Aegyptus

Province Description

1. Definitions.

Egypt was by ancient standards an intensively cultivated country; the large part of its territory that was stony or sandy desert only accentuated the fertility and cultivation of the small percentage of the land reached by the waters of the Nile or by water from wells (in the case of the oases). "Garden" is not a technical term, but it must be defined for our present purposes. Although Egypt was more intensively cultivated than, for example, Greece, the term will be used here to exclude the large part of the country under arable cultivation in wheat, barley, emmer, leguminous crops, hay, fodder, flax, and other plants grown in open fields.¹ What remains are vineyards, orchards and groves, vegetable gardens, and flower gardens. It is partly the intensity of cultivation, partly the plants grown, and partly the ancients' own vocabulary that define gardens in Egypt, but none of these by themselves is decisive. For example, olive trees grown in Greece on terraces without boundary walls in open country are not generally considered as gardens; but olive trees in Egypt were grown almost entirely, as far as we can see, in walled gardens. Dominic Rathbone's assessment of what κῆπος means in the papyri is a fair summary (Rathbone 1991: 186): "usually small, intensively worked, enclosed and irrigated plots of land which nestled around villages."²

Kepos was, however, not the principal technical term for gardens; in official language, that was παράδεισος. Paradeisoi did not, however, include vineyards, and in official use agricultural land not planted in arable crops tends to be divided between ampeloi (vineyards) and paradeisoi (orchards, including specifically palm trees and olive trees). The Roman term pomarium appears in Greek (πωμάριον) as a descriptor for orchards, but it does not replace paradeisos as the official designation of the category of land. Generally speaking the government was not interested in the precise plantings in paradeisoi.

A wide variety of combinations of the basic types of cultivation was known already in pharaonic times, and probably mixed types were normal rather than exceptional in Roman Egypt as well. Our evidence (cf. below) concerns mainly properties exploited for economic return or for subsistence use rather than ornamental plots or tomb gardens, but what can be discovered about these latter categories is included, and tomb gardens might well be of economic use. Our greatest difficulty lies with ornamental plantings. There is some evidence about plantings in temples, thanks mainly to excavation of pharaonic sanctuaries, but hardly any from villages or cities. This may be only a reflection of the types of evidence we possess, but habitation sites were mostly very densely built up and may not have afforded much opportunity for decorative planting. A special problem is indeed posed by the temples. What archaeological evidence we have is heavily concentrated on the sanctuaries built in the late period, some finished only in the early Roman period. But this evidence is intelligible only in comparison with that from the temples of earlier periods. For discussion, see below, section 9.

There are inevitably uncertainties about inclusion. Some vegetables were certainly grown as field crops, not in anything we would call a garden; this is particularly true of crops like "radishes," the seeds of which were used for oil production (the plant may not have been what is eaten today under that name). For this reason I have relied heavily on the use of vocabulary to identify a plot of land as a vineyard, orchard, or garden, and not only on the presence of plants characteristic of gardens.

2. Evidence.

The overwhelming bulk of our evidence comes from the Greek papyri. There is only a small number of Demotic Egyptian documents from the Roman period, and almost all of the relevant handful for our subject come from the period before 50 CE. Archaeological evidence is scanty: the exploration of habitation sites from Roman Egypt is very poorly developed. Excavations at a few sites have yielded botanical evidence, but actual garden sites have not been excavated. Nor is there much surviving representation of gardens from this period as there is for the Middle and New Kingdoms. The papyrus evidence is abundant but limited. To a large degree it is the product of what was necessary to assess, collect, and convey the government's taxes on agricultural land and activity. The government collected taxes both in kind and in money; those in kind are mostly in wheat and barley. Garden land of all types is rarely assessed or taxed in a way that reflects the actual produce, for the government was not interested in warehousing perishables like cucumbers and roses. Vineyards, as a distinct, common, and identifiable genre, show up in the texts much more often than any other non-arable type of cultivation, and this is probably a reliable indicator of actual gardening patterns, mainly because of the large volume of wine consumed. By the same token, we often know what was grown and sometimes the size of a plot, but only exceptionally much of the other physical characteristics of the garden itself, and almost never its internal arrangement.

