologiarius.

70 CIL 3.12074; 5.6785; 8.18291; AE 1968.605: miles leg. … beneficiarius. ILS 4510a: miles leg. … signifer. CIL 3.6108, 8201; 6.232, 3342, 3349, 3355, 3366; 10.1771: miles leg. … frumentarius. CIL 13.7943: miles medicus, etc.

71 See also CIL 13.7800, Rigomagus, with Bonnin (n. 6), 269.

72 See n. 21 above.

73 See n. 19 above.