The Week (Colson)/Chapter 3

by F. H. Colson

§ 3

The Planetary Week

The conclusions to which I am led by the evidence given in this section may be summed up as follows. By the beginning of the third century A.D. the habit of measuring time in cycles of seven days, each of them dedicated to one of the seven planets, had become universal or at least general in private life throughout the Roman Empire, though it had not received official recognition. Evidence of the practice, though hardly proving a general adoption, can be traced back to at least A.D. 79. Passing over some doubtful testimonies, we find, as already stated, that a century before the last-mentioned date, the idea that the Sabbath was 'Saturn's day' had become current. Before this no trace of it can be found, and speculations as to its origin and antiquity must remain at the best mere speculations.

For a complete understanding of the evidence a preliminary explanation of one point is necessary. This is the order of the planets in the week. Now the normal arrangement of the planets according to their supposed distance from the Earth is as follows:

Moon, Mercury, Venus, Sun, Mars, Jupiter, Saturn.

This order, which, if we leave the Moon out of consideration, and put the Earth in the place of the Sun, is a correct order for the distance of the planets from the Sun and was based by the astronomers on their calculation of the time taken by each planet in its revolution, was the accepted order possibly from the days of Pythagoras and certainly from the second century B.C.^[1] through the imperial period, the Dark Ages and the Middle Ages down to the establishment of Copernican astronomy. Thus, to take two examples from earlier and later times, we find it in Cicero's treatise on Divination in which he includes astrology^[2]. And it forms the framework of Dante's Paradiso. Dante, as every reader will remember, ascends through seven zones beginning with the Moon and ending with Saturn, whence he passes into the eighth and ninth heavens known as the Primum Mobile and Empyrean which later astrology had added to the Seven.

On the other hand, the order of the planets in the week, as clearly reflected in the existing names, is:

Sun, Moon, Mars, Mercury, Jupiter, Venus, Saturn.

A full discussion of this difference between the two orders must be deferred, but we may here note that different as they are there is a definite relation between them. The week-order is obtained from the normal order by always dropping two. The moon is in the third place from the Sun—then if we go back to the top Mars is in the third place from the Moon, Mercury in the third place from Mars and so on till we finally get back to the Sun. Now the importance of this for our present purpose lies in the fact that when we find the Seven arranged in week-order instead of normal order, we may be fairly sure that those who thus arranged them knew and observed the planetary week.

This being made clear, we may pass on to the evidence^[3]. It will be convenient to start from the point when we know the planetary week to have come into general use and then working backwards to see how much earlier we can trace it. This point is found in a passage^[4] of Dion Cassius, an eminent and careful historian, who wrote the history of Rome from the earliest date to his own time. The history was actually carried up to the year A.D. 229, but various remarks shew that the first part is to be dated earlier, and we may fairly place the passage which now concerns us somewhere between the years 210 and 220. In his 37th book (the total number is 80, the majority of which are lost or exist only in abridgments), he has come to the capture of Jerusalem by Pompey in 63 B.C. and notes that this would never have been accomplished but for the Jews' reverence for 'Saturn's day.' Not only as this came round did they allow the Romans to continue their siege operations unhindered, but the final assault was also made on a 'Saturn's day' and met with no resistance. He then proceeds to give some account of the Jews and in particular how 'they have consecrated the socalled day of Saturn and while performing on it many observances peculiar to themselves lay their hands to no serious work.' The passage then goes on as follows:

And as for Saturn, his personality, the source of the honours thus paid to him and the nature of the superstitious awe shewn to him have been treated by many writers and have no connexion with this history. But the dedication of the days to the stars called planets originated in Egypt, but is now universal though its origin is comparatively recent. Certainly the old Greeks, to the best of my knowledge, knew nothing about it. But since now it is an established usage among the Roman as well as all other nations, and indeed may be called an accepted tradition with them in a sense, I wish to give some little discussion of the method of arrangement of the days.

