Alexandra Villing, Marianne Bergeron, Giorgos Bourogiannis, Alan Johnston, François Leclère, Aurélia Masson and Ross Thomas
With Daniel von Recklinghausen, Jeffrey Spencer, Valerie Smallwood, Virginia Webb and Susan Woodford
- The Shelby White - Leon Levy Program for Archaeological Publications
- Christian Levett and the Mougins Museum of Classical Art
Share this catalogue
Naukratis: a city and trading port in Egypt
In the 5th century BC the Greek historian Herodotus visited Egypt and described what he saw and learned in the second book of his Histories.
It is here that we find the most important ancient account of Naukratis (Hdt. 2.178–9; cf. the commentary by Lloyd 1975–88, 2007a). According to Herodotus, the establishment of Naukratis as a settlement and trading post was a joint venture by Greeks from twelve different places: Ionians from Samos, Miletos, Chios, Teos, Phokaia and Klazomenai; Dorians from Rhodes, Knidos, Halikarnassos and Phaselis; Aiolians from Mytilene on Lesbos and the people of Aigina, the island close to Athens (see Fig. 1). Naukratis contained their sanctuaries and for a time acted as a privileged gateway for trade between Greece and Egypt:
Amasis became a friend of the Greeks and granted them a number of privileges; to those who came to Egypt he gave Naukratis as a city [polis] to live in. To those who sailed there but did not want to live there he gave lands on which they might erect altars and sanctuaries to the gods. The largest and also the most famous and most used is the sanctuary called the Hellenion; it was founded jointly by cities of the Ionians – Chios, Teos, Phokaia and Klazomenai –, of the Dorians – Rhodes, Knidos, Halikarnassos and Phaselis – and one city alone of the Aiolians, Mytilene. It is to these cities that the sanctuary belongs, and it is they who appoint the chief officers of the port [προστάται τοῦ ἐμπορίου].
Figure 1. Map of the Eastern Mediterranean showing Naukratis and its founding cities. Drawing Kate Morton and Ross Thomas.
Whichever other cities lay a claim to this, they claim for themselves nothing they have a share in. Besides these, the Aiginetans built a sanctuary of Zeus by themselves, and the Samians one of Hera, and the Milesians one of Apollo. Naukratis was in the past the only trading port in Egypt. Whoever came to any other mouth of the Nile had to swear that he had not come intentionally, and had then to take his ship and sail to the Canopic mouth; or if he could not sail against contrary winds, he had to carry his cargo in barges around the Delta until he came to Naukratis. In such esteem was Naukratis held.
For anyone reading this description, seeing the remains of the site today (Fig. 2) presents a severe challenge to the imagination: with no visible signs of ancient ruins and in a remote, rural location, it is hard to imagine that this was once the busy, thriving port city that Herodotus describes.
The site of ancient Naukratis is today located near the modern villages of Kom Ge’if, Nebira and El-Neqrash, 16km in direct line west of the Rosetta branch of the Nile, and 15km east of the Libyan plain, by the side of which the canal of El-Habir flows. It has no direct link to the Nile, yet in antiquity was on the (eastern) bank of the Canopic Nile branch, one of the major waterways linking the Nile valley with the Mediterranean sea, the mouth of which appears to have been the most accessible of the Nile’s branches in antiquity (cf. Stanley and Warne 2007; Cooper 2008). Located thus on the way to the capital city of Memphis south of the Nile Delta’s apex, Naukratis was in all likelihood also once connected, via a canal, with the city of Sais 19km to its north-east, seat of the Pharaonic court during the 26th (Saite) dynasty, for which it presumably functioned as its international port. 
In addition it was not far from the important salt and natron sources in the regions of Barnugi (25km west) and Wadi Natrun (70km south), on the likely route of shipment via the river port of Terenouthis, modern Tarrana/Kom Abu Billu, some 65km upstream on the Canopic branch (Bernand 1970, 625), while its rich agricultural hinterland provided ample opportunity to grow grain for export.
Figure 2. The site of Naukratis in 2011, looking south over the partly dried-up lake now covering much of the central part of the ancient town. Photograph Ross Thomas
1. The emporion Naukratis: trade between Greece and Egypt
With its strategic location – well-connected, close to Sais and Egyptian control – it is understandable that under the Saite dynasty Naukratis, together with its sister port of Thonis/Herakleion located downstream at the mouth of the Canopic branch (see below), functioned – as a port-of-trade – as one of Egypt’s main interfaces with the Mediterranean (Möller 2000a; Bresson 2000, esp. 62; Pfeiffer 2010). For a time at least, it was the Egyptian hub in a network of trade routes that linked the whole Mediterranean in the Archaic period, from East Greece to the Phoenician coast, Cyprus, mainland Greece, the Nile Delta, North Africa, Sardinia, Etruria and Spain. Indeed, the Greek cities involved at Naukratis are all well known as major trading and seafaring powers in the Archaic period (see the summary in Möller 2000a, 75–88). Their maritime and commercial exploits are epitomized in the story of the intrepid sea captain Kolaios from Samos, who, on a trading voyage to Naukratis, is blown off course and winds up in Tartessos in Spain, hitherto entirely unknown to Greeks. The fortune he made along the way was, Herodotus claims, only surpassed by that of Sostratos of Aigina – yet another trader from one of the founding cities of Naukratis (Hdt. 4.152).
