TIME is the basis of history, as space is of geography or matter of the physical sciences. Until some method of keeping accurate track of it was discovered, the data of history were like an uncharted land or an unanalyzed substance. To us with our almanacs this seems like the simplest matter of observation and arithmetic, merely a counting of days, weeks, months and years. But when history began there were no almanacs or calendars to consult. Weeks were unknown, months were observed only from the superstitious fears and beliefs attached to the changing moon, and the revolving years were too vast and vague extents of time to be measured off with any accuracy. There are really only two measures of time of which the primitive mind is fully conscious: the day — and one day is like another; and the season — and the seasons vary. A little thought shows that whole new sciences had to be evolved before the dates could be set along the margin of our annals — the sciences which make possible astronomy and through it a settled calendar for events that recur, and a fixed chronology for those which happen but once.1
Anthropologists point out that the greatest social revolution of primitive mankind came about when men, settling on the soil instead of wandering, and so accumulating goods which involved foresight, began to calculate for a future. From that dim sensing of futurity in which civilization dawned, the whole evolution of society has been conditioned by some reckoning of the passing of time. The calendars upon our walls make this now so simple and familiar that the fact escapes our attention. But it takes considerably more thought than most people are ever likely to devote to it, to realize that the calendar itself is an invention rather than a discovery, an art-creation, magnificent in its mathematical 41 perfection, but a product of human ingenuity all the same, and not the mere revelation of some laws of nature.
Yet the artificial character of our calendar can be seen very easily. Some of our time-divisions are artificial on the face of them, — the divisions of the day and the massing of days into weeks. We could do without seconds or even minutes without much inconvenience; and do so most of the time. Even hours vary greatly. The twelve-hour unit comes to us from Babylon, through Ionian Greece, — twelve being like our ten, the unit of measurement for anything. We might as well have had a decimal instead of a duodecimal system; it all depended on the arithmetical tables one used. But one should not put too much stress on the hour as a division of the day, for, in general, it is only the point of time, within the hour or at its beginning or close, of which we are keenly conscious, — especially the time for commencing or quitting work. It is the same with weeks. There were none in ancient Greece or Rome. They, too, like the hours, came apparently from Babylon. They mark off seven days, because seven was a sacred number. Habits and religious beliefs have settled this cycle upon our minds with the weight of centuries; the rhythmic Sunday pause in our busy week-day industries impresses itself upon the imagination so that poetically inclined people attribute to nature itself a restful note upon the sacred day. But this is merely our tribute to social convention and taboo. Every day is a sun-day. Weeks are a fiction based upon superstition but perpetuated for their social value. Even now, however, there are many people who pay no attention to them; in the mills of modern industry, on railways or ships, where work continues without ceasing, the weeks are practically unrecognizable. But days, months, and years are different. Here nature seems itself to mark an interval. The turning of the earth on its axis, of the moon around the earth and the earth around the sun, seems to furnish real units. It was undoubtedly these which first gave men a mathematical idea of time. But when we come to apply the lesson, it is not so easy.
The calendar began in registering these celestial phenomena. The first chronometer was the universe itself; its ever-recurring movements struck off the days, months and years as our clocks now strike off the hours. The days and years are 42 thus in reality on a par with the minutes and the hours, only they are the product of a larger clock. Unfortunately, however, the clocks of the universe do not run together. The days do not fit the years and the months fit neither one. The exact solar year is not even 365¼ days, awkward as that multiple would be; it is 365 days, 5 hours, 48 minutes, 46 seconds! We have frankly given up trying to keep track of months that really go by the changes of the moon — a cycle that has no relation to our night or day. Yet this was the unit for twenty or thirty centuries in that home of astronomy, Babylon. When we pause a moment to consider these things, we begin to realize what baffling mathematics lies behind our calendars and almanacs. For there are the stars, too, to keep track of, with their revolutions and conjunctions, coming and going at all sorts of intervals, planets zigzagging across the heavens in crazy patterns, out of touch with everything, and yet somehow forming, apparently with sun and moon, a final unit, composing a universe. What a tangled problem for Babylonian and Egyptian astronomers to work out! No Chaldæan shepherds, “killing time” in pastoral loneliness and innocence, were ever able to evolve the science of astronomy. That venerable myth still lingers in respectable books; but astronomy was the product of learned priests, those first scientists and intellectual leaders, who developed it, through astrology, for the service of religion.
