The following is my translation of an article called: Bautechnik beim Stonehenge, Eine Ergänzung des Aufsatzes "Astronomie der Vorzeit" von Dr H Hein. It appeared in a German popular science magazine, Kosmos Handweiser für Naturfreunde 1921, Heft 10, Seiten 253-256.
So as to annoy people, all of the fine illustrations won't be included here. The same applies for a couple of further translations on related themes: Stonehenge, from 1913 and The star-struck of yore, from 1921. Both receive mentions in the text. Also, some of the terms used in the article below proved unfamiliar to me, and these have been left in the original forms. Enlightenment would be welcome.
I'm not aware of any previous translations.
Trevor Dykes.

Construction methods of Stonehenge, an addition to the article 'Prehistoric Astronomy' by Dr H Hein.
Until recently, Stonehenge* was in private possession and, not all so long ago, opportunities to visit it were like great rarities. The state appeared to see no necessity to place this remarkable monument of the North European Stone Age under protection. A few insufficient measures were taken against the further collapse of the monument: stones were supported and one of them was even re-erected. But one can find it comprehensible that the then owner could not afford to bear the heavy costs necessary, that a longer term protection would have demanded. In the meantime, a number of important stones were brought down due to the influence of storms and weathering and, finally, the English conscience has now been awoken: generous resources flowed and, with careful supervision from prehistoric researchers, all possible stones that are still present should be re-erected and secured in the ground with concrete.
(* Compare with (Kosmos) Handweiser 1921, p.34. Kosmos 18, 1921, 10.) Modern cranes and winches will be able to complete work quickly which, previously, would probably have required many hundreds of workers over many decades. It also goes without saying that this rebuilding of Stonehenge will required the ground to be worked though and, in areas where that happens, it will be carefully examined for remains from prehistory, implements and bones. So far, researches show that this monument was built before the start of the Bronze Age. All processing was done without any metal blades.

It would not be without interest to follow the construction of this monument, as shown by the present results of research.

The very sources of the material give rise to all sorts of questions. Two sorts of stone were used: sarsen, a form of sandstone found commonly in Southern England, and blue stone, a type of rock which can no longer be found in the vicinity of the monument, but which does occur in Wales or Brittany. And that raises the question: Were a few pieces transported into the area during the Ice Age as so called erratic blocks, and they were all later used to build Stonehenge, or did the builders acquire the rock from Wales or even Brittany? The latter would have required impressively capable ships, and that during the Stone Age! The larger blocks have been worked from sarsen and the smaller from blue stone. Additionally, a few smaller columns are of Porphyr, Diabes and a particular type of sandstone.

The quarry that provided the builders of Stonehenge with sarsen is known. Careful examination has also revealed how the massive blocks (up to nearly 9m long, over 2m wide, more than 1m thick for the largest parts) were released from their surroundings: with fire and with water. A fire of wood charcoal was set along the line that was intended to crack. When the stone had been strongly heated along that line, the glowing charcoal was swept aside, water was poured in and this caused the rock to split. If, through testing, one has concluded how much water one would require to achieve this favourable effect (there must not be too much water, that would later leave the stone wet, and it must all evaporate), then an almost uninterrupted production is possible: the fire can immediately be pushed across to another position, and the slit of the channel moves ahead relatively quickly. The layering of the sandstone makes it easier to spring rock out from its bed. This springing may also have been assisted by wooden staves being driven dry into the fissure, and than being wetted.

Transport to the construction site would then have occurred on rollers.

Further preparation of the blocks took place at the construction site. The surfaces were crudely levelled or flattened for as far as they were to be above ground level, but only for the surfaces to be directed internally in the monument, and partly on the sides of pillars. The inner 17 stones also have a smoothed ceiling.

