Philip J. Stooke

Note: this article was originally intended to be published in "Astronomy and Space", a publication of Astronomy Ireland, but funding problems caused the magazine to cease publication. A more complete account of this research, but lacking the section on passage geometry, is to be found in:

Stooke, P.J. "Neolithic Lunar Maps at Knowth and Baltinglass, Ireland". Journal for the History of Astronomy, XXV: 39-55, 1994.

On March 17, 1980, Martin Brennan and Jack Roberts saw a beam of light from the rising sun illuminate a carved stone at the back end of the rock-lined passage in the great mound called Cairn T at Loughcrew, County Meath. Two weeks later, on the evening of the first of April, Brennan and his colleagues watched the rising moon from the same spot. As the moon appeared over the horizon a shaft of light was projected along the passage and onto the same carved stone.

The great passage mound of Knowth, also in Meath, contains two rock-lined passages, one facing east, the other west. Brennan observed the setting sun shine into the western passage on September 13, 1980. Attempting to observe sunrise on the following day, he found that the view of the rising sun from the eastern passage would be blocked by trees and the current level of the ground. Nevertheless, it seems likely that the eastern passage was also originally intended to be penetrated by light from the rising sun and moon. Although solar alignments have been described at many neolithic sites in Western Europe, less attention has been paid to lunar alignments, despite the fact that at certain times the moon can rise or set at any location on the horizon which may be occupied by the sun.

Brennan's experience, described in his 1983 book The Stars and the Stones, suggested that these Irish neolithic sites might have had some connection with the moon as well as the sun. I have identified another connection, previously overlooked by Brennan and others working at Knowth. If moonlight were to shine on the back stone of the eastern passage at Knowth, it would illuminate a map of the moon itself, the world's oldest known depiction of the lunar maria.

The carved stone which forms the end wall of the eastern passage was called Orthostat 47 by George Eogan, who excavated Knowth in the 1960s (Figure 1). The design has three sections, superficially similar but oriented differently. The right-hand section appears to be nothing less than a map of the lunar maria, as becomes clear when it is compared with a naked-eye map of the moon. At least a dozen points of correspondence are immediately obvious (Figure 2).

The remaining two sections of the carving are simpler but crudely similar to the first, sharing the overall arc shape of the maria surrounding the lunar central highlands as well as an isolated spot representing Mare Crisium. Why were they carved in different orientations? I believe they depict the apparent rotation of the maria on the disk of the full moon as it crosses the sky in the course of a night. Watch the full moon one night. The arc of maria opens to the right (like a letter C) as the moon rises, opens downwards (to a northern hemisphere observer) as the moon crosses the meridian, and opens to the left at moonset (Figure 3). The disk appears to rotate like a wheel, an illusion caused by our motion on a rotating Earth. We compare the moon with our apparently fixed horizon, but the plane of the horizon actually rotates with the Earth to trace out a cone in space.

The right-hand carving on Orthostat 47 depicts the maria as seen on a full moon a little after midnight. At upper left the maria are shown soon after moonrise. It would be reasonable to assume that the lunar disk continues to rotate beneath the horizon, and the carving at lower left appears to show the maria 'upside down' as if under the horizon. The only thing missing is a clear outline of the circular limb of the lunar disk. Perhaps it was originally chalked or painted on the rock.

Here on one stone are two important lunar observations, representations of both the maria themselves and their apparent rotation as the moon crosses the sky. Further examples of both are found at Knowth. One of the most diagnostic is Kerbstone 52 (as designated by Eogan. Brennan called it SW22 or the 'calendar stone'). On it a group of crescent shapes are arranged around the large boulder (Figure 4). Seven of the crescents at the top of the stone are replaced by circles. Brennan interpreted the crescents and circles as phases of the moon, though the characteristic first and last quarters are not convincingly portrayed. However, the crescents are not oriented as an actual lunar crescent would be in any obvious sequence of observations. Rather, their positions match the orientation of the maria as the full moon crosses the sky. Why do they turn to circles at the top of the stone? I attribute this to the excessive contrast between the brilliant full moon and the black night sky, which renders the maria harder to see than at dawn and dusk.

Near the base of Kerbstone 52 the crescents dip below a wavy line (a schematic horizon of waves or hills?) and pass horizontally back to the starting point to complete the cycle. Brennan interpreted the cycle as a month, whereas I see it as the daily motion of the moon. There are 29 symbols, the number of whole days in a lunation, but that number, obviously associated with the moon, does not by itself prove that the design is a depiction of a month. The symbols cross a spiral which may represent the moon's path through the heavens at a different elevation each month during the year. Here again we see the maria represented by an arc, whose changing orientation corresponds to the apparent rotation of the lunar disk during a night. This general pattern, in various forms and less fully developed, is repeated elsewhere at Knowth and at Loughcrew, Newgrange, Dowth and perhaps at Gavrinis in France.

The overall pattern of lunar maria from Humorum and Nubium past Imbrium to Nectaris and Fecunditatis may be seen as one broad arc. Alternatively, the eastern maria (Crisium, Nectaris and Fecunditatis) suggest a three-fold subdivision of the arc. This is extended westwards by Mare Frigoris, and the three-fold division is again suggested (a little less obviously) by south-western Oceanus Procellarum, Mare Humorum and Mare Nubium. Thus, various combinations of arcs might be interpreted as lunar maps. Several of the most complex examples are shown in Figure 5.

One of the most interesting of these additional maps is that carved on a stone basin which Eogan discovered in the right-hand recess of Knowth's eastern passage (Figure 5a). This shows a set of concentric circles (Mare Imbrium and its surrounding highlands), multiple lobes to the left and right (Oceanus Procellarum and the western maria respectively) and a small isolated circle near the top (Mare Crisium or possibly the eastern end of Mare Frigoris).