The calendar developed everywhere as a cycle of religious feasts. It was the gods, not men, for whom or by whom the days were first marked out. The times for hunting and fishing, for sowing and reaping, the phases of the moon, the summer and winter solstice, and the like, upon which the attention of primitive men was so forcibly directed, early became associated with some idea of miraculous power. The times themselves became “lucky” or “unlucky” — an idea still so common that we never stop to ask what it means.2 There was an uncanny power let loose in the world when the moon still hung visible in the sky by day, or under the blazing mid-summer sun. The primitive man cannot exactly tell whether the power is in the moon or sun or the day itself, but on that day he knows that it is there. So when animism produces its gods and 43 demons these days are consecrated to them. The time for reaping is sacred to the god of the harvest, and so forth. The old scruples take a more definite turn. A part of the time becomes the property of the gods. It is henceforth a violation of divine law to work or transact business on the days thus set apart. Holidays were at first genuinely holy days, and the calendar grew up around them. It was necessary to find some way by which the festival day, the dies nefastus, on which business was sacrilege,3 should not be violated. It was taboo; to violate it was not only wrong but dangerous. The power of an inherent curse, which is essential in the early idea of the sacred, protected it, and assured it social recognition. Accordingly it had to be kept track of in order to ensure that the proper ceremonies should be celebrated upon it. Hence the elaboration of that succession of religious feasts and fasts which still persists in our church calendar. The idea would not naturally occur to one that the lists of saints’ days and holy days which preface our liturgies are the historic remnants of the first marking of time. But in the practically universal superstitions about planting crops, gathering herbs or doing almost anything in the dark or the full of the moon we have a trace of something infinitely older than any sacred date in the prayer-book — a first vague fear of the unusual or uncanny, out of which theologies, as well as calendars, were born.
Once grant that days differ in their virtues, that some are good for one thing, some for another, it is of the utmost importance to know which is which. In Hesiod’s Works and Days we have the program outlined for the farmer of the earliest age of historic Greece. In the so-called Calendar of Numa we have the priestly reckoning for ancient Rome. But in Egypt and especially in Babylon, where the sky is so clear that, as the report ran in Rome, even the stars cast shadows, the mechanism of the heavens first produced an adequate system.
Babylon bears the proud title of Mother of Astronomy. It was a title already admitted by Greeks and Romans, to whom the words “Chaldæan” and “astronomer” (or rather astrologer) were practically synonymous. Modern scholars agree as to the 44 justness of the claim; but the careful study of newly found inscriptions places the scientific achievements of Babylonia and Assyria not at the opening but at the close of their long history. However much the priests of those distant centuries watched the heavens for portents and omens, their observations were not sufficiently systematic to enable them to measure the recurring periods of sun, moon and stars with that accuracy necessary for an unvarying calendar until after at least two thousand years of priestly lore. The Semites clung with the conservatism of superstition to the phases of the moon. Although they had grown civilized, — and civilization must arrange its work according to the sun, because nature does so too, bringing the recurring duties of the seasons, — these old desert-dwellers, and their neighbors who learned of them, never broke away from the lunar month and the lunar year.
No one knows when or how this reckoning was first adopted; but a study of primitive peoples the world over today shows that the moon and not the sun is generally the earliest guide toward the calendar. Wherever agriculture is not much developed, the moon dominates, owing both to its uncanny associations and to the shortness of its cycle. The origins of the lunar calendar of Babylonia, therefore, apparently lie beyond all the long story of its civilization. The records themselves carry us back, however, to the middle of the third millennium, when we find a Babylonian year of twelve lunar months, making up 354 days, with a thirteenth month thrown in once in a while — making that year 384 days — to bring the religious festivals and the business world right again. There was no absolute certainty as to what years should be lengthened and what ones should be left the normal length; the matter was in the hands of the priests. This unwieldy calendar spread throughout western Asia, wherever the cuneiform script carried the message of Babylonian culture, It was adopted by the Jews and, — apart from other fragments of it embedded in our calendar, — we still have a positive reminder of its difficulties in our festival of Easter.
