In Tunis the dolmen is not uncommon, and several groups or cemeteries have been reported. Near Ellez occurs a type of corridor-tomb in which three dolmen-like chambers lie on either side of a central pa.s.sage, and a seventh at the end opposite to the entrance. The whole is constructed of upright slabs of stone, and is surrounded by a circle formed in the same way.

Morocco, too, has its dolmens, especially in the district of Kabylia, while near Tangier there is a stone circle.

Off the north coast of Africa, and thus on the highway which leads from Africa to Europe, lie the Italian islands of Lampedusa and Linosa. The latter is volcanic in origin, and its surface presents no opportunity for the building of megalithic monuments. Lampedusa, on the other hand, consists of limestone, which lies about in great blocks on its surface.

On the slopes of the south coast there are several remains of megalithic construction, but they are too damaged to show much of their original form. However, on the north side of the island there are megalithic huts in a very fair state of preservation. They are oval in form and have in many cases a base course of orthostatic slabs.

Some miles to the north of Linosa lies the much larger volcanic island of Pantelleria, also a possession of Italy. Here megalithic remains both of dwellings and of tombs have been found. On the plateau of the Mursia are the remains of rectangular huts made of rough blocks of stone. These huts seemed to have formed a village, which was surrounded by a wall for purposes of defence. In the huts were found implements of obsidian and flat stones used for grinding.

[Ill.u.s.tration: FIG. 20. Plan of the Sese Grande, Pantelleria.

(Orsi, _Monumenti Antichi_, IX.)]

The tombs of the people who inhabited this village are, unlike the houses, circular or elliptical in form. They are locally known as _sesi._ The smaller are of truncated conical shape, the circular chamber being entered by a low door and having a corbelled roof. In one of the _sesi_ a skeleton was found buried in the contracted position. The finest of the tombs, known as the Sese Grande, elliptical in form (Fig.

20), has a major diameter of more than 60 feet, and rises in ridges, being domed at the top. It contains not one chamber, but twelve, each of which has a separate entrance from the outside of the _sese._ To judge by the remains found in the _sesi_ they belong entirely to the neolithic period.

The island of Malta as seen to-day is an almost treeless, though not unfertile, stretch of rock, with a harbour on the north coast which must always make the place a necessary possession to the first sea power of Europe. Much of its soil is of comparatively modern creation, and four thousand years ago the island may well have had a forbidding aspect.

This is perhaps the reason why the first great inroads of neolithic man into the Mediterranean left it quite untouched, although it lay directly in the path of tribes immigrating into Europe from Africa. The earliest neolithic remains of Italy, Crete, and the aegean seem to have no parallel in Malta, and the first inhabitants of whom we find traces in the island were builders of megalithic monuments. Small as Malta is it contains some of the grandest and most important structures of this kind ever erected. The two greatest of these, the so-called "Phoenician temples" of Hagiar Kim and Mnaidra, were constructed on opposite sides of one of the southern valleys, each within sight of the other and of the little rocky island of Filfla.

[Ill.u.s.tration: FIG. 21. Plan of the megalithic sanctuary of Mnaidra, Malta. (After Albert Mayr's plan.)]

The temple of Mnaidra is the simpler of the two in plan (Fig. 21). It consists of two halves, the more northerly of which was almost certainly built later than the other. Each half consists of two elliptical chambers set one behind the other. The south half is the better preserved. It has a concave facade of large orthostatic slabs with horizontal blocks set in front of them to keep them in position. In the centre of this opens a short paved pa.s.sage formed of fine upright slabs of stone, one of which is 13 feet in height. The first elliptical chamber (_E_) into which this pa.s.sage leads us has a length of 45 feet.

Its walls (Pl. III) consist of roughly squared orthostatic slabs over 6 feet in height, above which are several courses of horizontal blocks which carry the walls in places up to a height of nearly 14 feet. This combination of vertical and horizontal masonry is typical of all the Maltese temples. To the left of the entrance is a rectangular niche in the wall containing one of the remarkable trilithons (_a_) which form so striking a feature of Mnaidra and Hagiar Kim. It consists of a horizontal slab of stone nearly 10 feet in length, supported at its ends by two vertical slabs about 5 feet high. To the right of the entrance is a window-like opening (_b_, behind the seated figure in Pl. III) in one of the slabs of the wall, preceded by two steps and giving access to an irregular triangular s.p.a.ce (_F_). In the north-west angle of this triangle is fixed a trilithon table (_c_) of the usual type, 32 inches high; at a like height above the table is fixed another horizontal slab which serves as a roof to the corner. The south corner of the triangle is shut off by a vertical slab, in which is cut a window 29 inches by 17. Through this is seen a shrine (?) consisting of a box (_d_) made of five well-cut slabs of stone, the front being open. The aperture by which _F_ is entered was evidently intended to be closed with a slab of stone from the inside of _F_, for it was rebated on that side, and there are holes to be used in securing the slab. When the entrance was thus blocked _F_ still communicated with _E_ by means of a small rectangular window 16 inches by 12 in one of the adjacent slabs (visible in Pl.

