Reckoning Time in Ancient Rome

        Joseph Francis Alward


Some Christians1 claim that the ancient Romans counted hours relative to midnight, but nothing could be further from the truth. There exists in the historical records of ancient Romans an abundance of evidence that they counted daylight hours relative to sunrise and nighttime hours relative to sunset, but there is no document from that time which shows that the Roman's hour was referenced to midnight.


Few things about ancient Roman history are clearer than that the Romans reckoned daylight hours relative to sunrise and nighttime hours relative to sunset.


A search of the internet will confirm this.  One may find hundreds of references to the first, second, third, fourth, fifth, sixth, seventh, eighth, ninth, tenth, and eleventh hour of the day, or of the night, but nowhere is to be found a single reference to an hour beyond thirteen, and that's because at sunset (the twelfth hour) the counting started over for the nighttime hours.

Absence of evidence is not evidence of absence?  Not in this case.  Events of historical importance surely took place after 12:00 noon, so if counting hours relative to midnight actually was ever a Roman practice, as some Christians claim,  then surely one would find, for example, a 1:00 PM event--"the thirteenth hour" in the literature, or a 5:00 PM event—"the seventeenth hour."  But, no, there are no such hours, and that is clearly because the method of counting hours relative to midnight was never practiced. The total absence of the hours thirteen through twenty-four, and the huge number of examples in ancient writings of hours one through thirteen is almost conclusive proof that the Romans reckoned daylight hours relative to sunrise, and nighttime hours relative to sunset, and never reckoned any hours relative to midnight.

In the remainder of this article I will first provide an overview of the method of reckoning time in ancient Rome, and then provide several unequivocal examples of reckoning daylight time relative to sunrise, and nighttime hours relative to sunset.


Measuring Daylight Hours in Ancient Rome

The ancient Romans measured time relative to sunrise and sunset because these were unambiguous events and quite easily marked.  Thus, if sunrise occurred at 6:00 AM, then the "first hour" of the day would be from 6:00 AM to 7:00 AM, and the twelfth hour of the day would be from 5:00 PM to 6:00 PM, assuming that at this time of the year there were exactly twelve hours of daylight.2   Here is how Richard Welland, PhD, Latin editor of the Transparent Language series language translation computer software, explains it:

Beginning in the 3rd century B.C., Roman days were divided into two parts, the daytime and the nighttime, each with twelve hours. But since those two parts were defined by sunrise and sunset, which vary according to the season of the year, the individual hours of daylight were shorter than the hours of darkness in the winter, and longer in the summer. The hours were counted from sunrise: e.g., the "second hour" referred to the period between one and two hours after sunrise.

Confirmation that daylight hours were measured relative to sunrise is found on a number of different web pages.  Here is what the Latin literature instructors on the KET distant learning site have to say about the Roman method of counting daylight hours:



The sundial enabled the Romans to divide the day into 12 equal parts, or hours. The hours became a way to mark time and meetings. Courts opened at about the third hour, for example, and lunch was at midday, the sixth hour. People would go home to eat a leisurely lunch and take a siesta, returning to work in a few hours. People in Rome today still leave work at 1:00 and return to work from 4:00 to 7:00.



Measuring Nighttime Hours


Hours of the night were measured relative to sunset, so if the sun set at 5:00 PM our time, then the Roman "first hour of the night" would begin at 5:00 PM.  In summer, the sun might not set until 8:00 PM our time, so the first hour of that night in Roman time would have begun at 8:00 PM.  The following Commentary on the Pro Roscio Amerino 15-38 confirms this:


Day and night each had 12 hours, which were longer or shorter according to the season.  "After the first hour of the night" would correspond, in our terms, to after 9:00 PM or later in June, after 5:00 PM or earlier in December.


Water Clocks


Water clocks were filled at sunrise and used to measure the twelve "hours" of the day, then emptied and refilled at sunset to measure the twelve "hours" of the night.  Certain Christian apologists think that because official ("civil") Romans days began at midnight and ended at midnight, then the hours of the day likewise were counted from midnight, but this is quite false.  Rather than reckon hours from midnight to midnight, hours were reckoned relative to sunrise for the hours of the day, and to sunset for the hours of the night.  Here is how James Carcopino, author of Daily Life in Ancient Rome, describes it: 

The horologia ex aqua was built to reset itself, that is, to empty itself afresh for night and day. Hence a first discrepancy between the civil day, whose twenty-four hours reckoned from midnight to midnight, and the twenty-four hours of the natural day which was officially divided into two groups of twelve hours each, twelve of the day and twelve of the night.