Dion then proceeds to suggest explanations of the remarkable order of the planets in the week, and these I will give when we come to deal with that point. Meanwhile the passage so far contains three statements, (1) the planetary week originated in Egypt, (2) it is of recent growth, (3) it was in his time in general, indeed universal, use. Of these statements the first perhaps cannot be accepted as certain. The original source of a usage which spread so silently, as the week seems to have spread, is not an easy matter to determine for a modern and still less for an ancient observer. The second is couched in too vague language to be of much value. But the universality of the practice in his own time is a matter on which we may accept his authority as decisive, even if it were not confirmed, as it is, by contemporary evidence.

Of this evidence the most interesting item is a passage in Philostratus' Life of Apollonius of Tyana^[5], which is probably of almost exactly the same date as the part of Dion's history from which I have quoted. Philostratus was one of the greatest of the rhetoricians (a term which covers more or less our essayists, lecturers and preachers) of the third century. His most noteworthy work is this biography of Apollonius, a remarkable person, partly philosopher and partly charlatan of the first century. Philostratus relates how when Apollonius was travelling in India he received from Iarchas the chief of the Brahmins seven rings bearing the names of the seven planets, which he wore 'according to the names of the days.' Clearly this refers to the planetary week. Now Philostratus professes to derive this statement that his hero had a ring for every day of the week from a disciple or companion of Apollonius called Damis, whose memoirs had fallen into his hands. Whether there really was such a book, or if there was, whether it was really written by a companion of Apollonius has always been a moot question. If we take the story at its face value, since Apollonius' visit to India is supposed to be dated somewhere between A.D. 40 and 60 it would shew that some observance of a planetary week existed at that time. But this is not of real importance for my present purpose. We shall find other undoubtedly contemporary evidence that the planetary week was recognized by some people in the first century, and if so, it was only to be expected, indeed it might almost be taken for granted, that Apollonius, who had a strong leaning to the occult and was credited with a work on astrology, should have known and used it. The value of the passage lies in its wording. Philostratus wrote for the general reader. In fact his work has often been thought to be a counter blast to the Gospels. When he speaks of Apollonius wearing his planetary rings "according to the names of the days,' it is clearly assumed that the general public knows without further explanation that every day in a cycle of seven was named after a planet.

Another valuable witness is found or supposed to be found in a little Greek and Latin text book on grammar ascribed to one Dositheus and dated in A.D. 207, where we find a list of the days of the seven planets with the Latin and Greek names in the order Saturn, Sun, Moon, Mars, Mercury, Jupiter, Venus^[6]. Incidentally we observe that here we find the week beginning with Saturday, an idea which the next section will shew to be reasonable and natural. But a more important inference is that, as the grammar was no doubt intended for school uses we find that in the beginning of the third century, children had to learn the names of the week-days, much as I remember to have recited in childhood:

Thirty days have September, April, June and November, etc.

And at the same time it suggests to us that, though the week is in such general use that it has to be known, it is not so familiar that everyone is sure to know it without formally learning it. I never heard of school-children to-day being taught the names of the week-days. And this again is natural enough, for while our civil life is regulated by the week, it is, as we shall see, the most remarkable fact about the planetary week that it spread without any civil or official recognition—that nothing happened in it, as things happen to us on Sunday.

Further we may note that an inscription belonging to A.D. 205 has been found in Karlsburg in Transylvania, where the date is given not only by the year and the month, but also by the weekday, in this case Monday^[7]. Another, not quite so complete, but probably belonging to A.D. 231, comes from Kelheim in Bavaria, while representations of the Seven in week-order some of which at least are ascribed by experts to this period have been found in the Rhineland^[8]. All these go to shew that Dion's statement applies not only to the centre of the empire, but to its most outlying parts.

