Theran Floral Style in Relation to that of Crete

Theran art is conservative in its choice of subject, in pottery decoration and in technique. Stylistically, however, it is characterized by vividness, movement and an observation of nature not to be found in the pompous, stiffly symmetric works of the Knossos Palace School long before the LM II period. These differences enable the assignation of some Theran frescoes to local artists. Only in the West House and Xesté 3 frescoes are foreign, probably Cretan, elements to be found to a significant extent.

The conservative trends in Theran art, in the author's opinion, favour the possibility that the island was not deserted until a time when at Knossos the LM II Palace Style already existed.

Regrettably, Marinatos has hardly ever commented upon the stylistic qualities of the Theran works of art. Once (1971, 65) he calls Theran painting "more crude, but more sincere" than that of Cretan palatial art, the latter being defined as "a little effeminated and frozen". In another context (1973, 22f., 27) he emphasizes the sense of originality and vividness permeating Theran frescoes, and concludes that the manner of painting on Thera was relatively independent of both Crete and the Mainland. He provisionally calls its style "Theran", expecting that further finds will allow a definition as "Aegean" in general.

I should like to put those definitions to the test on the basis of a detailed comparison of floral subjects in Theran and Cretan art. We shall, for obvious reasons, have to confine ourselves to comparing the following groups of subjects:

-     Trees (A),

-     Crocus/Saffron (B),

-     Grass/Reeds (C),

-     Lilies (D), and

-     Papyrus (E).

With regard to the exchange of ideas between wall - and vase - painters (Evans 1921, 554, 603 ff., idem 1928, 468 ff.; Shaw 1972, 187), not only frescoes (A.1 etc.) but also floral ornaments on pottery vessels (A.2 etc.) will be considered. There will be a special purpose in regarding the matt-painted white-ground ware since this type of pottery was doubtlessly produced by the indigenous, non-Cretan population and thus bears witness to the taste and the outlook of this group among the inhabitants of Thera. Chronological differences between things compared can to some extent be disregarded since style itself evidently did not change much during the period in question either in Thera or in Crete.

• A.1.

Trees are not a common motif in Theran frescoes, being restricted to the West House and Xesté 3. In the miniature fresco on the South wall of Room 5 of the West House there are some trees (Fig. 1a) painted with a boldness somewhat reminiscent of Gauguin's style; irregularly bent trunks and rather schematic crooked branches end in cloud-like patches of light-brown intended as amorphous masses of foliage.

This crude rendering cannot be explained by the miniature character of the West House fresco, since in another miniature fresco in Xesté 3 (Marinatos 1974, pl. 23b) trees, or rather shrubs, are painted with individual branches and leaves without the patch type of foliage. These delicately rendered trees far surpass those in the West House.

Similar differences in quality are to be found within the miniature fresco on the East wall of the West House room (ibidem, Col. pl. 8). In this "Libya Fresco" some trees are without any foliage whatsoever, unless small groups of straight lines radiating from the ends of the spiky boughs are to be taken for leaves. On the other hand, two palm-trees (Fig. 1c) are rendered in a most accurate and naturalistic manner, even differentiating between (bluish) fresh fronds and (brownish) withered ones. These palm-trees, evidently painted by the hand of a master, were crudely and schematically copied by at least one other second-rate artist who apparently had never seen palm-trees himself. A couple of palm-fronds on a fresco fragment found in Arvanitis 1 (Marinatos 1969, pl. B 3) are of intermediate quality.

In comparison, Knossian artists not only outlined the foliage of trees as a patch of colour but, neatly and vividly, painted branches with individual leaves in front of this background (Fig. 1b), or (Evans 1967, pls. D 3.4; VIII 3), as in the miniature fresco from Tylissos (Hazzidakis 1921, pl. 3), they abandoned the patch type of foliage outline altogether. The local trees in the Xesté 3 fresco are of equal quality while those on the South wall of Room 5 in the West House are markedly inferior.

The "master's" palm-trees in the "Libya Fresco" even surpass Knossian examples such as MM IIIb palm-tree vases (Fig. 1f) or an undated fresco (Fig. 1d), although the general idea is already to be found there by MM III (Evans 1921, fig. 436c). This holds true for the general composition as well as for the choice and rendering of details, the manner of which conforms to that of the local trees in Xesté 3 which may originate from the same hand.