The figure at the right is a diagram of the water clock designed by Ctesibus, a 2nd Century BC Greek Alexandrian inventor.


The practical value of reckoning daylight hours relative to sunrise, and nighttime hours relative to sunset is obvious:  Sunset and sunrise are easily observed events.   The reason there was never a time reckoning system based on midnight is that there is no natural event easily observed which marks the arrival of midnight.

Examples from Ancient Roman Writings

Below I present several examples of ancient Roman writings which record events at particular "hours" of the day or night.  In interpreting these writings, I will assume that sunrise occurs at 6:00 AM, and sunset is at 6:00 PM.  For the reader's benefit, I provide in the table the correspondences between our time and the various hours of the day or night described in the ancient writings.




Hour of Day


Hour of Night

  6:00 - 7:00 AM


 6:00 - 7:00 PM


  7:00 - 8:00


 7:00 - 8:00


  8:00 - 9:00


 8:00 - 9:00


  9:00 - 10:00


10:00 - 11:00


10:00 -11:00


11:00 - Midnight


11:00 - Noon


Midnight - 1:00 AM


Noon - 1:00 PM


1:00 - 2:00


  1:00 - 2:00


2:00 - 3:00


  2:00 - 3:00


3:00 - 4:00


  3:00 - 4:00


4:00 - 5:00


  4:00 - 5:00


5:00 - 6:00 AM


  5:00 - 6:00






The following are a few dozen examples of the use of "hours of the day" and "hours of the night" in the writings and practices of the ancient Romans.


"Wills executed in provincial towns…should be opened and read in the presence of the witnesses… between the second and the tenth hour of the day." 


If the midnight reference was actually in use, wills would be being read at 2:00 AM.  It makes much more sense for wills to be read from 8:00 AM to 4:00 PM, which are the second and tenth hours of the day in the sunrise reference system.


"A given angle or the sun at this moment marks,

Telling across the dial,

outside, the hour five." --Satire, Persius 

In the proposed midnight reference system, the "hour five" marked by the sun would be 5:00 AM, which is an hour when the sun is not even above the horizon.  The actual "hour five" must be the fifth hour past sunrise, or 11:00 AM.


"After January 1 next no one shall drive a wagon along the streets of Rome or along those streets in the suburbs where there is continuous housing after sunrise or before the tenth hour of the day."  (Law of Caesar on Municipalities, 44 BC) 


The safety law makes no sense if wagon travel is only restricted only from sunrise to 10:00 AM in the imagined midnight reference system, for there would be large numbers of people on the streets from 10:00 AM through 4:00 PM.  Restricted hours are obviously from sunrise to 4:00 PM, as provided for by the method of reckoning hours relative to sunrise.

"I remember that Asinius Pollio, the great orator, was such a man, whom nothing detained after the tenth hour:  he did not even read letters after that hour, lest some new cares should arise:  but in those two hours he laid aside the weariness of the whole day.  Some rested at midday and reserved some  lighter work for the afternoon.  Our ancestors also forbade any new motion to be made in the senate after the tenth hour. "

Asinius didn't rest from 10:00 AM until noon, but from 4:00 PM to 6:00 PM, and motions in senate after 4:00 PM, not 10:00 AM, were prohibited.)


"Accordingly, having disposed guards here and there along the road, about the tenth hour of the night, he set out by narrow paths through the woods, to fetch the corn into the town. But their noise being heard by the sentinels of our camp, and the scouts which we had sent out, having brought an account of what was going on, Caninius instantly with the ready-armed cohorts from the nearest turrets made an attack on the convoy at the break of day."

Scouts were sent at 4:00 AM, followed by an attack two hours later at sunrise.  Note:  if "tenth hour of the night" really were measured relative to midnight, it would have been 8:00 AM that the guards and scouts were sent out, and eight o'clock in the morning is surely not "night."


"Vitruvius reckons the hours best adapted for bathing to be from mid-day to about sunset (v.10). Pliny took his bath at the ninth hour in summer, and at the eighth in winter (Ep. iii.1, 8); and Martial speaks of taking a bath when fatigued and weary, at the tenth hour, and even later" (Epig. iii.36 x.70).… ary/SMIGRA*/Balneae.html 

In the imagined midnight reference, the ninth hour would be nine in the morning, but it's clear that Pliny took his bath between mid-day (noon) and sunset (6:00 PM), i.e., at about 3:00 PM in summer. Furthermore, Martial would not already have been "fatigued and weary" at 10:00 AM, as the proposed midnight reference system would have, but instead at 4:00 PM.