Finally, before we leave the early decades of the third century, we may listen to the witness of the Christian Church. Tertullian, writing about or possibly before A.D. 200, in two very similar passages scoffs at the pagan week^[9]. The details of these passages, as so often with this most difficult of writers, are obscure. But the general sense is something like this. Pagans often say that the Christians worship the sun. They think this partly from our custom of turning to the east and partly because we hold a festival on the Sun's day. But our religion is not sun-worship and if we rejoice on the Sun's day we do but follow the pagans, who dedicate the day which precedes it, Saturn's day, to leisure and feasting. Here we seem to have the suggestion that the day of the sinister planet had first, because of its unluckiness, been held to be unsuitable for active work and then by a natural transformation had become a festival or holiday. We have seen much the same thing happen with the old Scottish Fast-days and our own Good Friday.

That the planetary week was a matter of common knowledge and observance after Dion's day is attested by a great body of inscriptional and other evidence. But I need not weary the reader with the details, for the truth of the statement is shewn most effectively by the simple fact that in spite of the opposition of the Church the planetary names held their ground in the Christianized empire at any rate in the West^[10]. It is true that where ecclesiastical names were available they could prevail. As already noticed, 'Dominica' and 'Sabbatum' ousted Sun's day and Saturn's day over most of Southern Europe. But the plain numbers for the 2nd, 3rd, 4th and 5th days, though regularly used by orthodox Church writers, took no popular root^[11]. Our task is to find how far before Dion's time we can trace the planetary week. The first writer as we move backwards through the second century is Clement of Alexandria, one of the greatest and most liberal-minded of the Fathers. The Christian fast-days he observes^[12] are the fourth day and the 'preparation,' i.e. Wednesday and Friday, and in this the enlightened will see an allegory. For these days are named after Mercury and Venus, and in fasting on them we symbolize the larger truth that our whole lives must be a fast from avarice (Mercury) and lust (Venus). This utilization of a piece of paganism to point a Christian moral is entirely after Clement's manner and we must not draw from his words an inference, which would be contradicted by plentiful evidence, that devout or strict Christians were reconciled to the use of the planetary names. But his words do shew that the planetary week was an accepted usage of the pagan world, in which the Christian necessarily moved.

We next come to a famous passage in the Apology of Justin Martyr^[13]. This defence of Christian doctrine and practice addressed to the Emperor Antoninus Pius and his adopted sons about A.D. 150 has a chapter in which the Christian weekly meeting is described. The whole chapter has an immense importance for the student of Christian antiquities, but I need only quote a small part of it. 'It is on what is called the Sun's day that all who abide in the town or the country come together,' (here follows an account of the reading of the Gospels at the meeting and the celebration of the Eucharist), 'and we meet on the Sun's day because it is the first day on which God formed darkness and mere matter into the world and Jesus Christ, our saviour, rose from the dead. For on the day before Saturn's day they crucified him, and on the day after Saturn's day which is the Sun's day he appeared to his apostles and disciples and taught them the things which I have transmitted for your examination.' Two points in this passage may be noted. First Justin assumes that two of the planetary names, Saturn's day and Sun's day, will be known to the Emperors, but he does not mention by name the day of Venus. It has been suggested that he does not wish to sully his page with the name of the adulterous goddess; I cannot think that this is at all likely. Saturn also is the hero of discreditable stories which Christian fathers were not slow to satirize. I have been inclined to suggest that Justin was hazy about the week-names, or at any rate thought that the Emperor might not be familiar with them all. The Sun's day is mentioned by name, because it is the day with which the chapter is concerned. The mention of Friday is merely incidental and therefore the Emperor need not be troubled with its special name. If this is right, we might infer that the planetary week was not so well established in popular usage and knowledge as it was when Dion wrote 60 years later, though on the other hand we must not make too much of a perhaps accidental omission. The other point is the position which Saturday holds in Justin's scheme. That day too comes in only incidentally, even more so than Friday, for nothing is mentioned as happening on it. But it is evidently the pivot on which the week turns, the day from which the others are measured. For the present I only note this in passing. I shall recur to it, when we come to deal with the relations between the Jewish and the planetary week.