Naukratis’ sister port was the harbour town of Thonis/Herakleion. It controlled and guarded (not least perhaps against marauding Greeks and other foreigners) the mouth of the Canopic branch of the Nile as it entered the ‘sea of the Greeks’, the Mediterranean. The two sites can thus be seen as forming stages in a series of (trading) ports along the Nile, controlling the river at strategic points and passing goods to the capital city or cities, with Naukratis, however, clearly the main foothold and residential base for Greek traders. Until very recently the site had been known only from a few literary and epigraphic sources, one of which interestingly mentions the site as an emporion, just like Naukratis (Diod. 1.9.4; Strab. 17. 1.16; cf. Hdt. 2.113). But detailed underwater geophysical survey and excavations since 1996 have now uncovered large parts of this important site in the bay of Aboukir (Goddio 2007, 2008; Bomhard 2008; Thiers 2009; cf. already Toussoun 1934). They have revealed extensive canals and harbour infrastructures as well as a main temple to Amun-Gereb and his son Khonsu (equated by the Greeks with Herakles: see Höckmann 2010), and in the harbour and canals, 62 (mostly Egyptian) sunken ships (or ship hulls) mostly of the 6th/5th to 3rd/2nd centuries BC.
Finds from the site suggest that Thonis/Herakleion would have been the first port of call for ships en route to Naukratis from the 7th century BC, i.e. from the likely time of its foundation. It clearly was a major Egyptian cultic centre, yet also a centre of administration for trade and taxation. The twin stelae erected by Pharaoh Nektanebo I in 380BC in both Naukratis and Thonis/Herakleion specify that one tenth of the taxes on imports passing through the town of Thonis/Herakleion and one tenth of the taxes on transactions at Naukratis (Per-Meryt) were to be given to the sanctuary of Neith of Sais, the main goddess in the royal city (Yoyotte 2001; Agut-Labordère 2012, 355–8; Bomhard 2012). Whether such double taxation already applied in periods earlier than the 30th Dynasty remains unclear. Imported finds from the site certainly go back to the 7th century BC and include East Greek and Phoenician/Levantine trade amphorae, Greek and Corinthian fine-ware pottery, as well as Achaemenid Persian finds (Goddio 2008, 228–31; Grataloup and McCann 2006; Grataloup 2010). However, the majority of the weights (Athenian more than Egyptian) date from the 4th century BC, and also the local minting of Athenian-style coinage does not appear to pre-date the late 5th to early 4th century BC (van der Wilt 2010; Bowman 2010, 103).
Figure 3. A wide variety of foodstuffs and other commodities could be transported in pottery amphorae: traces of figs were discovered on this fragment of a 6th century BC Ionian amphora from Naukratis (see Stacey et al. 2010). British Museum, 1886,0401.968
What kind of trade and what traders would have passed through Thonis/Herakleion and Naukratis? Among the earliest visitors and traders were, it seems, early voyaging Greek aristocrats – such as Sappho’s brother Charaxos, reported to have sailed to Naukratis with a load of wine from Lesbos (certainly popular in Egypt as finds of amphorae attest) as a means of financing his ‘sightseeing’ voyage (Strabo 17.1.33; Hdt. 2.182). The Athenian statesman Solon, too, is said to have travelled to Egypt ‘both on business and to see the country’ (Ar. Ath.Pol. 11.1; cf. also Hdt. 1.29). Some centuries later Plato supposedly traded in olive oil so as to finance his voyage in Egypt (Plut. Solon 11.4), though this may well be no more than an anecdote. In general, however, the increase of dedicated directional trade, by ‘professional’ traders, from the last third of the 7th century onwards, may well be reflected in the flourishing of Naukratis at this period, which Greek traders from a variety of backgrounds now made their permanent base (Möller 2000a, 196–203; Bresson 2000, 60–2). Other foreign traders may have joined them as well. Diodorus (1.68.8) notes that Greeks and Phoenicians were the main traders admitted into Egypt since the time of Psammetichos (Psamtek) I. Phoenicians were more likely to have been active in the Eastern part of the Delta, but, along with Cypriots – attested with certainty at Naukratis in the Classical period – may also have frequented Naukratis (Schlotzhauer 2006a; Villing and Schlotzhauer 2006b, 7 with n. 51).
Who was the trade aimed at? Some of the goods that went from the Mediterranean to Naukratis and beyond may well have been intended to supply foreigners in Egypt – such as the foreign mercenaries dotted around the country. But there was also clearly a genuine flow of goods and commodities destined for local Egyptian consumption.
Phoenician/Lebanese wine and wood from Lebanon had been imported into Egypt since the 2nd millennium BC (Möller 2000a, 29), with imported wine certainly playing a part in elite consumption (Meeks 1993; Wilson and Gilbert 2008). As mentioned earlier, from the mid to late 7th century BC onwards, we find Greek trade amphorae – including Samian, Lesbian, Chian, Klazomenian and Milesian – but also Phoenician and Cypriot trade amphorae in many parts of Egypt, and it seems likely that the majority of them (but see Fig. 3) would originally have contained wine (and some oil – both commodities also being mentioned in Greek literary sources, see above), even if many were later re-used for other purposes (see especially the various contributions in Marchand and Marangou 2007). Greek fine-ware pottery, by contrast, appears to have played virtually no role as an import into Egypt proper (Weber 2007; Schlotzhauer and Weber, forthcoming).