III).

[Ill.u.s.tration: PLATE III TEMPLE OF MNAIDRA, MALTA. APSE OF CHIEF ROOM To face p. 100]

Returning to the area _E_ we find in the south-west wall an elaborate doorway (Pl. II, Fig. I, p. 82) leading to a rectangular room _H_. The doorway consists of two tall pillars with a great lintel laid across the top. The s.p.a.ce between the pillars is closed by a fixed vertical slab in which is a window-like aperture similar to that which gives access to Room _F_. All the stones in this doorway are ornamented with pit-marks.

The rectangular room _H_ has niches in its walls to the north, south, and west. Each niche is formed by a pair of uprights with a block laid across the top. The west niche is occupied by a horizontal table or slab (_e_) supported at its centre by a stone pillar 39 inches in height, of circular section narrowing in the centre (visible through the doorway in Pl. II, Fig. I). The southern niche contains an ordinary trilithon table (_f_): the northern niche is damaged, but apparently held a table like that of the western.

The area _I_ consists of only half an ellipse, the southern half being replaced by the area _H_, which we have already described. It has a rectangular niche to the west containing a fine trilithon with a cover-slab nearly 10 feet long.

The whole of the southern half of the Mnaidra temple is surrounded by a wall of huge rough blocks of stone, presenting a great contrast to the dressed slabs of which the inner walls are formed. They are placed alternately with their broad faces and their narrow edges outwards. The roughness of this enclosure wall gives the structure a remarkably wild and craggy appearance from a distance. The northern half of Mnaidra is clearly a later addition.

There is no doubt as to the way in which the areas were roofed. In the apse-like ends of the elliptical rooms the horizontal courses are corbelled, i.e. each course projects slightly forward over the last.

Thus the s.p.a.ce narrows as the walls rise, until the aperture is small enough to be roofed by great slabs laid across. The corbelling of the apse is just perceptible in Pl. III. Whether the roofing of the Mnaidra temple was ever complete it is impossible to say: in any case the system we have described could only be applied to the apsidal portions of the areas, and their centres must either have been open to the sky or roofed quite simply with slabs.

In the still more famous temple of Hagiar Kim we have a complicated building, in which the original plan has been much altered and enlarged.

The main portion doubtless consisted originally of a curved facade and a pair of elliptical areas, the inner of which has been fitted with a second entrance to the north-west and completely remodelled at its south-west end. Four elliptical chambers, one of which is at a much higher level than the rest of the building, have been added. Here, too, as at Mnaidra, we find niches containing trilithon tables. In the first elliptical area, in which the apsidal ends are divided from the central s.p.a.ce by means of walls of vertical slabs, a remarkable group of objects was found. In front of a well-cut vertical block stood what must be an altar, cut in one piece of stone. It is square in section except for the top, which is circular. On the four vertical edges are pilasters in relief, and in the front between these is cut in relief what looks like a plant growing out of a pot or box. To the left of the altar and the vertical slab behind were an upright stone with two hanging spirals cut on it in relief, and at its foot a horizontal slab. Both the altar and the carved stone are covered with small pit-marks.

In the outside wall of the building, quite unconnected with the interior, is a niche partly restored on old foundations, in which stands a rough stone pillar 6-1/2 feet high. In front of this pillar is a vertical slab nearly 3 feet high, narrowing towards the base, and covered with pit-markings. This pillar can hardly be anything but a baetyl, or sacred stone.

The temple called the Gigantia, on the island of Gozo, is no less remarkable than the two which we have already described; in one place its wall is preserved up to a height of over 20 feet. The plan is similar to that of Mnaidra, though here the two halves seem to have been built at one and the same time. Several of the blocks show a design of spirals in relief, while on others there are the usual pit-markings.

Another bears a figure of a fish or serpent. At the foot of one of the trilithons was found a baetyl 51 inches in height, now in the museum at Valletta.

That these three buildings were sanctuaries of some kind seems almost certain from their form and arrangement. We do not, however, know what was the exact nature of the worship carried on in them, though there can be no doubt that the stone tables supported by single pillars and the trilithons found in the niches played an important part in the ritual.