Another, but to my mind, doubtful, piece of evidence in the earliest part of the second century is connected with the Emperor Trajan (98–117). A biographer of the fourth century, Lampridius^[14], notes that Alexander Severus, Emperor from 222 to 235, made certain restorations or changes in the public baths. One of them was that he named a basin after Oceanus, whereas Trajan 'had dedicated the basins to the Days.' Presumably by this he means the seven deities of the week. If we took the statement as it stands, it would be of some importance. The most mysterious fact about the spread of week-observance is that it was apparently accomplished without official recognition, and here we might seem to have some semblance of official recognition. But I fear it cannot be relied on. By Lampridius' time the planetary week was universal and if Trajan had recognized the seven planets or planetary gods Lampridius would naturally suppose that this involved recognition of the week-days over which in his time they presided. But this is not necessarily the case. It is all-important to remember that reverence for the great Seven does not involve the idea that time is measured by them. The implication in Lampridius' statement could only be proved if we knew that Trajan arranged the basins, which no doubt were distinguished by planetary statues or symbols, in week-order instead of normal order, and this we do not know.

When we add to these a coin of the time of Antoninus Pius found in Egypt and bearing, among other figures, seven, which are supposed to represent the seven planets in week-order^[15], we shall, I believe, have exhausted the evidence which may with any certainty be assigned to the second century^[16]. We may now pass on to the first, and here we have the testimony of no less a name than Plutarch^[17]. In the great body of his philosophical and general works which usually go under the name of Moralia, works which, though they have not exercised on posterity the same influence as the famous Lives, are a treasury of information on the culture and religion of his time, there is a set of essays in the form of dialogues called Symposia or Table-talk. One of these bears the title 'Why are the days named after the planets reckoned in a different order from the actual order?' By an unlucky chance only the title of this particular dialogue has survived, but this title shews the subject unmistakably. Still more important are some facts revealed by the discoveries at Pompeii^[18]. It is hardly necessary to point out that testimony from Pompeii has the merit that one limit of its date is fixed beyond dispute. It may be much earlier than the destruction of the city in A.D. 79 but it cannot be later. The clearest item is a wall-inscription in Greek headed 'Days of the Gods' followed by a list of the Seven in week-order beginning with Saturn and ending with Venus. Another wall-inscription (this time in Latin) has no heading and Mercury's name is omitted^[19], but the remaining six are in week-order, while two others mention the 'Sun's day' and the 'Moon's day.'^[20] There is also a fresco in which appear figures clearly representing the seven deities in week-order.

From this point the evidence becomes somewhat uncertain. One passage, which is often pressed into service, comes from the fragmentary Satyricon of Petronius, which may be dated about A.D. 60^[21]. By far the best known episode in this earliest of novels is the story of the evening spent in feasting at the house of the rich but vulgar parvenu Trimalchio in Naples. On one of Trimalchio's doorposts was fixed a tablet 'shewing the courses of the Moon and paintings of the Seven Stars and different marks to shew what days were lucky and what unlucky.' Here the 'Seven Stars' are clearly the seven planets, but it is not stated that they were in week-order. The only suggestion of a week comes in the words 'what days were lucky or unlucky,' and this may quite well refer to the days of the month. Even if we insist on connecting them with the 'Seven Stars,' they may be astrological in the proper sense and refer to the various combinations or positions of the planets on particular days. We know that calendars of this kind existed and were used even by astrologically-minded doctors in dieting their patients. And if it be objected that it would be difficult to represent these in a single tablet on Trimalchio's door, we have to remember the licence of the novelist who did not have to construct the tablet in actual fact. If the words are taken, as is no doubt quite possible, of the week, we may see in them a confirmation of the view which on other grounds seems probable that it was among the vulgar rather than the educated that the week originally obtained its currency.

