The Oxford Handbook of Neolithic Europe

by Chris Fowler

  We must then ask whether our data will allow us to establish the motivation underlying the custom. The royal tombs of Mycenae simply faced downhill. Islamic mosques—in theory at least—face towards the terrestrial ‘target’ of Mecca. In County Sligo, Ireland, the graves face towards the great cairn on top of Knocknarea. So we do not lack for examples of builders who followed a custom whose motivation had nothing to do with astronomy.

  How then can we satisfy ourselves that the custom called for an orientation towards a heavenly body? If we are dealing with passage graves confined to a limited geographical region, we may have suspicions but no proof—tombs facing in a given direction might be oriented on the rising of the Pleiades, but they might also be oriented on a distant sacred mountain. But if the custom is spread over a wide area, we can be confident that only by using the sky could the builders have articulated the custom.

  If the custom was indeed related to what is to be seen in the sky, what possibilities should the archaeoastronomer consider? We must disregard the planets, for their rising and setting points follow patterns so complex that if a group of passage graves were directed towards a planet, we would have almost no hope of discovering this. But what of the sun, moon and stars? The range of sunrise today extends a certain number of degrees north and south of due east, and that of sunset extends similarly north and south of due west (the extent of the range depending on the latitude). In the Neolithic the ranges were greater, but only marginally so, and for most purposes the change over the millennia can be ignored (hence the midwinter sunrise still penetrates Newgrange). The range of moonrise is (and was) similar, but extended a few degrees further in each direction, so that there can be graves that sometimes faced moonrise but never faced sunrise.

  If a passage grave faced southerly, it would never have faced the rising or setting of the sun or moon, but it would have faced these bodies each day when they were in the sky. But if a passage grave faced northerly—very roughly, between north-east and north-west, directions in which the sun and moon are never seen—and its target was astronomical, this target could only be a star.

  This however would be very difficult for the investigator to establish with confidence, because there are innumerable stars, and one or more will be seen rising or setting in almost any given direction. But even if we restrict ourselves to the very brightest stars, we have the further problem that the rising and setting points of any particular star change significantly over the centuries. This is because the gravitational pulls of the sun and moon on the earth (which is not a perfect sphere) cause the axis of the earth to wobble. As a result the position of the North Pole in the sky is constantly changing, and so the stars visible from any particular location on earth change too. These changes, which are fully understood, have important consequences for the archaeoastronomer. For example, the present writer has argued that the Bronze Age taula sanctuaries of Menorca were designed to face the Southern Cross (Hoskin 2001, ch. 4). Astronomy tells us that the Cross was visible when the sanctuaries were built, a millennium or so before Christ, although it has long since disappeared from sight of the islanders.

  The changes in the rising and setting of the stars pose a fundamental problem for the archaeoastronomer who suspects a passage grave was built to face in the direction of (say) the rising of a bright star; for by suitable choice of the supposed date of construction, it may well be possible to make the coincidence between the orientation and the rising position of the star exact. With so many stars to choose from, and so little known about the exact century of construction of a passage grave, it is usually so easy to ‘orientate’ a monument on a star that the exercise is meaningless.

  Fieldwork reveals that in western Europe it is in fact almost unknown for passage graves to face northerly, outside the range of moonrise (and of course sunrise). It is therefore most unusual for a star to be proposed as a convincing target for a dolmen’s orientation. But there is one plausible example. At Valencina de la Concepción, to the west of Seville, the remarkable Dolmen de Matarrubilla has a perfectly straight and uninterrupted corridor an astonishing 37m in length, giving a glimpse of sky (but no more) to the prehistoric observer in the chamber; and it is difficult to believe that the orientation of a corridor so long and so straight was of no interest to the builders. Most exceptionally, the orientation is 17o48′, far to the north of places on the horizon where the sun, moon and planets rose. If therefore it has an astronomical target, it must be a star, and the only bright star that rose anywhere in that direction was Arcturus, the third brightest in the sky as seen by the builders. Because of the exceptional accuracy with which the orientation is known, we can calculate that Arcturus was visible from the chamber only in the years immediately before 3100 BC, assuming of course the tomb existed at that time. Archaeologists confirm that the thirty-second century BC is very possibly the period for the construction of Matarrubilla; but unfortunately they cannot be more precise. In this most exceptional case, then, archaeoastronomy can confirm that if the orientation of the tomb was astronomically motivated, it was built to face the rising of Arcturus in the thirty-second century BC (Hoskin 2001, 77–80).

  This brings us to the sun and the moon, both of whose cycles must have been of the greatest significance in daily life. The seasonal (solar) calendar was fundamental to the growing of crops, and for people with limited artificial light the brilliant illumination from the full moon at night must have contrasted strikingly with the darkness at the opposite end of the monthly cycle. The role that the sun and moon played in religious practice must surely have been very great, and to the archaeoastronomer each comes with good credentials. We have seen that churches are commonly oriented on sunrise, and Strabo tells us of tribes in the north of Spain who in Roman times danced all night in front of their homes to celebrate each full moon.