Sir Arthur Evans in his famous article _Mycenaean Tree and Pillar Cult_ has suggested that in Malta we have a cult similar to that seen in the Mycenaean world. This latter was an aneiconic worship developed out of the cult of the dead; in it the deity or hero was represented by a baetyl, i.e. a tree or pillar sometimes standing free, sometimes placed in a 'dolmen-like' cell or shrine, in which latter case the pillar often served to support the roof of the shrine. In Malta Sir Arthur Evans sees signs of a baetyl-worship very similar to this. Thus at Hagiar Kim we have a pillar still standing free in a niche, and another pillar, which, to judge from its shape, must have stood free, was found in the Gigantia. On the other hand, at Mnaidra we have pillars which support slabs in a cell or shrine, and at Cordin several small pillars were found which must originally have served a similar purpose.

There can hardly be any doubt that Sir Arthur Evans is right in seeing in the Maltese temples signs of a baetylic worship. But is he right in his further a.s.sertion that the cult was a cult of the dead? Albert Mayr a.s.sumes that he is, and endeavours to show that the 'dolmen-like' cells in the niches are not altars, but stereotyped representations of the dolmen-tombs of the heroes worshipped. He thinks that the slabs which cover them are too large for altar-tables, and that the niches in which they stand are too narrow and inaccessible to have been the scene of sacrificial rites. Neither of these arguments has much force, nor is it easy to see how the cells are derived from dolmens. The fact is that the word 'dolmen-like,' which has become current coin in archaeological phraseology, is a question-begging epithet. The Maltese cells are not like dolmens at all, they are either trilithons or tables resting on a pillar. They are always open to the front, and instead of the rough unhewn block which should cover a dolmen they are roofed with a well-squared slab. If the pillar which supports the slab is, like the free-standing pillars, a baetyl, the slab is probably a mere roof to cover and protect it; if not, the slab is almost certainly a table.

At the same time, although we may not accept the hypothesis that the cell is derived from a dolmen, Sir Arthur Evans may still be right in supposing the worship to have originated in a cult of the dead. But he was almost certainly wrong, as recent excavation has shown, in supposing that the cells were the actual burial place of the deified heroes.

A number of statuettes were found at Hagiar Kim, two of which are of pottery and the rest of limestone. One figure represents a woman standing, but in the rest she is seated on a rather low stool with her feet tucked under her. There is no sign of clothing, except on one figure which shows a long shirt and a plain bodice with very low neck.

All these statuettes are characterized by what is known as steatopygy, that is, the over-development of the fat which lies on and behind the hips and thighs.

Steatopygous figures have been found in many places, viz. France, Malta, Crete, the Cyclades, Greece, Thessaly, Servia, Transylvania, Poland, Egypt, and the Italian colony of Eritrea on the Red Sea. The French examples are from caves of the palaeolithic period; the rest mainly belong to the neolithic and bronze ages. Various reasons have been given for the abnormal appearance of these figures. In the first place it has been suggested that they represent women of a steatopygous type, like the modern Bushwomen, and that this race was in early days widely diffused in the Mediterranean and in South Europe. Another hypothesis is that they represent not a truly steatopygous type of women, but only an abnormally fat type. A third suggestion is that they portray the generative aspect of nature in the form of a pregnant G.o.ddess.

Naturally there are considerable local differences in the shapes of the figures from the various countries we have enumerated, and it may be that no single hypothesis will explain them all.

There are other megalithic buildings in Malta besides the three which we have discussed, but none of them call for more than pa.s.sing mention.

On the heights of Cordin or Corradino, overlooking the Grand Harbour of Valletta, there are no less than three groups, all of which have been lately excavated. In all three we see signs of the typical arrangement of elliptical areas one behind another, and in the finest of the three the curved facade and the paved court which lies before it are still preserved.

It was for a long time believed that there were no dolmens in Malta.

Professor Tagliaferro has been able to upset this belief by discovering two, one near Musta and the other near Siggewi. It is hardly credible that these are the only two dolmens which ever existed in Malta. More will no doubt yet be found, especially in the wild north-west corner of the isle.

The megalithic builders of Malta did not confine their achievements to structures above ground, they could also work with equal facility below.

In the village of Casal Paula, which lies about a mile from the head of the Grand Harbour of Valletta, is a wonderful complex of subterranean chambers known as the Hypogeum of Halsaflieni, which may justly be considered as one of the wonders of the world.

The chambers, which seem to follow no definite plan, are excavated in the soft limestone and arranged in two storeys connected by a staircase, part of which still remains in place. The finest rooms are in the upper storey. The largest is circular, and contains in its walls a series of false doors and windows. It is in this room that the remarkable nature of the work in the hypogeum is most apparent. On entering it one sees at once that the intention of the original excavator was to produce in solid rock underground a copy of a megalithic structure above ground.