Another doubtful item belongs to a considerably earlier date. This is the fragment of a calendar which experts consider to belong to a few years before or after our era^[22]. I have already stated that Roman calendars marked the periodical recurrence of the nundinae or market days on every eighth day by the letters a to h. This calendar is an exception or rather a partial exception, for the days are marked from a to h in one column, but a to g in another. Here undoubtedly we have a seven-day week and it may quite possibly be the planetary week, though no planetary names are attached. But there are other possibilities. I do not think we can altogether rule out the suggestion that since this fragment was found in Sabine territory, the second column marks a local variety of the 'nundinatio' by which these market-days were held every seventh instead of every eighth day^[23]. That local needs might make such a variation desirable is prima facie possible enough. Again it is possible that it was intended for Jews. For the presence and influence of the Jews in Italy in the time of Augustus is certainly quite as marked as the influence of the astrologers^[24].

Finally, some few years probably before this we have the passage from Tibullus already mentioned, 'I alleged that I held the day of Saturn sacred.' That Tibullus is primarily alluding to the Sabbath seems to me, as I have said, almost certain. But does he in calling it 'Saturn's day' shew a knowledge of the planetary week? It seems to me the most natural conclusion that he does, but I do not think it is absolutely certain. The belief that the Sabbath was a Jewish form of reverencing the planet may have originated independently of the planetary week and indeed be anterior to it. That Saturn was an unlucky planet is a doctrine independent of the week and we may safely suppose that before the week came into use there were what we may call 'real' Saturn's days when the planet was actually in a position which the astrologers reckoned as unfavourable to enterprise. Then the Gentile world observing the Jewish abstention from work and misreading the day of rest and rejoicing as a day of superstitious dread of activity may have evolved the theory of its connexion with Saturn.

At any rate with Tibullus our knowledge of the planetary week comes to an end^[25]. Nothing that can with any probability be construed as involving or alluding to it has been discovered at any earlier date. That at the time when we first hear of it, whether we fix that time by the discoveries at Pompeii or the words of Tibullus, it was a comparative innovation in the Graeco-Roman world may be safely inferred from what we know of the literature and history of that world. Whether it had long existed elsewhere is, as I have said before, a matter of mere speculation, though a few remarks suggesting the contrary will be offered later. Meanwhile, reviewing the evidence discussed above, we see that the planetary week was known in some sense in the Empire as early as the destruction of Pompeii and most people will think a century earlier. But it should not be inferred that it was then known in the same sense as in Dion's time. The evidence, as we trace it backwards, tapers downward not only in quantity, but in quality. Justin is perhaps the earliest writer who assumes that the casual reader will understand the week without explanation: Plutarch was a researcher into out-of-the-way cults and might quite well discuss a subject which was only known to the astrologically-minded. Even the remains of Pompeii hardly shew popular usage. People who cherish beliefs which are not held or understood by their neighbours may embody them in inscriptions or frescoes for their own pleasure or edification. The evidence, in fact, if it does not actually support, is quite compatible with the natural view that knowledge of the week spread gradually through the Empire till it reached the universality expressed in Dion's words. We can easily imagine various stages in this from our own experience. Fifty years ago, I suppose, the feast of Corpus Christi was known only to Roman Catholics. To-day a great many Anglo-Catholics observe it and consequently it is known to a wider circle, who, however, would for the most part be puzzled to say how and when in the year it is fixed. To take a stage above this, probably most Scottish Presbyterians know that there is such a day as Ascension Day, but I should be surprised if the majority could locate it, or even know that it always comes on a Thursday. If the circle which observes either of these days should through some wave of feeling widen greatly, if for instance schools had to be closed, because so many of the pupils were required to treat them as days of religious observance, the general public would soon come to know their incidence, as well as it now knows Christmas or Easter Monday. Some such process as this we may suppose to have gone on in the Empire in the first two centuries under the stress of an increasing volume of popular belief and feeling.

What the nature of this belief was we shall have to enquire later. Meanwhile we may note that the most remarkable fact about the evidence we have reviewed is its meagre and casual nature. The casualness is in itself a testimony to its genuineness. It is not of a kind which could be invented or imagined. But at the same time it suggests strongly that the movement to week-observance was in a sense sub-conscious, that it was a movement of the masses and not of the educated. In this respect it stands in contrast to Judaism and official astrology, both of which had their adherents in high places and were matters of interest to the classes, from which our extant literature springs.