  The range of sunrise is virtually the same from one year to the next. The behaviour of the moon, however, is much more complex. The range over which the moon rises each month varies over a cycle that lasts a little over 18 years. If at a given time the range is at a maximum, extending from some five degrees north of midsummer sunrise to some five degrees south of midwinter sunrise, then nine years later the range is at a minimum, with its limits now five degrees short of the sunrise limits at either extreme. Archaeoastronomers term the maximum limits the ‘major lunar standstills’ (or the ‘major lunastices’), and the minimum limits the ‘minor lunar standstills’ (or the ‘minor lunastices’).

  At a major lunar standstill, the moon rises well outside the range of sunrise, and it is to be seen rising in such directions only at intervals of a generation or so. One can well imagine that moonrise at such almost unprecedented positions on the horizon would have seemed remarkable. At a minor lunar standstill, however, the moon rises at positions where it rises every month, the only singular feature being that it (so to speak) refrains from rising outside this minimal range. This can strike the modern mind, trained in geometry, as worthy of note, but whether Neolithic people found it remarkable is open to doubt.

  The sun is always a luminous disc, but the moon of course varies from the disc of the full moon to a tiny crescent (and even invisibility) around the time of new moon. We have seen that a passage grave facing sunrise necessarily faces moonrise, and for those archaeoastronomers (including the present writer) who favour a solar explanation for orientations, the situation is simple. Neolithic grave builders would have been well aware of the range of sunrise, and especially of the position of midwinter sunrise, for the days were then getting shorter and shorter, and colder and colder, and unless the sun reversed its course death would inevitably follow for humankind. The archaeoastronomers who favour a lunar explanation, on the other hand, have both major and minor standstills at their disposal but, more importantly, have a choice of phases of the moon—most plausibly full moon—and may focus on the full moon that occurred at a particular time of the solar year. Easter, for example, is related to the full moon (a lunar event) following the sp
ring equinox (a solar event). The opinions of respected archaeoastronomers are therefore divided between those who favour a simple solar explanation, and those who employ one of the more complex and more flexible lunar explanations. At present the debate continues and the outcome is unclear; but as the solar explanation is so much simpler to express, we shall adopt it in what follows (Hoskin 2001, 213–216; Belmonte and González García 2010).

  MEASURING THE PASSAGE GRAVES OF WESTERN EUROPE

  Of all the passage graves of western Europe, the most instructive for our purpose are the seven-stone graves of the Alentejo region of Portugal (Fig. 48.2; Hoskin 2001, ch. 6). These tombs, which are thought to date to the third millennium BC, have a tall backstone set vertically in the ground, and this is a true orthostat. Either side of the backstone is another stone of similar dimensions, but this is not an orthostat; instead, it leans on the backstone. Next to each of these sidestones is another similar leaning stone, and next to that yet another. In this way the chamber is octagonal and formed of seven stones, with an entrance and passage that face the backstone. The seven stones are large and sometimes enormous (the backstone of Anta Grande de Zambujeiro is no less than 5m in height), and the exceptional form of construction makes these tombs readily identifiable, and prominent in the landscape. The region in which they are found extends from the Atlantic across Portugal and into neighbouring Spain, perhaps 200km from east to west and a similar distance from north to south—an area so vast that any custom of orientation must surely be based on the sky.

  And the custom of orientation is absolute: every single one of the 177 graves the writer has measured faced within the range of sunrise (and therefore that of moonrise also). That this should happen by chance is of course inconceivable, for the range of sunrise occupies only one-sixth of the horizon. Now suppose the builders thought it necessary to guarantee that their grave would indeed face sunrise (on two days each year). The simplest way to ensure this would be to align the grave to sunrise on the day building started—as we know happened with the alignments of Christian churches. But the graves are often enormous and these would have required months to construct, and the cultivation of food, on which the survival of the constructors depended, must have taken priority; so it is likely that the Neolithic people of the Alentejo would have felt free to turn to building these graves only when the harvest was safely in. Therefore, if the graves were intended to face sunrise from time to time, and if this was achieved by aligning the tomb to sunrise on the day building started, then we would expect to find that most tombs faced sunrise in the autumn and early winter (but not in late spring and summer, the height of the agricultural season). And this is exactly what we do find (Fig. 48.3). This strongly suggests that these graves were indeed built to face sunrise, and some archaeologists see them as the most persuasive example in western Europe of a sunrise custom of orientation, and therefore of a role of astronomy in the cosmovisión of the builders.

  FIG. 48.2. One of the seven-stone passage graves typical of the Portugese Alentejo: Anta de la Marquesa, which is just across the frontier into Spain, near Valencia de Alcántara. The stone centre right is the backstone which is, or rather was, a true orthostat. The stone centre left leans on this, and further left we see the next two sidestones, each of which leans on its predecessor.

  FIG. 48.3. Histogram showing the orientations of 177 seven-stone passage graves of the Alentejo. All the graves without exception faced sunrise at some time of the year, the majority in the late summer and autumn. Two appear from the histogram to have faced a little south of midwinter sunrise, but this is not so, as they look towards a steep cliff.