A particular gap in our evidence is the lack of literary texts written in and referring to Roman Egypt. The question raised by Robin Osborne about classical Greece cannot even be asked: "...whether we are being given a description of the garden at all, or whether, rather, we are dealing in the imaginative product of ideological conviction" (Osborne 1992: 374). There is little direct access to how the inhabitants of Roman Egypt thought about their gardens. It is, however, fairly clear that gardens were, per unit of surface area, highly valuable properties; they required, moreover, considerable investment in equipment, labor, and supervision, and were thus disproportionately the domain of those with capital (cf. Rathbone 1991: 188).

3. Background influences.

With gardens as with most other features of Roman Egypt, disentangling the strands in the tapestry is difficult. Some characteristics are dependent on the physical realities of Egypt, particularly its unusual water regime and its desertic climate; these at least provide some boundaries to what is possible and make some solutions more successful than others (below, section 4). But they scarcely exhaust explanation. Quite a considerable amount is known about gardens in pharaonic Egypt, collected in a book by J.-C. Hugonot (1989). All types of gardens, for example, were highly ordered, with trees planted in neat rows. To be sure, our evidence is heavily biased in favor of monumental settings (palaces and temples, particularly) and of representations that may well show what was favored rather than what was true in everyday life.

The arrival of large numbers of Greeks in Egypt after Alexander's conquest certainly brought change, as settlers tried to replicate aspects of their familiar habitat (see Osborne 1992). But what we know of classical Greek gardens suggests that they already had much in common with Egyptian gardening practice: for example, Greek gardens, too, tended to be enclosed and planted in orderly rows. Given many centuries of connections between Egypt and the Aegean, it is impossible to know how far Egypt may have influenced pre-Hellenistic Greece in such matters. Nor do the architectural analogies of the hypostyle halls in Egyptian temples and the rows of columns found in Greek temples and other public buildings do much to clarify matters. The Greeks certainly brought some new plant types to Egypt, just as they were constantly trying to import new varieties into Greece and get them to grow. In some cases they were successful, in others not. We cannot assume that rare tree or plant types mentioned in papyri of the Ptolemaic period necessarily took root and were common in the Roman period. Balancing that caveat, however, is the likelihood that many common plants are mentioned only occasionally in the papyri, both for the reasons set out above and because the papyri come largely from Middle and Upper Egypt.

Whether the Romans had much influence on the gardens of Egypt under the Empire is hard to say. There was no large-scale Italian settlement in Egypt, and we must therefore be careful not to asssume diffusion of Italian notions of gardening, even supposing that we can identify innovations of the period. All the same, some species or cultivars do appear in the documents only in the Roman period, and it is quite possible that Roman estate managers brought some of them in. Fashion, imitation of the metropolis, may also have played a part.

4. Organization and physical character of gardens.

a. Location.

Most garden plots mentioned in the papyri were located outside the cities and villages, as was the case with other cultivated land. Information about location is generally either imprecise or linked to location descriptors we can no longer place. There was certainly a zone of intensively cultivated land around each city, in which gardens and vineyards were normally found. These were normally small (cf. section 5). If courtyards or vacant building lots inside cities were used for planting, there is little or no evidence of it. The fact that the Egyptians tried as far as possible to locate major settlements on land not usable for agriculture undoubtedly meant that the water supply necessary for gardening was hard to come by in these settlements, and any urban gardens were presumably part of very expensive properties (see Husson 1983: 147) or temples (cf. below, 9). There was in the Roman period a quarter at Oxyrhynchos named "the Paradeisos of Pammenes," attested in many documents. We do not know if this was a public garden, or indeed if it might even have been named after a garden that did not exist any longer. Imprecise wording of a third-century petition prevents us from being certain if it refers to a walled garden or to a garden inside the city walls of Oxyrhynchos; the former seems grammatically more likely.

Villages were also located in barren land as far as possible, but many certainly were inside the watered zone even if on higher elevations.³ Their fabric was also very dense, despite the presence of derelict and garbage-strewn lots. It is possible that a few large houses had trees around a courtyard, but even such "gardens" are not specifically documented.⁴ In general, then, it looks as if the particularities of Egyptian geography overwhelmed any Greek or Roman influences in the location of plantings, whether purely economic or intended both for ornament and for production.