The "West House Master" differs so much from his fellow artists that it is tempring to see in him a Cretan (or a Theran trained in Crete) called to Thera by the owner of the West House who, as Marinatos has shown, was probably a high officer in the Cretan navy. The existence of tree pictures in this very house and in Xesté 3, the character of which could not be identified hitherto) is possibly not merely due to chance. In any case, trees are absent from other published frescoes in Thera.

• A.2.

They are also practically absent from pottery decoration. In two cases (Fig. 1e; Marinatos 1974, pl. 82b) palm-trees decorate a polychrome pitcher and a "kymbe", all of them ugly and even distorted in the extreme. Apparently local potter-painters copied the motif from some original elsewhere (the "Libya Fresco" ?) without any understanding of, or interest in, the nature of the trees.

This contrasts with the keen observation of nature which is especially characteristic of the local class of white-ground ware to which the pitcher belongs, which underlines our observation that trees of any kind were uncommon subjects, probably of foreign origin, for the local artists of Thera.

• B.1.

Crocuses and similar flowers are more frequent. In frescoes such as those from Rooms 2 and 3 in Xesté 3 (Fig. 2a; Marinatos, 1976, pls. B - E; 39b), crocus plants are shown in a most naturalistic manner, thick bushes of slender leaves spreading irregularly in every direction.

The finest Cretan fresco representations of crocuses are some fragments of MM III / LM Ia date from the House of the Frescoes at Knossos (Fig. 2b). In their stiff parallel arrangement, however, they are far from reaching the liveliness of those from Thera. Some flowers in the "Saffron Gatherer" fresco from Knossos (Evans 1921, pl. 4) compare more favourably with Theran art although, in detail, they too are rather standardized. The sense of naturalism and individuality characteristic of the Theran flowers is absent from all the Knossos frescoes.

• B.2.

The fact that crocuses appear only in the Xesté 3 frescoes does not imply that were an unfamiliar subject to Theran artists; many representations on pottery are proof of the contrary. Besides specimens on imported vessels (Marinatos 1972 pls. 48; 64a; 88; idem 1974, pl. 78; idem 1976, pl. 50), one of them showing crocus blossoms radiating from a "LM Ia" whirl pattern (idem 1969, pl. D6), there are others on local white-ground ware (Fig. 2h), white-on-dark vessels (idem 1968, fig. 46; idem 1974, pl. 74a), offering-stands (Fig. 2f) and dark-on-light ware (op. cit., pl. 73b; idem 1976, pl. 47 a-b) which prove the popularity of crocuses in local pottery. It deserves mention that the triple group on the white-ground askos cited is the finest of them all. Curiously enough, though, there are two different ways of arranging the flowers, either vertically or bent to one side in the typically Cretan principle of torsion (e.g., Boyd Hawes 1908, pl. VIII - 21 from Gournia). In some cases (Fig. 2d), crocus-type blossoms are placed on lily-like stems, whereas in another (Marinatos 1971, pl. 79) single blossoms are scattered about the upper part of a stirrup-jar in a random fashion.

It is important to note that among published Theran vessels with crocus decoration, local and imported alike, there is not a single specimen of the stylized "pendant crocus" typical of Cretan LM Ib (Evans 1935, figs. 221 -222; Pendlebury 1939, fig. 38, 9) but only naturalistic renderings in the manner of LM Ia (Pendlebury op. cit., fig. 36, 12). Compared with the latter, Theran crocus designs offer a wide range in artistic quality reaching from the excellent white-ground example to extremely ugly ones (Marinatos 1971, pl. 64c), but even the poorest specimens avoid any tendency towards ornamental stylization of "pendant crocus" type. It will aid an understanding of the situation to bear in mind that also in Crete, as in Gournia which offers the closest parallels to Theran art, "old fashioned" naturalistic renderings of crocuses by far outnumber stylized "pendant" ones. Still more conservative an attitude may be inferred for Thera where in the same catastrophe layers vessels painted in MM III manner, with white ornaments on dark ground, are found side by side with polychrome white-ground ware and LM I (or later) dark-on-light ware. Confronted with a taste that was conservative, new ideas already commonly adopted at Knossos may well have found even less resonance in Thera than they did in eastern Crete; maybe even imported vessels were rejected when not conforming to local taste. This conservatism, expressing itself in a uniform predilection for naturalistic plant design and the rejection of all kinds of ornamental stylization, will be an aspect relevant to the discussion of the date of the catastrophe.

