Of all the great monuments of antiquity, none except the Great Pyramid has aroused more speculation than Stonehenge; and it is still, because of its unique character, one of the major puzzles of prehistoric Europe. However, as a result of excavations in 1919-26, and again in 1950-64, some things can be said with confidence, if not with certainty, about who built it, and how, and when.

There seem to have been three successive Stonehenges, built over a period of about four centuries at the end of the Late Stone Age and the beginning of the Bronze Age. Stonehenge I consisted of a circular bank and ditch, enclosing a ring of fifty-six pits named after their seventeenth-century discoverer, John Aubrey. In and beyond its entrance, to the north-east, there were outlying features of stone and timber, including the so-called Heel Stone. It appears to have been built by the local prehistoric population, some time between 2000 and 1700 B.C.; and other examples of embanked circles containing rings of pits, of the same date, have been found elsewhere in Britain. Stonehenge II was apparently built not long before 1600 B.C. by the Beaker people, who had colonized parts of Britain from Europe during the past few centuries, bringing with them the first knowledge of metals, in pursuit of which they established trade routes from Ireland through Wales to the Stonehenge area. Their monument consisted of a double circle of “bluestones” brought from Wales, never finished before it was dismantled, with an entrance pointing to the midsummer sunrise. Beyond the entrance of the earlier earthwork, now widened, this new axis was marked by a pair of stones and by the first stretch of an embanked earthen Avenue, or processional way, leading to the river Avon. In this or the next period the four so-called Station Stones were erected, roughly on the line of the Aubrey circle and symmetrical to the midsummer sunrise axis. Three outlying holes on the east side (F,G,H) may possibly be connected with them; but their excavators believed that they were of natural origin, and due to the former growth of trees.

Stonehenge III replaced the unfinished bluestone circle about 1600 B.C. It consisted of a circle of thirty large pillars of local sandstone, brought from about twenty miles away, enclosing a horseshoe of five pairs of even larger pillars, all originally capped by lintels of the same stone. Though doubtless built by the local population, now growing rich and powerful as traders on a European scale in tin and gold and bronze, the design and details of Stonehenge III strongly suggest influence from the contemporary civilizations of Mycenean Greece and Minoan Crete.

OF THE METHODS used to dig the ditch and pits, to haul and shape and erect the massive pillars, and to lift the lintels into place, we know a little and can reasonably infer rather more. In the last few years experiment with primitive tools and methods has given good estimates of the time and labor required for many of the operations involved.

The one question which archaeology cannot hope to answer about Stonehenge is: What was it for? In the absence of contemporary written records, this is and must remain a matter for conjecture, not for proof. The most common explanation, based on the midsummer sunrise orientation, is that it was a temple for sun-worship.

Professor Hawkins attempts to show in this book that it was more than this. It was, he claims “an observatory; the impartial mathematics of probability and the celestial sphere are on my side.” Unfortunately he has failed to substantiate this confident assertion, and has allowed his undoubted enthusiasm to cloud his scientific judgment.

For the more serious and informed reader he has prejudiced his case from the start, by adopting a frankly journalistic and distastefully overconfident style and approach. This is a mistake. Stonehenge, even for the layman. has its own intrinsic appeal and interest. It can do without the meretricious persuasions of literary giftwrapping.

This is a minor criticism. Far more serious are the many avoidable errors of fact, and the instances of special pleading and partiality, which have no place in an account of scholarly or scientific research, however popular. It is suggested, for instance, that the graves of the Beaker people contain swords and spears (p. 36); that the Aubrey Holes could have filled up by natural processes (p. 46); that the axis of Stonehenge II points to the Heel Stone (p. 48); that the Avenue follows the contours of altitude (p. 50); and that stone 23 has fallen and has been set upright twice (p. 113). All these statements are false. Again, Hawkins rejects the excavators’ own opinion that holes F, G, and H (which play a significant part in his theories) were caused by the growth of trees, on the ludicrous grounds that “there is no record of trees having grown there” (p. 156). What record does he expect, from prehistory, apart from the physical traces in the ground? He also attempts to excuse some of his least convincing alignments in Stonehenge III (p. 113) by suggesting that stones re-erected in 1958 were incorrectly positioned. He is evidently unaware that they had left accurate moulds of their bases in the chalk, into which they could be, and were, re-inserted.