Some of the larger garden properties were located around villages or in more open country. These often tended to be considerable country estates with their own management. Those owned by magnates were sometimes very extensive, although not necessarily territorially consolidated, and had large staffs. The best-known example is the unit of the estate of Aurelius Appianus located near Theadelphia, documented by hundreds of papyri from the middle of the third century CE (for a comprehensive description, see Rathbone 1991). In some cases there were houses on these properties, and such houses may have had gardens of their own (cf. Husson 1983: 147 for one possible such garden).

b. Physical characteristics.

These are known mainly from the language of leases and contracts for labor. Documents vary greatly in the specificity of their information, and the absence of a mention of a feature in a lease may not mean that it was not actually present. Most properties described in extenso were walled; sometimes it is specified that the wall was a πλαστή, a mudbrick wall, or a χάραξ, a stockade (presumably of reeds). Modern Egyptian gardens often have a mudbrick wall topped by a low reed stockade, and such combinations were probably common in antiquity, too.⁵ Larger units, at least, had water sources, whether wells, cisterns or pools, or water wheels (sakiyas) to draw water from distribution canals, or a combination. Such water supplies cost money to construct and maintain, and they are often discussed in the documents.

The most striking characteristic of gardens is the tendency to combine multiple types of planting in a single parcel, even a relatively small one (roughly a quarter-hectare or less). Many documents are incompletely preserved, and the diversity of plantings was thus probably even greater than we are able to see. Sometimes writers use a string of terms (e.g., "vineyard, orchard, and olive orchard" ); on other occasions we find compound nouns ("olive and palm orchard" ). There is no evidence to suggest that these represent different physical realities. The absence of excavation of gardens (almost all located in land continuously under cultivation since antiquity) means that we have almost no evidence about the physical arrangement of the plantings. It is, however, clear that vines were often trained up trees and that crops (sometimes arable, sometimes vegetables or flowers) were planted between rows of trees.

5. Quantitative measures.

Garden plots, even those of an economic character, were relatively small compared to plots used for field crops. A sense of the range typical in the intensive zone of cultivation near cities can be had from P.Panop. 14, an early 4th century CE list of garden plots in a particular section of land near Panopolis. Included are the following:

date palm orchard: 1.28 ha. olive-date palm orchard: 8.08 ha. to .57 ha., mean, mode, and median all .28 ha. rose and vegetable garden: 19.004 ha. to .31 ha., mean .12 ha. vacant sown land: 1.21 ha.

The totals show vegetable garden and rose garden separately (1.93 ha. of vegetable, .31 ha. of roses), but the individual plots are listed as "rose and vegetable" gardens. Evidently the orchard plots were on average about twice the size of the rose and vegetable plots, running around one aroura (.2756 ha.) where the vegetables are usually about a half aroura.

Detached country estates could be considerably larger, justifying an ampler investment in capital equipment. Unfortunately, most of the contracts involving such properties do not preserve size information; in some cases this is the result of damage to the papyrus, but more often the figure never was given. Probably the fact that the rents were not on a per-aroura basis, as is normally the case with arable land, made identification but not sizing of the property necessary.

The small size of many properties does not mean that the total amount of land in such gardens and vineyards was small. We have a few indications that vineyards were planted on a large scale. In the 6th century CE, 5 percent of the land of the Upper Egyptian city of Antaiopolis was devoted to vineyards, amounting to 2,578.5 arouras (710 ha.).⁶ The Fayum was heavily planted in vineyards already in the middle of the Ptolemaic period, as can be seen from the large wine production figures in P.Köln V 221; these suggest that at least 5 percent of the Fayum was in vineyards, and quite possibly more.⁷

6. The range of plants and trees grown.

A large number of different plants are referred to in the papyri or otherwise attested for Egypt. Some of these occur only rarely, but their rarity may indicate in part that they were normally grown in private garden plots that were not the subject of leases, and on which the owners paid a rent in money or in wheat. It is, for example, very unlikely that onions were grown as little as the small number of references in the documents would indicate. All the same, there is a marked difference between the staples of Egyptian gardening and the more exotic species. It is impossible here to give a complete list or discussion of this subject, but there follows a summary list in which the more commonplace items are listed first and distinguished from the rest by italics. See Keimer 1924, Keimer 1984, and Germer 1985 for detailed treatments of Egyptian plants, and Schnebel 1925: 239-315 for the growing of vines and trees.