The 'hammers' with which this, by no means small, labour was performed were simple round stones: hand-sized field rocks held by both hands, and perhaps with a bit of fur to protect the hands from the effects of rebound. They were used until they had worn down and became too small. These were then thrown away and, together with the countless hammered off stone splinters, now form a fairly stable stone paving above the originally chalk surface in which the pillars stand. The preparation for the five horizontal lintels, for the double columns, was conducted particularly carefully: these had concavities worked into the inner faces so that the curved outer sides could be set together precisely as a giant horseshoe. What is also remarkable is that cones were worked from the stones. The horizontal stones have corresponding sockets, and this achieved a very close joint of the columns and lintels.

As for how this construction progressed, partial clarity has resulted from examining the ground. The columns were erected into holes dug into the lower bed of limestone. The walls of these holes are close to the stone on three sides, and those for the large blocks are always four-sided; the fourth wall is slanted. From that side, the stone could be pushed into the hole and then tipped up. The underside of the stones was always somewhat pointed, and infilling helped to hold it secure. Given the relatively crude anchorage in the ground, it must be seen as remarkable that so many stones have remained upright for 3,600 years.

As to how the great pillars were raised, that is a question that studies have not yet decided upon. It is as good as certain that the law of leverage must have come into use. An experiment should be made to attempt a method of erection requiring relatively little labour and expenditure of energy. The stones of the outer circle are, on average, 5 metres long, 1.1 thick and 1.8 wide, have a volume of about 10 cubic metres, and thus a weight of some 25 tons -even if this is only approximate. To raise one end of such a stone requires an energy of 12.5 tons = 12,500 kg, ie. 125 men each with a weight of 100kg, and that must be applied to one end. With the help of fir trunks as levers, raising one end obviously becomes a different matter. But how to proceed? When one end of the stone is raised a certain distance above the ground, the lever, which can only be used close to the ground, is of no further help. Because of this, I would like to presume that the slanting side of the hole played an important role. The hole is 1.2m deep, and the sloping surface would be some 1.7m long (1.2 x sq. foot of 2). If one were to push on rollers (Ill. 1) 1.7m beyond the top of the sloping surface, then the projecting end would hold a similarly sized piece of stone in balance (reckoning from the top edge of the slope). Should the stone now be tipped over the edge of the hole, then it would only be a matter of lifting the burden of one end. That would only be a length of 1.6m for a 5m long stone. And in accordance with the law of leverage -the calculations will not be shown here, only a bit more than 3,000kg of energy are required. Such a force of energy could, should this stone have been raised to a limited degree by a lever, be supplied by some 50 to 60 men with only poles and bare hands. Everybody can learn from experience that, given an increase in the angle of slope, less energy is accordingly required to allow for its erection.

With the very large stones of the double columns -the largest are almost 9m long- the situation was also more difficult. The weight climbs to around 50 tons. Here, I would like to assume, that one raised the ground level in front of the sloping face of the hole, and built it into a ramp. If this had a height of only 1m, then the length of the slope would be increased by about 1.4m. If the hole was some 1.5m deep (one, for example, is over 2m deep and another 1.3), then the length of the sloping face would be 2.1m and, with the ramp, that would extend to 3.5m. A stone of more than 7m length like, for example, the double columns I, II, IV, V, would then be held in balance on the edge of the ramp (Ill. 2). It could then be tipped into the hole without great difficulty and then, certainly with greater difficulty, made entirely vertical.

Another method could have worked without a ramp (Ill. 3). The rollers 1, 2, 3 stick out from below at both sides of the stone. One uses levers to raise the end of the one side of the stone up, and force in one or more beams lengthwise below (Ill.3b). One now lifts in front of and behind the rollers, and the further beams 4, 5, 6 can be laid across and so on and so forth. In this way, the stone blocks grow, so to say, upwards on a pile of wood. Once the stone is high enough above the hole, the platform must be secured by a network of ropes or hammered in posts so as to give it stability. Then the stone has to be moved with the uppermost rollers until it loses balance and tips over the edge into the hole. The pile of beams must not prematurely collapse, as a relatively loose arrangement of wood could be life-threatening during a collapse. It should still be able to support the stone so it does not topple over in the hole. Therefore, a certain measure of stability is required for the wooden platform.

The choice between both methods of erecting the stone is left to the reader.