"Having been apprised of the arrival of Crassus by the scouts at about the third hour, he advances twenty miles that day." Gaius Julius Caesar "Military Commentaries


Advance of twenty miles after the third hour is not likely to have been achieved after 3:00 AM.  The third hour here is 9:00 AM.


The third hour was the ordinary time for Holy Communion, as may be seen from the decree (falsely) ascribed to Pope Telesphorus (a.d.. 127-138), in the Liber Pontificalis; "Ut nullus ante horam tertiam sacrificium offere praesumeret," and many other testimonies. 


Holy communion is at 9:00 AM,  not 3:00 AM.


"Public officers who attended on several of the Roman magistrates. They summoned the people to the assemblies, and those who had lawsuits to court; they preserved order in the assemblies and the courts, and proclaimed the time of the day when it was the third hour, the sixth hour, and the ninth hour."….html 


Court officers obviously did not call out the time at 3:00 AM, which would be the third hour in the non-existent midnight reference system.  The times were called at 9:00 AM, noon, and 3:00 PM.


"About two years after the last victory of Belisarius, the emperor returned from a Thracian journey of health, or business, or devotion. Justinian was afflicted by a pain in his head; and his private entry countenanced the rumor of his death. Before the third hour of the day, the bakers' shops were plundered of their bread, the houses were shut."   Chapter XLIII: Last Victory And Death Of Belisarius, Death Of Justinian. Part IV. 


Houses wouldn't be shut because of panic in the street at 3:00 AM, for they already would have been shut at that hour.  The time was three hours past sunrise, or 9:00 AM.


"About sun-rising; and that the first stone of Rome was laid by [Romulus] the ninth day of the month Pharmuthi, between the second and third hour." 


Second and third hours are relative to sunrise.


With the approach of midnight Domitian became so terrified that he jumped out of bed; and at dawn condemned to death a soothsayer …Presently he asked for the time. As had been prearranged, his freedmen answered untruthfully: 'The sixth hour,' because they knew it was the fifth he feared. (Robert Graves' translation)

Dawn had already come, so it was past 5:00 AM.  The freedman was concealing the fact that it was the fifth hour, but dawn had already come, so the time could not have been 5:00 AM; it must have been 11:00 AM, six hours after sunrise.


No duumvir holding an investigation or conducting a trial in accordance with this law, unless such trial is by this law bound to be concluded in one day, shall hold the said investigation or conduct the said trial before the first or after the eleventh hour of the day.

Trials cannot be conducted before 7:00 AM or after 5:00 PM.  Clearly, the sunrise reference system is being used here to reckon daylight hours.

And now the foreign envoys are introduced. The king hears them out, and says little; if a thing needs more discussion he puts it off, but accelerates matters ripe for dispatch. The second hour arrives ; he rises from the throne to inspect his treasure-chamber or stable.  "Measuring Time in Ancient Rome," Richard Welland Crowell, PhD.

King rises from throne at second hour, which is 2:00 AM in the now discredited midnight reference system.  Actual time is 8:00 AM.


Bonded commitment made to appear by Lucius Marius Hemeris the Kalends [Ides] of next November  at Puteoli in the Forum in front of the Hordionian Altar of Augustus at the third hour. 

If the third hour were referenced to midnight, then poor Lucius would have to appear before the judge at 3:00 AM.  The third hour is actually the third hour past sunrise, or 9:00 AM.

"The celebration in the strict sense of the word began at the second hour of the night of May 31. Sacrifices were offered to the Fates, on altars erected between the Tarentum and the banks of the Tiber"…*.html 

Celebration would not begin at 2:00 AM, so the second hour of the night must have been 8:00 PM.

"To-day I studied from the ninth hour of the night to the second hour of day, after taking food. I then put on my slippers" 

In the proposed midnight reference system, studying would have begun at 9:00 in the morning and ended at 2:00 AM, which is impossible.  This example makes it very clear that there were two different reference times.  Sunrise was the reference for reckoning daylight hours, and sunset was the reference for nighttime hours.  Thus, the studying was from 3:00 AM to 8:00 AM.

Next night, about the fourth hour, Ben-Hur stood on the terrace of the great warehouse with Esther. 