Thus the walls curve slightly inwards towards the top as do those of the apses of Mnaidra and Hagiar Kim, and the ceiling is cut to represent a roof of great blocks laid across from wall to wall with a s.p.a.ce left open in the centre where the width would be too great for the length of the stones. The treatment of the doors and windows recalls at once that of the temples above ground. The mason was not content, when he needed a door, to cut a rectangular opening in the rock; he must represent in high relief the monolithic side-posts and lintel which were the great features of the megalithic 'temples' of Malta. Nor has he failed in his intention, for, as one moves from room to room in the hypogeum, one certainly has the feeling of being in a building constructed of separate blocks and not merely cut in the solid rock. No description can do justice to the grace of the curves and the flow of the line in the circular chamber and in the pa.s.sage beyond it, and we have here the work of an architect who felt the aesthetic effect of every line he traced.

Behind the circular chamber and across the pa.s.sage just referred to lies a small room which, rightly or wrongly, has been called the 'Holy of Holies,' the idea being that it formed a kind of inner sanctuary to the chamber. It contains a rough shelf cut in the wall, and in the centre of this a shallow circular pit. It has been suggested that this pit was made to hold the base of the cult-object, whether it was a baetyl or an idol. This, however, is a mere conjecture. In the pa.s.sage just outside the door of this room are two small circular pits about 6 inches in diameter and the same distance apart. They connect with one another below, and are closed with tightly fitting limestone plugs. In one of them was found a cow's horn. Their purpose is unknown, but similar pairs of pits occur elsewhere at Halsaflieni.

In two of the largest chambers in the hypogeum the roof and walls are still decorated with designs in red paint. The patterns consist of graceful combinations of curved lines and spirals. Many other rooms, including the circular chamber, were originally painted with designs in red, which have now almost wholly disappeared.

Many of the chambers are extremely small, too small for an adult even to stand upright in them, and their entrances are merely windows, perhaps a foot square and well above the ground.

What then was the purpose of this wonderful complex of rooms? Before attempting to answer this question we must consider what has been found in them. When the museum authorities first took over the hypogeum practically all the chambers were filled to within a short distance of their roofs with a ma.s.s of reddish soil, which proved to contain the remains of thousands of human skeletons. In other words, Halsaflieni was used as a burial place, though this may not have been its original purpose. The bones lay for the most part in disorder, and so thickly that in a s.p.a.ce of about 4 cubic yards lay the remains of no less than 120 individuals. One skeleton, however, was found intact, lying on the right side in the crouched position, i.e. with arms and knees bent up.

With the bones were found enormous quant.i.ties of pottery and other objects, buried with the dead as provision for the next world. The pottery is rough in comparison with the fine painted wares of Crete, but it is extremely varied in its decoration. One particularly fine bowl shows a series of animals which have been identified by Professor Tagliaferro as the long-horned buffalo, an animal which once existed on the northern coasts of Africa. Ornaments of all kinds were common, and include beads, pendants, and conical b.u.t.tons of stone and sh.e.l.l. The most remarkable of all are a large number of model celts made of jadeite and other hard stones. These are of the same shape as the stone axes used by neolithic man, but they are far too small ever to have been used, and they must therefore have been models hung round the neck as amulets. Each is provided with a small hole for this purpose. The popularity of the axe-amulet makes it probable that the axe had some religious significance.

Finally Halsaflieni has yielded several steatopygous figurines. Some of these resemble those of Hagiar Kim, but two are of rather different type. Each of these represents a female lying on a rather low couch. In the better preserved of the two she lies on her right side, her head on a small uncomfortable-looking pillow. The upper part of her body is naked, but from the waist downwards she is clad in a flounced skirt which reaches to the ankles. The other figurine is very similar, but the woman here is face downwards on the couch.

The bodies themselves were so damaged with damp that only ten skulls could be saved whole. These, however, afford very valuable anthropological evidence. They have been carefully measured by Dr.

Zammit, and they prove to belong to a long-headed (dolichocephalic) type usual among the neolithic races of the Mediterranean.

We have still to discuss the purpose of this great complex of underground chambers and pa.s.sages. It is quite clear that its eventual fate was to be used as a burial place for thousands of individuals, but it is far from certain that this was the purpose for which it was built.

The existence of the central chamber, with its careful work and laborious imitation of an open-air 'temple,' is against this interpretation. It has therefore been suggested that the hypogeum was meant for a burial place, and that the central chamber was the chapel or sanctuary in which the funeral rites were performed, after which the body was buried in one of the smaller rooms. This, however, does not explain the presence of burials in the chapel itself, and it is far more likely that it was only after Halsaflieni had ceased to be used for its original purpose that it was seized upon as a convenient place for burial.

The question of the date of the Maltese megalithic buildings is a difficult one. It is true that no metal has been found in them, and that we can therefore speak of them as belonging to the neolithic age. But the neolithic age of Malta need not be parallel in date with that of Crete for example. It is extremely probable that Malta lay outside the main currents of civilization, and that flint continued to be used there long after copper had been adopted by her more fortunate neighbours.

CHAPTER VIII