The greatest difficulty is posed by the question: How were the horizontal lintels set onto the large columns? Here are weights of about 12 tons (in the case of the double columns) being raised to a height of 5 to 7 metres, and then being pushed to the side over the cones, and then being lowered down. There have been a few researchers who really assumed that massive, steep banks must have been constructed so that the lintel stones could be steered over the columns. In such a way, the entire object would once have been buried by soil, and must have subsequently been dug out again. But there are no traces of such enormous quantities of earth having been moved. And there is also a simpler method. The second of the erection techniques mentioned would suffice for this. Such a platform could be built around the crown of each column (Ill. 4a). The horizontal stone would lie on this scaffold behind the columns. Then, the surface of the scaffold could be exposed with alternate movements to the left and right, and alternately forwards and backwards, and the platform would be further raised with beams being pushed in. Eventually, the stone would be higher than the cones of the columns. Then the entire platform could be carefully steered forward on rollers below it until the sockets of the lintels are precisely over the cones on the columns (Ill. 4b). Next, the rollers and a few beams would be removed. The stone sinks down over the cones and, soon, it contacts the columns, and then the platform can be quickly dismantled.

In the South Sea -for example on Easter Island, on Tinian, Hawaii, Tahiti and other islands- can be found constructions of similar dimensions. No significant movements of earth have been required for their erection. In these cases too simply levers and platforms were utilized, and these also grew higher by beams being inserted. Whether the builders used round or, more probably, rectangular cut beams will be left for the reader to consider.

And now we turn to the astronomical significance. Due to the obvious numerical relationships, one may think of a use as a calendar for lunar eclipses, as was suggested previously. The summer solstice can be easily identified as, on the longest day, the Sun rises through the northeasterly openings of both circles when seen from the so called altar stone. In the preceding and following days, the Sun shines through two great stones, the "astronomical stone" and the butcher / sacrificial stone, which stand directly south of the elongated axis outside of Stonehenge.

The altar stone lies precisely across the axis (see Ill. page 35). A few researchers are convinced that, previously, it would have been standing, and explain it would have stood on the axis, but that it then came into this remarkable position with both ends lying 2m from the axis. I cannot really share this view. I prefer to assume that the stone has always lain like that, and this is because it serves as a pointer. Namely, on the afternoon of the longest day, the shadow of the lintel of the middle double columns is cast directly onto the altar stone. The length and width of this stone have often been cited but, remarkably, not the height. As this "column" is said to be 90cm wide, the stone may well reach about 75cm over the ground. It appears to be rather like that from photographs. If that is really the case, then the shadow of the lintel stone would fall precisely onto the edge of the altar stone on the longest day, but it would fall below it on all other days. The approach of the shadow would thus be a means for predicting the approach of the summer solstice. At the winter solstice, the sun sets to the southwest, in the direction of the axis.

Hopefully, valuable material will not be unconsciously destroyed be the reconstruction of Stonehenge. Such unintentional destruction of an important object of research has occurred, among other cases, due to the activities of a very well known researcher at the Odilienberg in Alsass. There was a series of stone circles there which, according to the scant description, had similarities with those at Stonehenge and Avebury -but much smaller: only stones of about a foot in length, and the whole complex was covered with a heap of sand. The entire thing was simply accounted for as being the toy of a child. The pieces were carelessly packed together and carried off to the museum. But has one ever heard of a prehistoric child's toy being produced from a particular type of sandstone? It would have been better to have carefully cleared the area of sand, and to have precisely recorded the positions of the standing and fallen stones. I do not hold it for impossible that, at a time of emergency or threat, a calendar man of prehistory could have buried hid miniature calendar with sand. The giant works of Stonehenge and Avebury must also once have had their much smaller models.

May this example of the unintentional destruction of prehistoric remains show just how exceptionally careful one must be, even when one is dealing with apparently trivial things. May the wide distribution of this periodical serve to make this warning much more accessible!

An index of more of my translations of old Kosmos articles can be found at:

Kosmos Translations Archive

A number of Mesozoic (and post-Mesozoic) location summaries can be found at Localities.