Ben-Hur would not  likely have been standing with Esther at 4:00 AM, which is the fourth hour according to the proposed—but nonexistent--midnight-reference system.  The fourth hour of the night is measured relative to sunset, or about 6:00 PM.  Thus, Ben-Hur was with Esther at 10:00 PM.

But when daylight returned, the conquerors…marched in a dense column upon Hadrianopolis…And to prevent the ardor of the soldiers from being cooled by delay, the whole city was blockaded by the fourth hour 

The fourth hour could not have been 4:00 AM, as the midnight reference system would provide, because daylight had already come.  The actual time, of course, must have been 10:00 AM.

"Thereafter, when the fourth hour in heaven has gathered thirst and the note of the shrill tree-crickets pierces the copses, by wells or by deep pools I will bid the flocks drink the wave that runs in troughs of ilex; but in the noonday beats seek some shady dell" (Roman poet Virgil)

Virgil would bid the flocks drink at 10:00 AM, not 4:00 AM.

But at last, urged by Decimus Brutus not to disappoint the full meeting, which had for some time been waiting for him, he went forth almost at the end of the fifth hour.

Brutus had meeting at 11:00 AM, not 5:00 AM

"The time for business began with sunrise, and continued to the fifth hour, being that of dinner, which with them was only a slight repast. From thence to the seventh hour was a time of repose; a custom which still prevails in Italy. The eighth hour was employed in bodily exercises; after which they constantly bathed, and from thence went to supper."  Ancient History Sourcebook: Pliny the Younger (61/62-113 CE): Selected Letters, c 100 CE


Business begun at sunrise wouldn't continue to 5:00 AM, which is what the fifth hour in the imagined midnight reference method of reckoning hours would be. The fifth hour is 11:00 AM, five hours after sunrise.

"While we were engaged in these discussions as fancy prompted each, appears an envoy from the cook to warn us that the moment of bodily refreshment is at hand. And in fact the fifth hour had just elapsed, proving that the man was punctual, had properly marked the advance of the hours upon the water-clock . The dinner was short, but abundant"  (Letter of Sidonius to his friend Donidius 461-467 AD) 

Dinner was five hours past sunrise, at 11:00 AM, not five hours past midnight, at 5:00 AM.

On the ninth day before the Kalends of February [January 24, 41 A.D.], at about the seventh hour he hesitated whether or not to get up for luncheon.  Ancient History Sourcebook: Suetonius (c.69-after 122 CE):  De Vita Caesarum: Caius Caligula The Lives of the Caesars: Caius Caligula), written c. 110 CE


Lunch seven hours past sunrise, 1:00 PM, not seven hours past midnight, 7:00 AM.

"It was now the eleventh hour and the chariot was still stuck in the ditch because the chariot driver was not able to move it. The Cornelians thought that maybe they should spend the night in the inn. It was getting darker now."  Accident on the Appian Way.


It couldn't be the eleventh hour past midnight, for it was getting darker.  It was 5:00 PM and getting darker.

"Yet, though assailed by so many disadvantages, [and] having received many wounds, they withstood the enemy, and, a great portion of the day being spent, though they fought from day-break till the eighth hour, they did nothing which was unworthy of them."  Commentaries:  Gallic Wars 

A "great portion of the day" fighting means the fight was eight hours of fighting from dawn to 2:00 PM, not two hours of fighting from dawn to 8:00 AM)

At the third hour, Silanus formed the soldiers into long ranks so that they could [or "to"] salute Agricola. When they saw [literally: "had seen"] Agricola entering the camp, they raised a great clamor: 

If the midnight reference system existed, the third hour would have been 3:00 AM.  Instead, the time is the third hour past sunrise, or 9:00 AM.

Next day, encircling the city from the sea by ships furnished with all kinds of missiles under the command of Laelius, and sending forward on the land side two thousand of his strongest men together with the ladder-bearers, he began the assault at about the third hour. Polybius: The Histories /ancient_rome/E/Roman/Texts/Polybius/10*.html

Assault at 9:00 AM not 3:00 AM.

"No Jew, my dear Tiberius, ever keeps such strict fast upon the Sabbath, as I have to-day; for while in the bath, and after the first hour of the night, I only ate two biscuits, before I began to be rubbed with oil."   Twelve emperors: by Suetonius Life of Augustus Chapter: 76

Augustus' bath at the first hour must have been at 6:00 PM, not 1:00 AM.

The festival took place in summer…The solemnities began at the second hour of the night, and the emperor opened them by the river side with the sacrifice of three lambs to the Parcae ("Ludi Saeculares") Ludi_Saeculares.html 

Festival began at 8:00 PM, not 2:00 AM.