Figure 4. Bronze stamp seal of the Persian period with the name of an official in Aramaic, which may have been used e.g. for stamping the mud sealings of amphorae. British Museum, 1886,0401.1706
Yet archaeology can also be misleading: many of the goods that were the staples of ancient trade – such as grain, wood, textiles or many natural resources – were perishable and not traded in pottery containers, and are thus not reflected in the surviving archaeological evidence. However, we are fortunate enough to have a few written documents that record some of the goods that were shipped between the Mediterranean and Egypt. One is the Aramaic tax register of an unnamed port found at Elephantine (where the papyrus had been re-used), most likely dating to about 475 BC, the time when the Persian king Xerxes reigned over Egypt (cf. Fig. 4). It records Greek ships from Phaselis, one of the cities involved in the founding of Naukratis, arriving in Egypt with a cargo of primarily wine and oil, and Phoenician ships (probably from Tell Ghazza) carrying Sidonian wine, cedar wood, bronze, iron, wood, Samian earth, tin (possibly) and building materials to Egypt (Bresson 2000, 67–9; Tal 2009; Villing 2013). This information is complemented by the decree written on the twin stelae erected by Pharaoh Nektanebo I in 380 BC, which mentions imports of gold, silver, timber and worked wood (Yoyotte 2001; Agut-Labordère 2012; Bomhard 2012).
Quite possibly, however, what was exported from Egypt was at least, if not more, important than what was imported. Raw materials such as, perhaps, alum (Bresson 2008, 150; van Alfen 2002; cf. Hdt. 2.180) and certainly natron – used in glass and faience production (and mummification) (Shortland et al. 2006) – played a major role here. In fact, natron is the only cargo that the Elephantine tax register records as being exported (Briant and Descat 1998, 95–7). Is this because natron was really the only product carried back by the Greek and Phoenician ships, or was it merely the only export product that was taxed and therefore recorded? At any rate, the fact that natron features so large in the tax register may suggest that the unnamed trade port to which the register refers could in fact be Naukratis, located not too far from the main Egyptian natron sources at Barnougi, and en route from the likely shipping port of the natron sources of the Wadi Natrun at Terenouthis, upstream from Naukratis on the way to Memphis (Bernand 1970, 625; Bresciani 1995, 108; 1996; Shortland et al. 2006, 523; Boardman 2013).
Grain was another important factor. Certainly from the early 5th century BC onwards, and probably already earlier, Egypt – along with the Black Sea region and Sicily – was a major source of grain for Greek cities, essential for securing sufficient supplies of grain particularly in times of crisis. In the 5th century the city of Teos, one of the founding cities of Naukratis, passed a law placing a curse on anyone obstructing its grain imports, suggesting that grain imports via Naukratis could be of vital significance (Möller 2000a, 81). Grain may well have been a major incentive for Greek interest in trade with Egypt from the very beginning – although it remains a matter of debate among Greek economic historians to what extent it was subsistence commodities such as grain, rather than luxury goods or semi-luxuries that were the mainstay of regular trade (Foxhall 1998; Morley 2007, 42–3). Certainly numerous other goods were also imported from Egypt to Greece, such as papyrus and flax/linen, perfume and linseed oil, calcite alabastra, Egyptian or Egyptianizing amulets (again partly produced at Naukratis itself) and foodstuffs such as salt or Nile fish (Möller 2000a, 211–12; Briant and Descat 1998, 97; van Neer et al. 2004, 130–3 cf. also Boardman 2011).
Figure 5. Fragment of Ptolemaic mould for terracotta figurines of an old man, 2nd–1st century BC. British Museum, 1886,0401.1542
In the Ptolemaic period Naukratis received (wine) amphorae from a number of Greek cities of the Aegean Sea, particularly Rhodes, Knidos and Kos, but also Miletos, Attic and other black glazed Greek pottery, and figurines as well as marble for sculpture. The imports appear to match what has been found in Alexandria, though their quantity and range during this period are less than that of earlier periods. It is possible that direct trade links decreased. New research into the material by Ross Thomas as part of the current project indicates that whilst in the Late Period about half of the terracotta figurines and a significant proportion of the lamps were brought to Naukratis from East Greece, Cyprus and Greece, the material from the Ptolemaic levels is almost all locally made (Fig. 5; cf. Thomas forthcoming). Still, whilst Alexandria had become the main entrepôt for Greek imports into Egypt, Naukratis’ location on the Canopic branch of the Nile and at the beginning of a canal to Mareotis (and by extension Alexandria’s Nile harbour Kibotis – following Blue and Khalil 2011, 9–10 and their interpretation of Strabo 17.1.22–3), meant that the settlement could benefit from the river trade between Alexandria and the rest of Egypt, as well as from the expanding industrial production in the Alexandria–Mareotis region that included glass, textile, papyrus, pottery and wine (Blue and Khalil 2011, 1).