• C .1.

Grasses and reeds are a common subject in both Theran and Cretan art although, inconspicuous as they are, they are less often used in frescoes than in pottery decoration. Papyrus ("waz") is to be excluded from this statement since, in view of its current combination with lilies, it will be treated separately below.

Grass designs in Theran frescoes (Fig. 3a; Marinatos 1969, pl. 5,2) are treated in the same vivid way as are crocus plants. Stalks and leaves are drawn distinctly, resulting in representations very close to nature. In Knossos (Fig. 3b; Evans 1928, fig. 264) or Katsamba (Alexiou 1955, fig. 2), for comparison, the stalks are not shown altogether but rather symmetric pairs of leaves are placed on top of each other, the plants themselves being stiffly and vertically arranged side by side without any of that overlap which adds so much to the vividness of the Theran designs. It should, however, be mentioned that in some fresco fragments from the South-East House at Knossos (Fig. 3c) and Ayia Triada (Evans 1928, fig. 201) grasses are shown in a style approaching the quality of the Theran. It would seem that the stiff and formal rendering of these plants was more characteristic of the Royal Court at Knossos than of Crete as a whole.

• C . 2

In Theran pottery decoration the same formulae are used as in Crete (Fig. 3e), whether it be local matt-painted ware (Fig. 3d) , white-on-dark ware (Marinatos 1969, fig. 24) or dark-on-light ware (idem 1971, pl. 68b). On pithoi (idem 1970, pl. 49,1) the reeds may even be distorted into awkward groups of curved lines which have become scarcely recognizable. As in Crete, vertical (cf. Boyd Hawes 1908, pl. VII - 24) and diagonal arrangements (Pendlebury 1939, fig. 36,7) occur side by side.

• D .1.

Lilies, together with sea-daffodils (pancratium), may be called a favourite subject of Theran, as well as of Cretan art. The magnificent frescoes in the "Lilies' Room" (Δ2) (Fig. 3f) or Room 4 in the West House (Fig. 3h) are masterpieces of their kind. The lilies in the former, sprouting from rocky cliffs in an idealized landscape, are treated in an unusually animated manner, no two plants being exactly identical. The stems are curved, sometimes in S bends, in different directions, crowned by manifold groups of blossoms and buds without any symmetry or standardization. Marinatos (1971, 50) has aptly commented on these features. It is astonishing to note that the West House specimens lack these qualities. They are, of course, placed in vases which might account for the rigid symmetry of the stems; but the fact that blossoms and buds are also symmetrically arranged on the stems suggests, in my opinion, that these frescoes as a whole are characterized by a different artistic attitude. We find the same rigid symmetry again in the groups of sea-daffodils in the "Room of the Ladies" (Marinatos 1972, pls. E - F; 11b) which in arrangement and size alike are regally pompous rather than inspred by nature.

In Knossos (Evans 1928, fig. 266c) or Amnissos (Fig. 3i) fresco lilies and sea-daffodils (Evans op. cit., fig. 268) tend, as Marinatos (1971, 50) noted, as a rule towards a style as stiff and symmetric as that of the West House and "Room of the Ladies'" designs. Most characteristic of this attitude are the lilies in the "Priest-King" relief (Evans 1928, pl. XIV) which are mixed, in a manner popular in Crete, with papyrus (waz) blossoms. Conversely, in frescoes from Ayia Triada, both lilies  (Fig. 3g) and sea-daffodils (Evans 1928, fig. 201) are painted in as animated a style as that of the Theran lilies in room Δ2. It appears as though the pompous style was characteristic of Knossos and its vicinity well in advance of the Palace Style of LM II, which also had its centre in the same area.

• D. 2.