• Grapes, mainly for wine but also for consumption dried (raisins) and fresh (σταφυλή).
• Trees: Date palm (Phoenix dactylifera, φοῖνιξ καρυωτός = dates); Dum palm (Hyphaene thebaica, Cucifera thebaica, κούκιον);⁸ Persea (Mimusops schimperi, περσέα); Fig, summer and winter (Ficus carica; σῦκον); Sycomore (Ficus sycomorus, συκάμινος συκόμορος); Acacia (Acacia arabica, ἄκανθα); Olive (Olea europaea, ἐλαία); Apple (Malus sp., μῆλον); Pomegranate (Punica granatum, ῥόα); Tamarisk (Tamarix nilotica, μυρύκη); Willow (Salix, ἰτέα); Peach (Amygdalus persica, περσικόν); Pear (Pyrus communis); Citron (Citrus medica, κίτριον); Carob (Ceratonia siliqua, κεράτιον); Sebesten (Cordia myxa, μύξα); Jujube (Zizyphys spini Christi, παλίουρα = Great Jujube); Nuts (κάρυον and καρύα used for various types, incl. walnut, hazelnut, filbert, chestnut, almond); Pistachio (Pistacia vera); Quince (Cydonia vulgaris). Bibl.: Drew-Bear 1995; Kramer 1993.
• Other fruit: Melon (Cucumis melo, κολοκύνθη); Watermelon (Citrullus vulgaris, πέπων)
• Legumes: Fava beans (Vicia faba, κύαμος)
• Other vegetables: Beet root (Beta vulgaris, σεῦτλον); Gourds and squash (Lagenaria vulgaris?, σίκυος); ; Plantains (θρυαλίς); Leeks, sweet (Allium pomum, πράσον γλυκύς); Leeks, headed ("tree onions") (κεφαλώνων λαμπάδες); Lettuce, gardener's = Romaine? (Lactuca sativaθ, ρίδαξ κηπουριακή); Lettuce, wild (Lactuca scariola, μηκωνίς); Cabbage or Cauliflower (Brassica oleracea var., κράμβης καυλός); Kitchen vegetables (ἑψανόν); Asparagus (ἀσπάραγος); Onions (Allium cepa, κρόμμυον); Garlic (Allium sativum, σκόρδον)
• Flowers: Rose (ῥόδον); Laurel? (Laurus nobilis)
• Herbs: Thyme (θυμος)
• Other: Reeds (κάλαμος); Mistletoe (στελεῖς)

The ancients were well aware that some plants and trees would not grow, or would not produce fruit, in particular climates. Except for one variety, pines would not grow in Egypt, and even that one variety appears with great rarity.⁹

7. Regional variation.

Our evidence is not systematic enough to allow any comprehensive statements about ways in which gardens differed in the various parts of Egypt. But individual cases show that such variation existed. There is a considerable climatic difference between the Delta and Upper Egypt, and particularly between the area along the Mediterranean coast and those inland.¹⁰ Extensive wetlands were probably available only in the Delta, and papyrus (not a garden crop) was manufactured mainly in the Delta.

An interesting case study, revealing in many ways about the deficiencies of the evidence, is provided by the Madonna lily (Lilium candidum; Greek κρίνον, λείριον, σοῦσον: the first two of these can also refer to other plants). Lilies were used for making perfume, by steeping them in oil and then squeezing out the oil. The process is known through representations in reliefs from the West Delta of the Saite period, Dyn. 26 (Leahy 1988: 179-94, 292-98). The plant is not native to Egypt and was probably introduced from Syria. There are a few references in the early Ptolemaic papyri. One of these (P.Cair.Zen. I 59009 fr. F.19 = P.Cair.Zen. IV, p.285) refers explicitly to a purchase in Palestine. The others appear as items in accounts from Middle Egypt (P.Petr. III 127; P.Hib. II 271), and it is impossible to tell where the lily oil was produced. There are no references after this time until the 5th century CE, when lily oil is referred to (P.Lond. III 928.12) and a recipe refers to putting lily in wine (PSI IV 297.8). It is likely enough that lilies were grown in the Delta throughout the Roman period; they prefer wet, marshy areas (and are shown with waterfowl in the Saite reliefs). They may also have been grown elsewhere in ponds in gardens. Because of the lack of papyri from the Delta, however, lilies are largely lacking in the documentary record.