THE FIRST FOUR CHAPTERS of the book outline, with variable accuracy, the legendary and archaeological background of Stonehenge. The fifth, on the neighboring prehistoric monuments, is in large part a direct paraphrase (though with errors) of the British Ministry of Works guide, Stonehenge and Avebury, from which illustrations have been reproduced without acknowledgment. Neither this, nor the official guide to Stonehenge itself, is cited in the bibliography.

The rest of the book concerns Stonehenge as an “observatory.” What Hawkins has done is to calculate with a computer, from small-scale plans unsuitable for the purpose, the directions of sight-lines between pairs of stones and other features. He has then compared these with the directions of significant risings and settings of the sun and moon (stars are apparently not involved) as they would have been seen in 1500 B.C. He assumes, arbitrarily, that coincidence within two degrees either way constitutes an acceptable correspondence between observation and hypothesis. Experiment with primitive methods suggests that the realistic limit of error would be twenty-four times smaller than this.

On his own conditions, Hawkins claims to have found twenty-four significant alignments for Stonehenge I (actually periods I and II) and eight for Stonehenge III, although five of this total fail to satisfy his own criteria for significance. If these claims were acceptable, they would indeed imply that Stonehenge was a prehistoric observatory; but unfortunately they are not. In spite of his appeal to “the impartial mathematics of probability,” the author has failed to show that these results are due to any factor other than chance, as any competent statistician will confirm. A critical examination shows that no more than six of the alignments claimed may be significant; and these must be tested by measurement on the site, not from plans, before they can be fully assessed.

The second main contention is that the fifty-six Aubrey Holes were used to predict movements of the moon and eclipses, for which Hawkins believes that he has discovered a fifty-six-year cycle not previously recorded by astronomers. This is an ingenious theory deserving of serious consideration, even if there are serious objections to it on practical and archaeological grounds. That the holes could have been used to keep track of such a cycle is clearly possible; whether they were actually so used is, and must remain, a matter of opinion. For an illiterate and uncivilized community (we are dealing here with the builders of the earliest Stonehenge), which has left us no evidence for a system of numerical notation, there are obvious difficulties in recording with sufficient accuracy observations which can lead to the recognition of a fifty-six-year cycle only after the passage of several generations. Furthermore, if one wants to establish a set of permanent tally-marks, to dig and then immediately to re-fill a series of holes in the ground is not perhaps the best way of going about it. The evidence from excavations leaves no doubt that the Aubrey Holes were refilled deliberately very soon after they were dug.

IT IS ALL THE MORE unfortunate that Hawkins has spoiled his case by exaggerated claims and by ignoring the possibility that his results could be due to chance; for among his suggestions there are some which could contribute materially to our understanding of Stonehenge, if they are confirmed. He argues, for instance, that whenever an observer at the center of the circle sees the full moon rise over the Heel Stone near the time of the winter solstice, an eclipse of the sun or moon inevitably follows. This is by far the best explanation of this stone which has so far been put forward. He confirms, too, a proposal made independently by an English researcher, Mr. C. A. Newham, that the latitude of Stonehenge was chosen so that the extreme northerly and southerly risings and settings of the sun and moon, marked by the sides of the Station Stone rectangle, were at right angles one to another.

That the layout of Stonehenge has some astronomical significance is not in doubt, for the orientation of the axis of the stones upon the midsummer sunrise has been recognized since the early part of the eighteenth century. The interpretation of that alignment, however, and of any other astronomical characteristics of the monument, requires the exercise of scholarly caution. Let us remember that although the majority of Christian churches are aligned accurately on the equinoctial sunrises and sunsets, they are not for that reason regarded either as observatories or as temples of sun-worship.


To test the more acceptable of the theories put forward in this book, we need data measured on the site itself, and not from plans; we need a realistic estimate of the accuracy of which the builders of Stonehenge were capable; and we need the impartial application to this material of reliable measures of statistical probability. If Professor Hawkins will give us all this, he will do much to redeem the shortcomings of this unsuccessful attempt to apply the methods of science to one of the major problems of archaeology.

This Issue

June 23, 1966