Meanwhile, imports from Republican Rome gradually increased in importance and included amphorae from Brindisi and other Italian cities, and with the annexation by Rome, fine red-slipped terra sigillata. In Roman times imports from Turkey, Cyprus, Syria, Greece, Italy and North Africa continued to find their way into the city. Following a decline in the later 3rd and early 4th century AD Naukratis recovered, again importing North African, but increasingly (like Alexandria and Mareotis) Cypriot and Turkish tablewares and amphorae. It is during this late 4th or 5th century resurgence that the Canopic branch is thought to have started silting up (Blue and Khalil 2011, 9). Such silting would have required increasing efforts to maintain effective canals to Alexandria for transport and water. In spite of this Naukratis’ position at the start of a major canal to Alexandria may explain its resurgence in the 5th century AD.
As mentioned above, rather than merely channelling goods, Naukratis also produced some popular export goods itself. This is certainly true for calcite (‘alabaster’) alabastra (perhaps once filled with Egyptian perfume – cf. Athen. 12.553d–e; Bäbler 1998, 69–77), from the production of which Petrie found numerous discarded drill cores in his excavations at the site (Petrie 1886a, 15), and of course for faience scarabs and amulets produced at the site in the late 7th to early 6th century BC (Gorton 1996, 91–131; Möller 2000a, 152–4; Hölbl 1979, 141, 207–9; 2005). The terracotta figurines and lamps produced at Naukratis, by contrast, appear to have been made primarily for local or regional consumption. Their production seems to go back to the Archaic period and is confirmed by finds of moulds (albeit from later periods), with standards of production comparing reasonably well with products of other areas. Pottery, too, was locally produced, at least from the 6th century BC onwards. Among the earliest preserved examples are pieces painted in an East Greek style, while later on a local version of the Nile silt ‘black ware’ and other fabrics are attested (Schlotzhauer and Villing 2006, 62–5; Mommsen et al. 2006, 71; Dupont and Thomas 2006, 81; Schlotzhauer and Weber forthcoming). A potter from Naukratis is mentioned in a Ptolemaic papyrus (Redon 2012, 81 no. 17) and Athenaios (11.480d–e) refers to the presence of many potters in the city, as well as to a gate named Keramike. While a painted amphora of the Archaic potter’s workshop at Naukratis was found as far afield as Tell Dafana, there are no signs, at least to date, to suggest that locally produced wares were particularly widely distributed, or indeed of exceptional quality – making the reference in Athenaios (ibid.) to the local production of excellent cups (kylikes), coloured to look like silver and shaped like phialai with a broadly extended base and four handles (‘ears’), stand out as an inexplicable oddity.
2. Early Naukratis between literature and archaeology
One of the main results that emerged from Petrie's fieldwork at Naukratis was an earlier date for the foundation of the site than had previously been assumed. As Petrie realized after the first season’s fieldwork, Naukratis must date back to the late 7th century BC, not just to the time of Amasis, as one might be tempted to think (though would not necessarily have to conclude) from glancing at Herodotus. Even though Petrie’s dating of the site was by no means immediately or universally accepted (Hirschfeld 1887; cf. see Leonard 1997, 17–19) and still remains debated (esp. James 2003, 2005; cf. also Bowden 1986, 26–8; Agut-Labordère 2012, 360 with n. 19; Lloyd 1975–88, vol. 3, 222–4), a pre-Amasis date of the foundation is today commonly accepted. Historically, it certainly makes sense if seen against the background of Psamtek (Psammetichos) I’s Greek and Carian mercenaries and the military support he received from the Kingdom of Lydia, in the hinterland of the Ionian cities in the late 7th century BC (Fantalkin 2014). Most importantly, however, such a date is supported by historical dates for Archaic Greek pottery in archaeological contexts in the Levant (e.g. Stager et al. 2011; discussed in detail by Schlotzhauer 2012), which independently confirm a (late) 7th century dating for the earliest pottery finds from Naukratis (Figs 6 and 7).
Figure 6. Fragment from Early Corinthian black figure aryballos from Naukratis, c. 620–590 BC. Ashmolean Museum, Oxford, 1896-1908 G.120.30. © Ashmolean Museum, University of Oxford. Photography by British Museum staff
A foundation under Psamtek (Psammetichos) I or soon after is, finally, also suggested by some historical sources, notably the Roman geographer Strabo. According to Strabo 17.1.18, ‘in the time of Psammetichos (who lived in the time of Kyaxares the Mede) the Milesians, with thirty ships, put in at the Bolbitine mouth, and then, disembarking, fortified with a wall the above-mentioned settlement [‘Milesian Wall’]; but in time they sailed up into the Saite Nome, [having?] defeated Inaros in a naval fight, and founded Naukratis, not far above Schedia.’ In addition to its chronological implications, the passage is noteworthy also for the questions it (and some other literary references) raises over the nature and date of the early foundation and the role played by Miletos, all of which has been much debated in scholarship (see e.g. Haider 1988, 184–99; Drijvers 1999; Möller 2001a; Quack 2006; and notably the recent reinterpretation by Carrez-Maratray forthcoming).