1. V. infra, p. 58
2. Cic. De Divinatione, II, 91. In another treatise, the De Natura Deorum, II, 53, he makes Venus and Mercury change places.
3. Most of the examples given in the next few pages will be found in an article by Schürer in the Zeitschrift für die Neutestamentliche Wissenschaft, 1905, entitled 'Die siebentägige Woche im Gebraucheder Christlichen Kirche der ersten Jahrhunderte.' The title is misleading in a sense, for the article contains a vast amount of facts and references bearing on the week in general. In fact its only inadequacy is that Schürer was not acquainted with the astrological documents edited by Kroll. I should add that I have accepted Schürer's statements with regard to coins, wall paintings and the like. All literary and inscriptional references have been verified.
4. XXXVII, 18.
5. III, 41.
6. Corpus Glossariorum Latinorum, ed. Goetz, III, p. 58. But I think this testimony must be received with some caution. I am not sure that the date which is given as 'the consulship of Maximus and Aper' applies to the whole of the contents.
7. V. Schürer, p. 33.
8. Ib.
9. Apologeticum, 16; Ad nationes, I, 13.
10. The Eastern Church seems to have been much more successful. The names of the days both in Modern Greek and Russian are not planetary; v. Appendix, pp. 117 ff.
11. Portugal is an exception; v. App. p. 118.
12. Stromateis, VII, 12 (p. 877, Potter).
13. Ap. I, 67.
14. Scriptores Historiae Augustae, XVIII, 25. The text is in part corrupt, but there can be little doubt about the general meaning.
15. Vide Schürer, p. 23.
16. Among the monumental and inscriptional evidence there is probably a certain amount or perhaps much which may belong to this century or even to the preceding one. The only literary point known to me and not included above is to be found in one Ampelius, whose Liber Memorialis, though it does not mention the week, in describing the planets puts them in week-order. There are strong reasons for placing Ampelius in the second century, but this early date is not universally accepted.

    I should perhaps also mention an 'oracle' adduced by Porphyry the celebrated Neoplatonist and quoted from him by Eusebius (Praeparatio Evangelica, V, 14, 1). This oracle, which is in a fragmentary form and as it stands partly corrupt, probably named the seven planet-gods, though not in week-order, and bade the worshipper invoke them. It certainly (and this is the point which concerns us) bade him invoke the Sun on the Sun's day and the Moon on the Moon's day. Porphyry wrote somewhere in the second half of the third century, but it is a fair supposition that the oracle dates from a considerably earlier time. It may probably therefore belong to the second century and may quite conceivably be as early as any evidence of the planetary week. But the matter is too uncertain for any stress to be laid upon it.

17. Sympos. IV, 7. The greater part of Plutarch's life belongs to the first century, but possibly this work may date from the earliest years of the second.
18. V. Schllrer, pp. 27, 28.
19. Have we in these two cases attempts of schoolboys to memorize the days of the week, after the manner of Dositheus' pupils, but in the second with less success?
20. One of these is definitely dated as A.D. 60.
21. Sat. 30.
22. V. Schürer, p. 26.
23. This view is adopted by Marquardt, loc. cit. (v. p. 3).
24. Another probable employment of the seven-day week for calendar purposes, in the first century A.D. (this time with almost certainly the planetary names in week-order), appears in a fragment, which seems intended to mark the market-days in various towns of central Italy. A representation and attempted restoration is given in Corp. Lat. Inscript. I, p. 218, but I cannot pretend to understand it fully.
25. A passage in Horace (Sat. II, 3, 291) sometimes quoted as evidence of week-observance would be about the same date as Tibullus. I regard this interpretation of the line as very forced and unnatural, but as it has been accepted by respectable authorities, I have discussed it in Appendix, pp. 124 f.