I was writing this on the 23rd at the ninth hour of the night. Milo already holds the campus. The candidate Marcellus was snoring thus that I--his neighbor--hear it.  Cicero, Letters to Atticus 4.3 

Snoring at 3:00 AM, not 9:00 AM.

Accordingly, having disposed guards here and there along the road, about the tenth hour of the night, [Drapes] set out by narrow paths through the woods, to fetch the corn into the town. But their noise being heard by the sentinels of our camp, and the scouts which we had sent out, having brought an account of what was going on, Caninius instantly with the ready-armed cohorts from the nearest turrets made an attack on the convoy at the break of day.   "Bellum Gallicum" Book 8  50-51 BC 

There is no "night" ten hours after midnight, so it is obvious that nighttime hours are measured relative to sunset.

At Vipasca, in Lusitania (Portugal), the regulations for the Hadrianic procurators of the pithead baths include the duty of heating the baths for women from the first to the end of the seventh hour, and for men from the eighth hour to the end of the second hour of the night.

Women took baths from 6:00 AM to 1:00 PM, while men took baths from 1:00 PM to 8:00 PM.

The first and second hours cause those involved in the salutatio to rub shoulders,
The third sees lawyers active,
Rome extends its labors into the fifth hour,
The sixth will be a respite for the weary, the seventh, the end of labor.
The eighth and part of the ninth hour is sufficient for the sleek exercise rooms,
The ninth commands people to wear out couches piled up with pillows...

The poet Martial (Epigrams 4.8.1-6) gives the schedule for a typical Roman workday.

These times are obviously incompatible with the proposed midnight reference system.





1.  In the Gospel of John, the author's account of the time of Jesus' crucifixion apparently contradicts Mark's account.  John thinks that Jesus wasn't crucified until after about 12:00 PM noon, while Mark thinks that Jesus was already crucified three hours earlier, at about 9:00 AM. Here is the evidence:

About the sixth hour (hektos hora)…they shouted, "Take him away! Take him away! Crucify  him!" (John 19:14-15 NIV)

And it was the third hour when they crucified Him. (Mark 15:25)


Christian apologist have tried to make this problem go away by proposing that John was using what the apologists think was a Roman system of reckoning hours that refers all hours to midnight, in exactly the same way as time is measured in modern times: 



John wrote his gospel in Ephesus, the capital of the Roman province of Asia, and therefore in regard to the civil day he would be likely to employ the Roman reckoning. (Encyclopedia of Bible Difficulties, by Gleason Archer, page 364.) 



Gleason Archer believes that since John was writing his gospel in a Roman city, he would have used the Roman custom of reckoning time.  Archer believes that since the Roman "civil" day legally was the second half of night and all of the rest of the day and the first half of the next night—just as it is in our modern system, he assumed without evidence that perhaps the Romans counted their "hours" from midnight. Thus, midnight to 1:00 AM was the Roman first hour, and so on, in Archer's imagination.  In this system—which was in fact nonexistent, John's "six hour" was really Archer's 6:00 AM, not noon, and thus there would have been enough time after the sentencing in John's gospel for Jesus to be at the 9:00 AM crucifixion described in Mark's gospel.



The "midnight" reference system is all a figment of Archer's mind, as the remainder of this article will show.


2.  Only twice a year were there exactly twelve hours of daylight, and twelve hours of darkness:  at the vernal (spring) and autumnal (fall) equinoxes.  In the winter, there was less daylight, and more darkness.  On December 22 in Rome, there were only eight hours and 54 minutes of sunlight, and fifteen hours and six minutes of darkness.  Dividing the eight hours and 54 minutes (534 minutes) by 12, we see that the "hours" of daylight were only 44.50 minutes long, while the twelve nighttime "hours" each lasted 75.50 of our minutes. On December 22 in Rome, sunrise was at our 7:33 AM.  The following table shows the twelve hours of the Roman day on December 22 (adapted from data in Daily Life in Ancient Rome by Jerome Carcopino):



Hour of the Day


Rome, December 22


  7:33 (Sunrise) - 8:17


  8:17 - 9:02


  9:02 - 9:46


  9:46 - 10:31


10:31 - 11:15


11:15 - 12:00


12:00 - 12:44


12:44 - 1:29


  1:29 - 2:13


  2:13 - 2:58


  2:58 - 3:42


  3:42 - 4:27 (Sunset)