On balance, the archaeological evidence gives no clear chronological (or other) precedence to Miletos, even if it is becoming increasingly clear that the city does have a more significant stake in the early archaeological record of the site than hitherto admitted (e.g. Kerschner 2001; Schlotzhauer in Schlotzhauer and Weber 2012 forthcoming), nor can the site’s establishment be dated significantly earlier than the final third or quarter of the 7th century BC. A consensus has also emerged regarding the unlikelihood of the establishment of a settlement purely on Greek initiative, against Egyptian resistance. Strabo’s account is most likely coloured by a later desire – attested also otherwise – to emphasize the notion of Miletos as a great colonising power, as well as to embellish the historical record and credentials of Naukratis (Möller 2001a; Redon 2012, 63–9).
Figure 7. Fragment from rim of Milesian plate from Naukratis, late 7th¬–early 6th century BC. British Museum, 1888,0601.544.f
Amasis may nevertheless have played a vital role for Naukratis (though see the arguments by Fantalkin 2014), reorganizing the site, establishing additional traders and temples and granting the administration of the site to the προστάται τοῦ ἐμπορίου. As many scholars have argued, such reforms may have been intended both to foster trade and alliances with certain cities while possibly minimizing the influence of others, and to keep tight control of foreign traders, an aspect that may have been of consideration particularly following the nationalist backlash that followed the reign of Apries (589–570BC) (Pébarthe 2005, 172; Bresson 2000; Bresson 2005; Carrez-Maratray 2005; Herda 2008, 49).
3. Cult, feasting and pottery
Following its establishment, and throughout the 6th century BC, the city and port of Naukratis clearly flourished. Traders and travellers, mostly Greek but also Cypriot and presumably Phoenician, came together here to do business and to attend elaborate feasts in the sanctuaries, which saw their heyday in the Archaic period. If we believe the literary sources, Archaic Naukratis was the playground of exceptionally beautiful hetairai (Hdt. 2.135), alluded to also by the poetess Sappho (fr. 7 Lobel – Page), whose brother Charaxos supposedly fell in love with a local hetaira, Doricha, or even the famous Rhodopis (Möller 2000a, 199–200). Attempting to explain the origin of the legendary ‘Naukratite wreath’ that was worn by symposiasts, Athenaeus (15.675f–676c, quoting Polycharmos) refers to one such feast organized by the Naukratite sailor Herostratos following his rescue at sea by Aphrodite and upon his safe arrival at Naukratis: ‘And having sacrificed to the goddess and dedicated the statue [previously bought at Cypriot Paphos] to Aphrodite, he invited his relations and closest friends to a feast in the temple itself, giving to each a wreath made from the myrtle, which even at that time he called a Naukratite wreath.’
Figure 8. Laconian black figure pottery kylix from the sanctuary of Apollo at Naukratis, 570–560BC. British Museum, 1886,0401.1063
The rich finds of high-quality late 7th- to 5th/4th-century BC Greek symposium pottery in the sanctuaries of Naukratis (Fig. 8) – studied for the present project by Marianne Bergeron and others – bear direct witness to such sacred feasting at Naukratis right from the end of the 7th century BC. Communal ritual meals must have been of particular importance in a place like Naukratis, reinforcing a communal spirit and identity, and enlisting the gods’ help for the success of voyages and trade ventures. The profile of the Greek painted pottery finds in Naukratis (cf. Möller 2000a; Kerschner 2001; Schlotzhauer and Villing 2006) in general supports the literary account of who was involved in the emporion – even if it also illustrates the fact that the equation ‘pots = people’, here as elsewhere, does not quite work out (cf. Schlotzhauer and Villing 2006; Villing forthcoming a). Pottery from Ionia (north and south) and Aiolis is much in evidence, far more so (as far as we can judge from the preserved asemblage) than anything that might reflect the Dorian involvement of Rhodes, 
Knidos, Halikarnassos and Phaselis, and there is only very little Carian pottery. Greek pottery from elsewhere – especially from Attica but also from Corinth and Laconia – rather than necessarily suggesting visitors from these places, reflects the general pattern of pottery distribution in the Archaic Mediterranean, with Attic pottery in particular presumably, at first, carried by Aiginetans. Also the few pieces of Etruscan bucchero pottery were most likely brought by East Greeks or Aiginetans, even if later, in the later 4th century BC, a votive graffito actually attests a Syracusan at Naukratis. Close connections with the homeland are indicated, for example, by the import of specially made sanctuary pottery with dipinti to Hera of exactly the same type as have been found in large numbers in the Samian Heraion (Schlotzhauer 2006a, 311–13), while finds of numerous early Cypriot figurines confirm the importance of Cyprus in trade networks. All together these examples testify to the attraction Naukratis held as the gateway to the ancient civilization of Egypt.
4. Naukratis: the later history
Scholarship on Naukratis, particularly from an archaeological perspective, has traditionally focused on these earliest phases of the site. One of the site’s early excavators, Ernest Gardner, claimed that ‘almost all the things found at Naukratis date from an earlier [pre-Ptolemaic or even pre-Classical] time’ (Gardner 1888, 27), and even though this is clearly something of a misrepresentation, even when looking at Petrie’s and Gardner’s own publications (note, for example, the ‘abundance of Roman pottery’ reported by Petrie 1886a, 13, as originally covering the temenos of Apollo), a focus on the Archaic period has dominated scholarship on Naukratis through the ages. Partly this is owing to the fact that as far as Greek painted pottery at least is concerned, Gardner’s statement was not entirely incorrect, with the majority of the (preserved) Greek fine-wares dating to the 6th century BC. One has to consider that by the time Petrie and his successors began working at Naukratis, the sebbakhin in many parts of the site were just reaching the Classical–Archaic levels, having cleared away later evidence, thus leaving less of the earlier periods to be discovered by the archaeologists. Nevertheless, both the results of the later American fieldwork at the site and literary and epigraphic sources clearly indicate that Naukratis continued to flourish in later periods, even if historical developments occasioned a variety of changes in its fortunes across time.