If the distribution of documents is to be trusted, λάχανον was grown much more extensively in the Fayum than in the valley, with its frequency rather diminishing as one moves south (Drexhage 1990). Unfortunately, it is not always clear what plant is referred to as lachanon; the sources give a variety of conflicting evidence about its characteristics. The plural, moreover, may encompass more than one plant even if the singular refers to a particular type (lettuce or cabbage, for example).

8. Tomb gardens.

Egyptian funerary practice sought to provide in microcosm whatever was needed for the deceased to regenerate in the afterlife. Water and vegetation dependent on water were certainly central to that vision. In the Nile valley, however, tombs were generally situated in the desert regions, where no water was readily available. Direct realization of a fertile nature there was therefore limited to royalty and the occasional noble; others used imagery, particularly wall paintings (see Hugonot 1989: 169-221). For the Roman period, there is no evidence of garden installations in this zone.

In the Delta, however, conditions were different. Along its edge desert land was available for tombs, but elsewhere it usually was not. Genuine funerary gardens were thus possible. Strabo (17.1.10, 795) says that there were "many gardens and burials" in the western suburb of Alexandria called Necropolis; there is no reason to reject the normal assumption (Fraser 1972: I 26-27, notes II 81-83) that this refers to garden tombs. Such tombs were also found to the east of the city (Fraser 1972: 31-34 with notes). One Augustan document from Alexandria documents the lease of three such tomb gardens near Canopus, a suburb of Alexandria. No information on size or arrangement is provided, but the variety of plants grown shows that they were of some size. An inscription of (probably) the second century CE, and probably from the eastern suburban necropolis zone of Ibrahimiya, mentions another garden tomb. The tombs in Alexandria's extensive necropolis zones were largely underground, and much of the surface should thus have been available for cultivation apart from monuments and entrances into the underground tombs.

9. Temple gardens.

Pharaonic Egypt's most striking monuments other than the pyramids are the temple complexes, especially those of the New Kingdom. Sanctuaries continued to be built along traditional lines in the Ptolemaic and early Roman periods, and although the temples then declined those built under the Ptolemies remained in use for more than two centuries after the Roman conquest. Plantings occurred in Egyptian temples in several locations, documented by texts and excavations (see Hugonot 1989: 21-85): (1) processional ways (dromoi) were regularly lined with trees and flowers along with sphinxes; (2) entries were flanked with trees; (3) there were sacred parks and groves in sanctuaries, mainly in the space between the temple and its enceinte wall; (4) the required sacred lake contained aquatic plants (lotus, papyrus, flowers, etc.) and was bordered by trees; (5) temples had gardens to produce aromatics, herbs, and the like required for religious and medicinal purposes.

These practices undoubtedly continued in the Hellenistic and Roman periods. At the temple of Chnum at Elephantine, trees formed a kind of avenue from the pylon to the sanctuary entrance, remains of which have been found. Illustrations reconstructing some of the sanctuaries of these periods, like Dendera, Ombos, and Hibis, give a general idea of the plantings (see Aufrère, Golvin and Goyon 1991: 222, 236-37, 261; 1994: 76-77, 80). Temples built on desert land no doubt had only a very limited amount of plantings, but trees were nonetheless a characteristic feature.

Evidence for temple gardens is not always clear. An inscription from Koptos, dated to 219 CE, records the reconstruction and painting of a κῆ from its πος foundations, along with its plantings. A companion inscription records the construction and painting of a chapel, ναός, which was probably adjacent. Some scholars have taken this to refer to an actual garden adjacent to a temple, others to refer to a painted representation of a garden or an enclosure with painted walls (see Boyaval 1994; J. Bingen, Chronique d'Égypte 59 [1984] 367). The findspot at Coptos of the inscription is not recorded in Petrie's excavation report.