But so much observation of the moon ultimately produced an astronomical cycle of great importance, that of the moon with reference to the sun. It was discovered that in nineteen years the moon returned to almost its original position with reference 45 to the sun,4 a period destined to be used for chronology by the Greeks. This discovery was not made until the eight and seventh centuries B.C., however, in that period when the study of the universe began to assume more calculable form, and astrology — still rooted in religion, but verging toward science — rose to supersede the crude old fantasies of the earlier and barbarous priestcraft. Then we come upon a strange and happy interworking of calendar and chronology. To foretell an eclipse, or a conjunction of the stars, it was necessary to know the period of time which had elapsed between such eclipses or conjunctions in the past. So, looking forward to forecast the future, the astrologer found himself obliged to consult the records of the past, and the more he sought for accuracy in his calendar the more he needed it in the royal or priestly annals which supplied him with the data upon which he had to build. In short, mathematics began to emerge from the position of a mere tool of superstition, in which the luck of numbers combined with that of the stars in a jumble of folly, and to assume its proper rôle as the basis of definite knowledge.
This was an epoch in the history of thought, an epoch of fundamental importance for history, for from that time to the present the years have been numbered in regular, unbroken succession. The list of the kings of Assyria whose dates are thus fixed and accurate began in the year 747 B.C., the first year of a somewhat insignificant monarch, Nabonassar. This list was used by the great astronomers of Alexandria, who finally worked out the problem of calendar and chronology as far as they were solved in antiquity, and it has been preserved in what is called the Canon of Ptolemy. Through these savants the Babylonian-Assyrian year was translated into the “fixed” year of Egypt, i.e. 365¼ days; and to the “Era of Nabonassar” were added those of the Persian and Alexandrian empires, and finally the list of Roman emperors, down to the year 160 of our era. So that from 747 B.C. until the present, the years have been kept track of in continuous, if varied, reckoning. But the Canon of Ptolemy was used by astronomers, not by antique historians.546
The mention of Alexandria naturally suggests the contribution of Egypt. But it was not Egyptian so much as Greek science which made the name of Alexandria so illustrious in antiquity, and the great astronomers who worked there found little in the long centuries of Egyptian culture to help them in their study of astronomy or chronology. This seems strangely paradoxical when one reads in modern histories of ancient Egypt of the great achievements of its science and, above all, that it bears an even prouder title than Babylon as the land which produced the solar year. The date when that event took place is a matter of dispute among Egyptologists; but if the calendar year of 365 days was introduced at a time when it fitted the solar year day for day, the nineteenth of July,6 4241 B.C., would be the first day of the year one of the new calendar. This date is reached by calculating back from a known date in the third century of our era, when a Latin writer, Censorinus, tells us that the solar year of Egypt was two months behind the calendar year. As the calendar year was about a quarter of a day short in length, it had been gaining on the solar year that much yearly, so that in 1460 years (4 X 365), it would gain a whole year. Thus, the two had coincided about 140 A.D., of which fact further evidence exists, and again at 1460-year intervals. The third of these, 4241 B.C. is, in the opinion of Professors Eduard Meyer and J. H. Breasted, the starting point for the invention of the calendar.7 It was a remarkable achievement. What long and puzzling computation, what tables of priestly science and records were at the disposal of those who inaugurated it, no one can tell. When one compares this solar year, only a little over six hours wrong, with the grossly inaccurate lunar year of 354 or 355 days in use in the rest of the world throughout most of antique history, it seems at first to indicate something like a Hellenic rationalism 47 at work in Egypt as long before the Greeks as we are since them. But this impression of Egyptian superiority is hardly borne out by fuller study. For, not only did Egypt fail to make good its early promise in astronomy,8 but by failing to rectify the error of a quarter of a day, its calendar year came to have no real correspondence with the solar year, as we have seen. And, finally, the ancient Egyptian did not know how to make the discrepancy between the official and the true solar calendar the basis of a calculation of dates in history. There is no trace of his having used the (Sothic) cycle of 1460 years. It is the modern scholar who uses it to check up his calculations.