  We have mentioned the passage graves of western Europe whose orientations are ambiguous, but these are atypical and few in number, and we shall disregard them in what follows. Most passage graves have a well-defined orientation. The majority are megalithic, built of a small number of large stones, but some are of tholos (‘false-cupola’) construction and made of a large number of small stones. Many of the megalithic graves are modest in size, but others are gigantic and almost defy belief, their construction involving a workforce numbering in the hundreds. By contrast, the tholos tombs with their small stones could have been built by a team of limited size.

  In making his specific contribution to the study of passage graves, the archaeoastronomer abstracts from the form of the construction before him, and focuses his attention on measuring the grave’s orientation. Going from one grave to the next, he compiles tables of azimuths (angles measured clockwise from true north), angular horizon altitudes, and latitudes. With these terrestrial data he can readily calculate the ‘declinations’ (celestial latitudes) of the heavenly bodies whose rising or setting the graves faced. And we find that there is much to be learned from this.

  Except along the French Mediterranean coast and parts of Catalunya (to which we shall come shortly), the graves of western Europe, from the Channel Islands to Gibraltar, overwhelmingly faced the eastern horizon, and south of midsummer sunrise. Almost none faced northerly, and (with the exceptions just mentioned) almost none faced westerly. Nearly all faced directions in which the sun rose at some time of the year, or faced further south, in directions where the sun had risen and was climbing in the sky or was around culmination. I characterize the former as ‘SR’ (sunrise oriented) and the latter as ‘SC’ (sunclimbing oriented). The seven-stone tombs of the Alentejo are without exception SR. The megalithic sepulchres of Montefrío, to the west of Granada, and the tholos tombs of Los Millares in Almería, both of the early third millennium BC, are—despite their very different constructions—also SR (Fig. 48.4; Hoskin 2001, 59–62, 66). Within each of these two groups there are a tiny number of tombs with exceptional orientations—at Los Millares, for example, in the very midst of all the SR tombs, and of identical construction, there are two that face south-west, a bizarre anomaly that is obvious to any visitor to the site—but the overall pattern is beyond dispute.

  FIG. 48.4. (left) The orientations of 41 megalithic sepulchres at Montefrío, some 40km west-north-west of the city of Granada. (right) The orientations of 48 tholos and related passage graves at Los Millares, 100km from Granada in the opposite direction. In structure the two sets of graves could hardly be more different, but when we abstract from this and measure their orientations, we find that the two patterns are almost identical.

  Montefrío and Los Millares are but two sites among many, and it is in the west of Iberia that a strict SR custom prevails—so much so that there it is almost unknown for a grave not to face sunrise. Of the 334 tombs I have measured (including those of the Alentejo), no fewer than 324—an astonishing 97.0%—are SR. But as we move further from the Atlantic coast of Portugal increasing numbers of sites have tombs whose orientations are a mixture of SR and SC, and it is tempting to see the latter as a relaxation of the strict SR custom. Orientations that are neither SR nor SC are very unusual: of the 945 graves I have measured in Iberia and the adjacent Pyrenees, no fewer than 911 (96.4%) are SR or SC. Similar patterns are to be found in western France, on the French causses, and in the Channel Islands.

  PASSAGE GRAVES THAT FACE WESTERLY

  The major exception to this simple schema is a group of tombs near to the French Mediterranean coast. At Fontvieille, near Arles and close to the Rhône, are communal tombs that, instead of being on the surface, are quarried out of the bedrock (discreet access being provided by steps) and covered with slabs at ground level. These slabs are carefully shaped on the interior, as are the sides and base of the elongated rectangular chamber, but on the exterior the slabs are left in their natural state. As a result, the visitor may be standing on top of one of these tombs and be quite unaware of this; this is quite different from what we encounter with many megalithic graves, which seem to be intentionally prominent in the landscape, no doubt warning others that ‘we have been in this land since time immemorial’. Still more remarkably, instead of facing easterly, the Fontvieille tombs faced westerly, within the range of sunset (and
moonset) (Sauzade 2000; Hoskin 2001, 140–147).

  At one place at Fontvieille, Dolmen de Coutignargues, the builders found themselves excavating rock of such poor quality that they were unable to shape the sides of the chamber as they would wish. They therefore cut out of the bedrock a sizeable trench, and within this trench they constructed a passage grave with an elongated rectangular chamber whose sides were of drystone. The orientation of the grave is again westerly.

  To the east of Fontvieille, throughout Provence and towards the Italian frontier, we find passage graves (presumably later) that seem to imitate the salient features of this Fontvieille exemplar, although these graves are not subterranean but on the surface. Like Dolmen de Coutignargues, they have rectangular (if less elongated) chambers, and they have drystone walls, although these are often interspersed with orthostats. Their orientations are invariably westerly, but whilst many face sunset, others face the sun when it was descending to the horizon. The latter’s ‘relaxation’ of the strict sunset custom seems to mirror the SC relaxation of the SR custom that we apparently find in Iberia and elsewhere.