Figure 9. Faience libation vase from the foundation deposit of the gate (pylon) of the Egyptian Temple enclosure, built by Ptolemy II (285–246 BC). British Museum, EA 58329
One major change for Naukratis was the gradual decline in its role as an ‘international’ port of trade. The surviving assemblage of Greek pottery – the vast majority in all likelihood dedications by traders both resident and passing – indicates that convivial feasting clearly continued across the watershed that was the Persian conquest in 525 BC for Egypt, and throughout the Classical period, albeit somewhat reduced. The two stelai erected in 380 BC by Nektanebo I at Naukratis and Thonis/Herakleion (Yoyotte 2001; Bomhard 2012), attest to significant trade passing through the site still in the 4th century BC. In addition, late 5th to 4th century BC inscriptions record traders from Naukratis being elected as proxenoi of Greek poleis and major religious centres such as Athens, Ios, Kos, Delphi and Delos (Bresson 2000, 61–2, 67; cf. Redon 2012. 63–4).
As mentioned earlier, following the foundation of Alexandria trade along the Canopic branch was once more boosted, with the Canopic branch allowing access through Lake Mareotis to Schedia and Alexandria (Blue and Khalil 2011), even if the spotlight soon moved to Alexandria. It was a native of Naukratis, Kleomenes, who, as Alexander’s financial officer, governed Egypt from 331 BC and after Alexander’s death briefly acted as hyparchos under Ptolemy I, until the latter had him executed in 322 BC (Le Rider 1997). Naukratis now almost certainly had a status as a Greek polis. Although this term is first unequivocally attested for the 2nd century BC, it must go back to an earlier date, possibly to the Archaic period, though maybe not necessarily to before the late 4th century BC (for an early dating, see e.g. Herda 2008, 46–51; for a later one, e.g. Möller 2000a, 184–91; Bresson 2000, 74–9).
Figure 10. Fragment of a handle from a Thasian transport amphora with a stamp of a seated sphinx, 380–360 BC. British Museum, 1955,0920.62
At this time, Naukratis also briefly issued its own bronze coinage, presumably under Kleomenes (Le Rider 1997, 91–3; Bresson 2000, 75). The city’s administration, at least from the Hellenistic period, centred on the prytaneion, where feasting for Apollo Komaios (perhaps the main polis god: Herda 2008, 46–51) also took place. Bouleutai, timouchoi and at least four phylai are attested. The use of the Milesian calendar in Hadrianic Antinoupolis, which modelled its laws on those of Naukratis, suggests that this calendar might also have been used in the latter city (see Athen. 4.149d–150a; Herda 2006, 143–50; 2008, 46–51; Scholl 1997).
A major programme of embellishment appears to have taken place in the earlier part of the 3rd century BC, when Ptolemy I (323–283 BC) erected a new temple to Amun-Ra in the Egyptian temple enclosure, and his son Ptolemy II (283–246 BC) the precinct’s new massive pylon (Fig. 9; cf. Leclère 2008, 131–5). Successful traders are still attested in this period too – though they now feature in Egyptian inscriptions: Nekhtnebef, son of Tefnakht (Jansen-Winkeln 1997; cf. Agut-Labordère 2012, 370), as well as Horemheb, son of Krates and Shesemtet. The latter is represented by an impressively – and exceptionally – huge statue of late 4th to early 3rd century BC date and entirely Egyptian type, even though its inscription designates its owner as a Greek (Yoyotte 1994–5, 672–3; Agut-Labordère 2012, 369).
The large number of imported stamped amphorae (studied by Alan Johnston) from Greece (notably Rhodes) and, increasingly, Italy found at Naukratis from the early 4th century BC (Fig. 10) to the end of the Hellenistic period (and beyond) is a strong indicator that Naukratis continued to receive imported goods from the Mediterranean on a considerable scale throughout this period, though this is not necessarily unusual for sites in Ptolemaic Egypt; it is quite possible that these increasingly reached Naukratis via Alexandria, rather than through direct trade with the source. By the mid-Ptolemaic period, Naukratis was certainly still a busy regional, but perhaps no longer international trading hub.
Figure 11. Flask with the Greek ευλογία (blessing, gift or alms) written on the shoulder, possibly a reference to the Eucharist, 6th century AD. British Museum, OA.9431
Naukratis continued to be occupied through the Roman period and well into the Byzantine period at least until the 7th century AD. Imported and local pottery, lamps, glass, coins, jewelry, stamps and seals, games, figurines and sculpture dating from the late 1st century BC to the 7th century AD have been found in all seasons of excavation as well as the survey at Naukratis. The settlement seems to have been in slow decline from the late Ptolemaic period until the 2nd century AD, by which time the geographer Claudius Ptolemy in his Geography (4.5) describes the site as a town, and no longer a city or polis: ‘Sais Nome and metropolis: Sais 61°30' E 30°30' N. On the Great river toward the east: Naukratis town 61°15’ E 30°30’ N.’ Naukratis retained some status, however, as games were performed there in the 3rd century AD. They are mentioned in a papyrus document found in Oxyrhynchus that dates from c. AD 226 to 230 (P.Oxy. XXII 2338) that lists victorious poets, trumpeters and heralds at the Naukratis games.