Lilies are also popular in Theran pottery decoration but, curiously enough, among the material published the subject is confined without exception to local white-on-dark ware and, to a lesser degree, to matt-painted ware (Fig. 4d; in this singular case the blossoms are highly stylized). The predilection for white-on-dark lilies is so marked that on a strainer from Xesté 3 (Fig. 4b) crocuses and swallows of the main frieze are painted dark-on-light whereas the lilies in the foot panel are executed in the opposite technique. Regarding the arrangement of white lilies there is considerable variety. A large cylindrical vessel (Marinatos 1970, pl. 64 a-b), for instance, is decorated with pompous symmetrical lily bushes whereas on a flower-pot from Δ8 (op. cit., pl. 83a) the lilies, though drawn in a rather schematic manner, are arranged in a lovely animated composition with some dead blossoms being borne away by the breeze. In other cases (e.g., Marinatos 1969, pls. E7; 25, 1) single blossoms drift over the vessel's surface.

There is no fundamental difference between these white-on-dark lily designs from Thera and Cretan ones (Fig. 4c), the latter being typical of the MM III period. The fact remains, however, that among published vessels from Thera there is not a single one painted in dark-on-light technique, a style which in Crete (Fig. 4a) was popular till LM II times.

If this statement is correct - and in the preparation of this paper I had to rely on the literature - it emphasizes that Theran art was not only conservative in general, but even conservative in a selective way that has to be taken into consideration when dating the catastrophe.

• E.1.

It has been mentioned that papyrus, the depiction of which goes back to Egyptian prototypes ("waz"), is a popular subject in Cretan art (Fig. 5c), either rendered in a naturalistic way or admixed with lilies. It is all the more surprising that in Thera papyrus has been depicted in only two frescoes in the West House. In the "Libya Fresco", some papyrus plants are shown growing by the side of the river (Fig. 5a). They are painted in a rather naturalistic manner, although less detailed than most of the Cretan frescoes. The only other evidence are waz symbols, or rather waz-lilies, topping some of the captains' huts ("ikria") in room 4 of the same building.

• E. 2.

On a matt-painted pitcher from room Δ3 (Marinatos 1971, pl. 71), there are two signs which might be interpreted as waz-lilies. If this were correct, their stylization would, as in the case of the matt-painted trees and lilies, contrast with the naturalism characteristic of the white-ground ware. In my opinion this maybe taken as an indication that all of these subjects were exotic for the indigenous Theran potter-painters.

CONCLUSIONS

There exist differences between Theran and Cretan art even more marked than were hinted at by Marinatos. Characteristics of Theran style are a keen observation of nature and the animated arrangement of plants without rigid symmetry. Some designs include overlapping while in others even the action of the breeze is indicated. Cretan, and especially Knossian art, on the contrary, might be defined as formal and standardized with a predilection for symmetrical or parallel arrangement of the individual plants. "Cretan" elements, however, occur in the frescoes of the West House and Xesté 3. It would seem that their owners were Cretans who had their houses in Thera decorated by artists from the school of Knossos.

At the same time Theran art preserves techniques such as light-on-dark painting long outdated in Crete. This makes us ask how long this survival went on, or, to put it in another way, at which period of the Cretan chronology Thera was destroyed. Whereas Marinatos, despite some hesitation (1970, 68), preferred until his death to date the catastrophe in the LM Ia period, J.V. Luce (1976) suggested a dating in LM Ib, and I myself (1974) pointed out arguments for an event later dating to the third quarter of the 15th century B.C.

In this context a non-floral detail from Xesté 3 (Fig. 5d) deserves mention. A fresco panel there consists of interlaced patterns of stylized rocks to be found neither elsewhere on Thera nor in Cretan frescoes. Single stylized rocks, however, are known from Cretan pottery of the LM Ib Alternating Style, while such patterns, interlaced to form a network covering the vessel's surface, frequently occur in me LM II Palace Style and later. One such vessel (Fig. 5e) was found in the catastrophe layer at Zakro which, according to vulcanologists (Kleinmann 1974), was destroyed simultaneously with Thera. At this time in eastern Crete LM Ib Marine Style pottery predominates over both surviving LM Ia elements and occasional intrusive specimens, such as the Zakro vessel cited, which testify to influence from the LM II Palace Style of the Knossos area. In my opinion the Xesté 3 fresco (Fig. 5d) favours the possibility that Thera was not deserted until a time when at Knossos the LM II Palace Style already existed. If this is correct, MM III elements survived on Thera for several generations after having gone out of use in central Crete, which would indicate a really surprising degree of conservatism in Theran art.

--------------------------------------

-----------------------------------