Under the circumstances prevailing in Egypt and the given the nature of our evidence, a catalogue listing each individual "garden" known in the papyri would be of enormous length but little use, for most entries would contain the barest of skeletons, and the interesting material would be buried under the mass of uninformative data. Instead, included are a selection of gardens about which there is enough information to be of interest, chosen with an eye to representing some of the regional and typological variety known.

Bibliography

• Aufrère, S., J.-Cl. Golvin, and J.-Cl. Goyon (1991). L Égypte restituée 1: Sites et temples de Haute Égypte. Paris. (worldcat)
  • (1994). L Égypte restituée 2: Sites et temples des déserts. Paris. (worldcat)
• Boak, A. E. R. (1933). Karanis: The Temples, Coin Hoards, Botanical and Zoölogical Reports, Seasons 1924-31 (Univ. of Michigan Studies, Humanistic Series 30). Ann Arbor. (worldcat)
• Boyaval, B. (1994). Un jardin d Égypte, Chronique d Égypte 69 (1994) 147-51. (worldcat)
• Drew-Bear, Marie (1995). Le bois en Égypte d après les papyrus d époque romaine, L Arbre et la forêt, le bois dans l antiquité (Publ. de la Bibl. S. Reinach, Univ. Lumière-Lyon 2, VII, Paris) 3-9. (worldcat)
• Drexhage, H.-J. (1990). Λάχανον und λαχανοπῶλαι im römischen Ägypten, Münstersche Beiträge zur antiken Handelsgeschichte 9.2: 88-117. (worldcat)
• Fraser, P. M. (1972). Ptolemaic Alexandria. Oxford, 3 vols. (worldcat)
• Fraser, P. M. and B. Nicholas (1958). The Funerary Garden of Mousa, Journal of Roman Studies 48: 117-29. (worldcat)
• Germer, Renate (1985). Flora des pharaonischen Ägypten (DAI Kairo, Sonderschrift 14). Mainz. (worldcat)
• Hohlwein, Nicholas (1939). Etudes de papyrologie 5 (worldcat)
• Hugonot, J.-C. (1989). Le jardin dans l'Egypte ancienne. Europäische Hochschulschriften, ser. 38, Bd. 27. Frankfurt. (worldcat)
• Husselman, Elinor M. (1979). Karanis Excavations of the University of Michigan in Egypt, 1928-1935. Topography and Architecture (Kelsey Museum Studies 5). Husson, G. (1983). ΟΙΚΙΑ. Paris. (worldcat)
• Keimer, Ludwig (1924). Die Gartenpflanzen im alten Ägypten I (Hamburg-Berlin). (worldcat)
  • (1984). Die Gartenpflanzen im alten Ägypten II, ed. R. Germer (DAI Kairo, Sonderschrift 13). Mainz. (worldcat)
• Kramer, Bärbel (1993). Akanthus oder Akazie? Bemerkungen zu Bäumen, Zeitschrift für Papyrologie und Epigraphik 97: 131-43. (worldcat)
• Kruit, Nico (1992). The Meaning of Various Words Related to Wine, Zeitschrift für Papyrologie und Epigraphik 90: 264-76. (worldcat)
• Leahy, Lisa M. (1988). Private Tomb Reliefs of the Late Period from Lower Egypt. Oxford D.Phil. thesis. (worldcat)
• Osborne, Robin G. (1992). Classical Greek Gardens: Between Farm and Paradise, in John Dixon Hunt, ed., Garden History: Issues, Approaches, Methods (Dumbarton Oaks Colloquium on the History of Landscape Architecture 13) 373-91. (worldcat)
• Rathbone, Dominic (1991). Economic Rationalism and Rural Society in Third-Century Egypt. Cambridge. (worldcat)
• Schnebel, Michael (1925). Die Landwirtschaft im hellenistischen Ägypten (Münchener Beiträge zur Papyrusforschung 7). Munich. (worldcat)

Places

• Africa Proconsularis (late Roman province) (Pleiades:991341)
• Africa (province) (TGN:6000082)