The archaeological material suggests that a significant decline occurred in the mid 3rd century AD, lasting well into the 4th century AD, possibly related to economic and demographic difficulties in Egypt and the wider Roman Empire at this time. A resurgence in the late 4th and 5th centuries AD made the settlement significant enough for Naukratis to have a bishop based there in 459 AD (Timm 1988, 1749; Munier 1943, 23), and Christian motifs and prayers can be recognized on seals and pottery (Fig. 11). The settlement continued until the 7th century AD, after which it appears to have been abandoned, with no artifacts that can be conclusively dated to the early Islamic period.
5. The people of Naukratis: debating the Egyptian presence
Figure 12. Donation stele for Amun-Re Baded, reign of Apries (589-570 BC). Berlin, Ägyptisches Museum und Papyrussammlung. Inv.-Nr. ÄM 7780. Foto: Fotoarchiv.
One of the issues that has long been a bone of contention, particularly between Classicists and Egyptologists, is the question of the date and extent of Egyptian Naukratis. Was Naukratis a Greek foundation on virgin soil, to which later, in the 4th or 3rd century BC, an Egyptian ‘suburb’ and sanctuary was added? Or was it the ‘Greek suburb’ of a pre-existing Egyptian town? Scholarship on this topic reveals a significant (ideological) divide between the disciplines, with many Classicists favouring a foundation on ‘virgin soil’ (e.g. Möller 2000a, 118–19: ‘Naukratis was a purely Greek empórion, the assumption of an Egyptian quarter being based on misinterpretation’; see also Möller 2001a, 5–11), and Egyptologists an earlier Egyptian presence exerting firm control over the foreign settlers (cf. Yoyotte 1993/4; Leclère 2008, 113–47, esp. 121: ‘C’est au nord de ce noyau urbain égyptien que s’est développé l’établissement grec’). Relatively little work, however, has to date been done on Egyptian Naukratis and particularly on the Egyptian finds from the site, thus leaving much open to speculation.
The fact that Herodotus does not mention anything Egyptian in his description of the site might create the impression that Naukratis was a purely Greek city at least in the Archaic and Classical periods. However, this is contradicted by Egyptian epigraphic evidence (summarized by Leclère 2008, 117–18, cf. also Bresson 2000, 66–7). Already in the third quarter of the 6th century BC an Egyptian inscription clearly refers to Nokradj, the Egyptian name of Naukratis, naming an Egyptian resident there, while another inscription notes the renewal of a donation connected with the temple of Amun-Ra of Baded, located at Naukratis (Fig. 12). 
The large Egyptian temple precinct discovered at the site thus must date back to the early 6th century BC at least, even if the surviving securely datable elements all seem to be connected with rebuilding phases in later periods. A further Egyptian name of Naukratis, Pi-emroye or Per-Meryt, ‘the Harbour/Port House’, occurs in later documents, most famously on the two stelai erected in 380 BC by pharaoh Nektanebo I in Naukratis and Thonis/Herakleion (Yoyotte 2001; Bomhard 2012).
Figure 13. North Ionian large bowl with a painted inscription (dipinto) on the inside naming ‘Aphrodite at Naukratis’ [ο[...]τηι: τηι εν Ναυκρατι], c. 575–550 BC. British Museum, 1888,0601.531.
Per-Meryt also appears to have been applied to other places in Egypt and is a reference to the town’s function (Leclère 2008, 117 n. 24; Agut-Labordère 2012, 359–60). By contrast, the name Nokradj – etymologically of Libyan ancestry – is unique and appears to be the primary name of the site. For linguistic reasons, as Egyptologists point out, it also must be the original name of the site, from which the Greek name ‘Naukratis’ (the earliest known occurrence of which is a painted inscription on an early 6th century BC pottery bowl: Fig. 13) derives – hence giving precedence to an Egyptian foundation of Naukratis (esp. Yoyotte 1991–2, 640–1; Leclère 2008, 117 n. 24; contra Lloyd 1975–88, 222; Sullivan 1996, 189; Möller 2000a, 195).
Figure 14. Small bronze relic box or coffin of a snake from Naukratis, once containing parts of a dead snake, decorated with a coiled snake on the top, date unknown. Bolton Museum, 1886.31.69.b. Photograph © Bolton Library and Museum Service. Photographer François Leclère, The British Museum.
To date it has proved difficult to thoroughly assess these arguments with the help of the site’s archaeology, which remained poorly understood particularly as far as Egyptian evidence is concerned. Hardly any of the once abundant Egyptian pottery survived the early excavator’s find selection (see below). A small number of securely dated and inscribed finds of Saite or even earlier date were assembled and discussed by Jean Yoyotte (Yoyotte 1991/2; 1993/4, 1994/5; cf. Leclère 2008, 116–19; Agut-Labordère 2012)as part of his case for an early Egyptian presence at the site, but the vast majority of surviving Egyptian finds remained barely studied (for bronzes see recently Weiss 2012, 442-6). All this resulted in a picture of Naukratis that was heavily biased towards the Greek. It is only now as part of the present project that the Egyptian material from early fieldwork at Naukratis has been fully catalogued, and that the size and significance of the Egyptian element in the site's material culture can be appreciated (Figs 14 and 15).