All these carvings, but especially the map on the stone basin and the carvings on Orthostat 47, which might have been illuminated from time to time by the moon itself, suggest that the markings on the moon - as well as the moon itself - were important to the builders of Knowth. God, home of the gods or of departed spirits? We will never know, but we can be certain that the builders of Knowth were keen observers of the sky. Without using sophisticated mathematics or fine angle measurements (as suggested by Alexander Thom and others), these neolithic astronomers were able to observe, depict and hypothesize about the moon and its motions. These are the actions of scientists.

The carvings are about 4800 years old. The next oldest depiction of the maria known to science is that by Leonardo da Vinci in about 1505 AD. Other ancient lunar maps may lie unrecognised among neolithic (and later) artifacts. For instance, I note that the plan of Stonehenge is itself very like the shapes I have described - a circle containing a horseshoe - and so might be considered a very simple map of the moon. Its famous solar alignment happens every year, but at certain times the full moon would also rise over the Heel Stone, and would illuminate a map of itself. This, of course, is even more speculative than the Knowth alignment, and it is also somewhat more recent than Knowth. I expand on this point below.

For now the products of the artists - and scientists - of Knowth are by far the oldest apparent representations of lunar markings and motions known to us. Their significance to the history of science is considerable. Knowth is one of the most important ancient scientific sites in the world.

Lastly, I will raise a still more speculative matter. The historian Diodorus Siculus (writing shortly before the time of Christ but quoting Hecataeus, some five centuries earlier, 'and certain others') wrote of the 'Island of the Hyperboreans':

... there is also on the island both a magnificent sacred precinct of Apollo and a notable temple which is adorned with many votive offerings and is spherical in shape... They say also that the moon, as viewed from this island, appears to be but a little distance from the Earth and to have upon it prominences, like those of the Earth, which are visible to the eye. The account is also given that the god visits the island every nineteen years....

The nineteen year period is presumably that between almost identical eclipses. The spherical temple has usually been assumed to refer to the circular Stonehenge, but might equally well refer to Knowth or Newgrange, which being prominent mounds at least resemble hemispheres. Newgrange appears to have been partly covered with white quartz fragments, giving the appearance of an egg (according to Brennan) or of the shining moon itself. Quartz is also found at Knowth and Dowth. Its original distribution at each site is uncertain. Were the mounds themselves decorated to look like the moon? This deserves further study. The 'prominences' referred to are presumably the lunar maria visible to the eye, and the reference to them at least indicate a local interest in those markings which is more in evidence at Knowth and Newgrange than at Stonehenge. The text may even be a confused reference to the very carvings which are the subject of this article.

Figure Captions

Figure 1. Carvings on Orthostat 47 at Knowth, the vertical stone at the end of the eastern passage.

Figure 2. A naked eye map of the moon (left), a carving from Knowth Orthostat 47 (right) and the two superimposed (centre) to illustrate their similarity.

Figure 3. The full moon crossing the sky during the course of a day. As it moves, it appears to rotate like a wheel. The effect is an illusion caused by the motion of the observer on the rotating Earth.

Figure 4. Kerbstone 52, the 'calendar stone', showing arcs in the orientation of the lunar maria (not the crescent phase) crossing the sky.

Figure 5. Other possible lunar maps. (a) Carving on the stone basin found in the eastern passage at Knowth. (b) Knowth Kerbstone 97. (c) Knowth Kerbstone 68. (d) Stone G, Baltinglass, Co. Wicklow.

Let me share with you a few ideas which I have not yet published. First, a striking carved stone basin was found inside Knowth. I learned recently of an ancient Irish tradition of placing bowls of water at doorways in the spring, to reflect sunlight into the house, presumably for good fortune in the coming year. I can't say how widespread this was, of course, or if anybody still does it. The eastern passage of Knowth slopes down a bit towards its entrance, so it might be difficult to get sunlight or moonlight up onto the back wall. Maybe the structure settled a bit after construction, and the bowl was filled with water to reflect the rays of the rising sun or moon up the passage after rising. That bowl was found in the eastern passage, and it also has a carving resembling lunar markings on its outer surface, or so I interpret it (a small outer circle could be Mare Crisium, Figure 5a). Could the later tradition be a memory of that ancient ritual?

Newgrange is well know for its mass of white stones, found all over the site and now reconstructed as a wall around the hill. Some white stones were also found at Knowth and Dowth. The original distribution is not really known, and the idea they formed a wall is pure speculation. I have a different idea, also mere speculation, but it is that the white stones covered the 'dome' of Newgrange (and maybe the other mounds) either completely, to resemble the lunar disk, or in a pattern which made - yes - a map of the Moon! - white stones plus grassy areas arranged in the shape of the lunar maria. These mounds may have been the first lunar globes. We'll never know, though records of the distribution of white stones before excavation began might help.

Also I believe that the later structure of Stonehenge is also a map of the Moon. Astronomical alignments are frequently invoked for Stonehenge, but there is absolutely no reason for the alignments to require the shape of Stonehenge. That shape has some other purpose. It consists of a circle containing a horseshoe shape, like a full moon. It is not built on the ground, but raised up into the sky on tall pillars, as if its builders were saying to the Moon - "We worship you - here is your likeness" - and the full moon could look down on that offering, a picture of itself. Stonehenge faces the rising midsummer sun, but on rare occasions the full moon would also rise at the same place. On very rare occasions the sun would rise on the axis of Stonehenge, and later the same day the rising full moon would rise at the same place. On those special occasions the geometry is right for an eclipse of either the sun or the Moon.