In chronology, therefore, as in the calendar, the Egyptians have no such contribution as might be expected from the promise of their early texts. Moreover, the more detailed data for chronology are as irregular as the calendar. The years were numbered, not in a straight and continuous succession, but according to striking events, campaigns, the years of the pharaoh’s reign, or (especially) the levy of taxes. When the state was thoroughly organized, the treasury officials “numbered” the royal possessions every two years, and the regnal years were known as “Year of the First Numbering,” “Year after First Numbering,” “Year of the Second Numbering,” etc. Whatever knowledge the priests may have had of the period involved in the long succession of Egyptian dynasties, — and Hecatæus and Herodotus show that they had some, — it was left for the twentieth century A.D. to disentangle the problem for the world at large; and much is still to do.
The Babylonians and Assyrians had the practice of naming rather than numbering their years. There was some priestly or royal functionary whose duty it was to proclaim what event or man should give the name to the year. It was to be the year of the magistracy of so-and-so, or the year when a battle was fought or a city taken. There is a touch of casual history in this, but it is too haphazard to be of much use. For in the first place, one never knew, until the functionary made up his mind — perhaps toward the end of the year — what the year really was! Combine that with a lunar calendar, and one can see that there is work for the Babylonian scholar as he struggles with the problem of Sumerian date-lists, 48 which contain the names of the years, as recorded by the Babylonian scribes.9 Neither Greeks nor Romans worked out by themselves any adequate reckoning of time. The lunar year was the basis, and with all their ingenuity, they could not make it work. In Greece it was easily seen that the 354 days did not exactly fit the twelve lunations of the year, being short by 8.8 hours. So (if the old accounts are correct) they put in a month every second solar year, which brought the total up to about 7½ days more than the right amount. In order to meet this inaccuracy, the intercalation was then omitted every eighth year. This octaëteris or luni-solar cycle of eight years was in itself not rigorously exact and was not systematically carried out. In 432 B.C. the astronomer Meton proposed the 19-year luni-solar cycle, of which we have spoken above. It was not adopted, however, until the second half of the fourth century. Once adopted, it was naturally destined to play a very important rôle in later classical and ecclesiastical chronology. The astronomical cycle is really slightly less than 19 years, however, and further corrections were necessary. In fact so long as the motions of the moon remained the basis of reckoning, the calendar was sure to be imperfect.
The Romans began with a lunar calendar, but since they regarded odd numbers as the lucky ones, they made the year 355 instead of 354 days. Then they added a month every second year, inserting it between the 23d and 24th of February, so that the mean length was 366¼ days. To get rid of the extra day they had recourse to a clumsy device, — perhaps based upon the old Greek eight-year cycle, — ordering that every third period of eight years should have three instead of four intercalary months, and that they should be of 22 days each. This made the year 365¼ days. But the pontiffs were left discretion in adjusting the calendar to the needs of astronomy, and they seem to have adjusted it (in some cases at least) rather to the needs of their friends, — having long years when those who were in office whom they wished to favor, and short ones when their enemies were in power! In any case, the calendar fell into such confusion in the last years of the republic that it was out by three months, judging by the solar year. The decree 49 of Julius Cæsar was the result, fixing the year at 365 days with an extra day in every fourth year. The ancients have attributed the reform to the intercourse with the savants of Alexandria, but there is also some ground for connecting it with a simple old-fashioned solar year of Italian farmers, of which we have fragmentary but definite traces even in the official calendar, and which in its turn may have been affected by the farming calendar of the Greeks. If this be true we have a single line from Hesiod to Cæsar.
The first reformed year began on the first of January, 46 B.C. (708 A. U. C.). The months took their place in it,10 and then Christianity brought in the weeks from Judæa — and Babylon. The year remained, as we have seen, a fraction of a day too short, and there was no absolute agreement yet as to when it should begin. But these were matters never settled until the sixteenth and even the eighteenth centuries of our era.