Figure 15. Seal of Amasis-Sa-Neith, a high official at the court of Pharaoh Amasis (570–526BC), found at Naukratis. British Museum, 27574
The only other fieldwork conducted at Naukratis, by the American mission under Coulson and Leonard in 1977–83 (Coulson 1996; Leonard 1997, 2001; cf. also the discussion below) did little to solve the outstanding questions on early Egyptian Naukratis. The mission did in fact uncover a certain amount of Egyptian Late Dynastic pottery, notably in the area of the ‘Great Temenos’, but failed to fully recognize and adequately acknowledge it (Spencer 2011; Thomas and Villing 2013; see also below), hence perpetuating the misconception of a pre-Ptolemaic ‘void’ of Egyptian evidence.
Judging from written sources and from the archaeological evidence as it presents itself today, biased as it may be, it is clear that Naukratis was not just a Greek but also an Egyptian town, with the Egyptian presence at the site contemporary with the Greek – though not necessarily any earlier. The town had close connections with the then royal capital city of Sais (Fig. 15), in the nome (administrative precinct) of which it lay and the main temple of which, that of Neith, received a tithe from tax income raised at Naukratis. The local temple of Amun-Ra presumably functioned as the centre of Egyptian administration of Naukratis, which ultimately would have been under the jurisdiction of officials such as the ‘Overseer of the gate of the Foreign Lands of the Great Green’ (i.e. the Mediterranean), officials who since Saite times would have been in charge of securing the borders as well as administrating trade taxes (Pressl 1998, 70–3; Posener 1947; cf. also Vittmann 2003, 220–1, fig. 111; Carrez-Maratray 2005, 202–3).
The population of Naukratis was thus undoubtedly ethnically diverse from the outset, accommodating at certain points in time, in addition to Greeks and Egyptians, also Cypriots, Persians, Macedonian Greeks and quite probably others. The site’s Greek residents would have included traders, craftsmen, sailors, cult officials, hetairai (if we believe the literary sources) and perhaps retired mercenaries; its visitors – some of them staying for extended periods – traders, architects and builders, mercenaries, philosophers and other travellers with an interest in Egypt. Among the Egyptian residents we must expect officials and administrators of various kinds, interpreters (cf. Hdt. 2.154), but also traders and craftsmen, some of them perhaps involved with the local faience scarab workshop (Gorton 1996, 92). Egyptian women, too, lived in Naukratis: on an inscribed late 5th-century BC Attic cup, a certain Gorgias and Tamunis declare their mutual love, and Tamunis – judging from her Egyptian name (though this may of course be misleading) – appears to have been of Egyptian descent. The idea that the population of Naukratis was characterized by cross-cultural integration at this time – if not already earlier – is supported by a further large and important body of evidence: the numerous, previously largely unstudied and unpublished terracotta and stone figurines of the Saite through to the Roman periods, which are now being investigated by Ross Thomas as part of the current project. Certainly from the 5th century BC these not only reveal certain hybrid elements of iconography and style and the transfer of technical features, but they also suggest that Egyptian domestic religion related to fertility was practised in the town, hinting perhaps at mixed marriages. For later periods, the huge Egyptian statue of Horemheb, mentioned above, who self-identifies as a Greek, may stand as emblematic for a certainly fluidity.
- ^  - Summary in Bernand 1970, 620–1, 624; cf. also Wilson and Gilbert 2007 on Sais, and Cooper 2008 on Nile branches; for a more sceptical view of Naukratis’ nautical accessibility see O. Höckmann 2008/9. Although one could argue that the existing evidence cannot support his conclusions concerning the site’s origins as a military fort (considered also by a number of other scholars) and settlement for retired Greek and Carian mercenaries, Höckmann’s article is an important study of seafaring and river transport in and off the Nile Delta region.
- ^  - It is important to emphasize here once more that new research on the production centres of East Greek Wild Goat style (once commonly labelled ‘Rhodian’) and Fikellura style pottery, as found in Naukratis and elsewhere, has conclusively shown that Ionian cities both north and south (and some of their colonies) were by far the dominant producers of these styles, with only very small groups of such pottery still attributable to possible production on the island of Rhodes (cf. e.g. Cook and Dupont 1998; Akurgal et al. 2002; Schlotzhauer and Villing 2006; Mommsen et al. 2006; Posamentir and Solovyov 2006, 2007). Scholars thus need to be careful when interpreting ‘Rhodian’ pottery at Naukratis as being indicative of connections between Naukratis and the island of Rhodes (cf. Malkin 2011, 83).
- ^  - Berlin 7780 dating from the reign of Apries (589–570 BC - the precise date is disputed, see Guermeur 2005, 128-9) and St Petersburg, Hermitage 8499, dating from 554 BC. An ‘Egyptian from Naukratis’ is also mentioned in the later Lindos decree, cf. Bresson 1980, 2000, 2005; Möller 2005