We need to know this much of the origins of the calendar in order to complete our survey of antique chronology. In both Greece and Rome, — after the fashion of Babylon and Egypt, — the year bore the name of the ruling magistrates. In Rome it was named after the consuls, in Athens after the first archon, in Sparta after the first ephor, etc. As it was found necessary for practical purposes to keep lists of these, from the calendar we pass not only to chronology but to the crudest of annals.11 Thucydides, for instance, had only the Athenian list of archons, the Spartan lists of ephors and the lists of the priestesses of Hera in the temple of Argos to rely upon, in addition to the festivals.12 The cycle of 50 the Olympiad, the four-year period based upon the celebration of the Olympic games, by which later ages reckoned Greek history, was never used officially by the city states, and really was not taken over by historians and chronographers until about the end of the third century B.C. The credit for its introduction seems to belong to Timæus (c. 350 B.C.), an indefatigable antiquarian and historian whose philosophical cast of mind apparently left him free to indulge a singularly un-Hellenic taste for dates. But it was a geographer rather than a historian who finally attacked the problem of chronology in a critical spirit. Eratosthenes, who flourished about 276 to 194 B.C., and who, as librarian of the Alexandrian library, was equipped with the science of the East as well as with his native Hellenic genius, fixed the dates of the great epochs of Greek history in what was destined to be the accepted chronology of antique as well as of Christian historians. Into this we cannot go further at present.13 Nor need we do so for this chapter of our history of History. The crude old reckoning of Rome, from the fabled founding of the city, 753 B.C., and the Olympiads remained, for later classical antiquity, the two eras in general use.
Looking over this chapter of our intellectual evolution one is impressed with the slowness of its progress. The ancient world could come to its full maturity without any clear idea of the passing years, with even no accurate knowledge of what a year should be. Yet does not such vagueness correspond with our own experience? The past is all one to us; yesterday as dead as the centuries of Egypt. Only by the magic of memory can we even recall its faded color or catch an echo of its silenced voices. How that memory has become a social and undying heritage, a heritage that hallows its own possessions, is the theme of the chapters which follow on the history of History.
1 Cf. J. T. Shotwell, The Discovery of Time, in The Journal of Philosophy, Psychology and Scientific Methods, vol. XII (1915), Nos. 8, 10, 12.
2 See Hutton Webster, Rest Days, a Study in Early Law and Morality (1916), for an exhaustive survey of time taboos.
3 The Romans, characteristically viewing things from the practical point of view, had the terms inverted: the dies fasti were those on which business was permitted.
4 The time between eclipses was seen to be 18 years, 11 days, or 223 lunations (“Saros”).
5 The importance of the “Era of Nabonassar” for chronologists was first seen by Panodorus, the creator of the Alexandrian school of chronologers, in the opening of the fifth century A.D. See H. Gelzer, Sextus Julius Africanus und die byzantinische Chronologie (1898), Part II, p. 227, who traces the development of the Canon of Ptolemy through the Syncellus into Byzantine chronology and so opens up the connection with the Middle ages.
6 The day when the star Sirius rose at dawn, at the opening of the Nile floods.
7 For discussion, see J. H. Breasted’s Ancient Records of Egypt, Vol. I, Sects. 38 sqq. There is a short, clear account by H. R. Hall, The Ancient History of the Near East, p. 19. The chief protagonist of the longer chronology of Egypt, Professor Flinders Petrie, is now regarded as having been extreme.
8 It even failed to note eclipses.
9 There is a clear, short summary in R. W. Rogers, A History of Babylonia and Assyria (2 vols., 6th ed., 1915), Vol. I, Chap. XIII.
10 Julius Cæsar’s months were to be of alternate lengths, the odd numbers being 31, the even numbers 30 (except February). That would have made a a simple year to reckon with. But when the eighth month (the fifth in the old year) was named after Augustus, his vanity was gratified by adding a day to it to make it as long as that of Julius. Then, in order to avoid having three months of 31 together, September and November were reduced to 30, and 31 was given to October and December.
11 The vagueness of an idea of extent of time in Greek history can by seen by the fact that “generations” were used to help reckon time and this was roughly put at 33 years, although the period varies. In Herodotus one comes upon a system of 23 years.
12 The only continuous list of the Attic archons which has come down to us is a copy preserved in the history of Diodorus, but a growing body of inscriptions supplements it now, and enables the modern scholar to recover more than the ancients knew themselves.
13 Apollodorus of Athens, applying the conclusions of Eratosthenes, drew up a metrical Chronica in four books, dedicated to Attalus of Pergamum, which became the popular handbook on the subject. Both this and the works of Eratosthenes are lost, but fragments were preserved by the Christian chronologers, Julius Africanus, Eusebius, Jerome and Georgius Syncellus, and so this still is a primary base for the old Greek chronology. Vide H. Gelzer, Sextus Julius Africanus und die byzantinische Chronologie.