Retracing the olive ( Olea europaea L. subsp. europaea ) origin and its early history diffusion among the civilizations overlooking the Mediterranean coasts: an integrated approach

1 Introduction

The olive (Olea europaea L. subsp. europaea) thousand-year history overlaps and intersects with the social, cultural, religious and economic development of the Mediterranean, birthplace of the first human civilizations, from the IV–III millennia BCE onward. This tradition means that even today, this plant retains significant economic, cultural (ranging from gastronomy and medicine to art, mythology, and history), social, and environmental importance. This long-standing tradition has meant that still today this plant in this area has a prominent social, economic, scientific, cultural (encompassing diets and traditional medicine, as well as art, folklore, and historical heritage) and environmental value (Lombardo et al., 2022; Mazzei et al., 2026). Yet the precise origins of its domestication, the routes and time of its primary diffusion, and the relationships between cultivated olive varieties, their wild relatives, and associated pests such as the olive fly (Bactrocera oleae) remain the focus of intense interdisciplinary research (Nardi et al., 2010; Besnard et al., 2013a; Khadari and El Bakkali, 2018; Julca et al., 2020).

Palynological, archaeobotanical, and genetic data converge in identifying the Levant as the original cradle of olive cultivation as early as the V millennium BCE, with subsequent centers of large-scale cultivation in the Aegean (especially Crete) by the early/mid IV millennium BCE (Langgut et al., 2019; Barazani et al., 2023; Sabatini, 2024). Population genetics and plastid phylogeography similarly point to a first domestication in the northern Levant, embedded within long-term wild lineages that diversified since the Late Quaternary in three main refugial regions: the Near East, the Aegean area, and the Strait of Gibraltar (Besnard et al., 2013b; Mousavi et al., 2017; Besnard et al., 2018). Genomic and genealogical reconstructions refine this picture by showing that cultivated olives derive from a major eastern domestication but have repeatedly admixed with genetically rich wild populations, particularly in the western Mediterranean, generating the modern mosaic of cultivars (Julca et al., 2018; Lombardo et al., 2019; Mariotti et al., 2023; Gómez-Gálvez et al., 2026). Strikingly, the genetic differentiation of B. oleae and its vertically transmitted symbiont Candidatus Erwinia dacicola, as well as the broader phylogeography of olive-fly metapopulations, also follow an East-West pattern that parallels olive’s domestication and diffusion history around the Mediterranean basin (Lantero et al., 2022: Martinez-Sañudo et al., 2024). However olive diffusion was neither linear nor uniform until the III millennium BCE, whereas archaeobotanical, architectural, and textual records collectively document the first emergence of integrated olive oil economies and cultures in the southern Levant, Aegean, Anatolia and Egypt, accompanied by sophisticated pressing installations involving large-scale oil extraction, storage technologies and long-distance trade, and the embedding of olive oil in culinary, ritual, cosmetic, and lighting practices (Archi and Biga, 2003; Nowicki, 2009; Cuyler, 2012; Pratt, 2021). This preceded a complex, stagewise westward expansion mediated by maritime exchange networks, colonial foundations, and regional sociopolitical transformations (Valamoti et al., 2018; Manzano-Agugliaro et al., 2025; Langgut and Sasi, 2023; Palli et al., 2025).

Within this context, a multidisciplinary approach integrating palynological, archaeobotanical, genetic, entomological, and textual evidences, has been here used to chronologically reconstruct a coherent, cross-disciplinary narrative of the emergence of a shared oleoculture throughout the Mediterranean. Specifically, this article explores (1) the origin of O. europaea subsp. europaea and its early diffusion among the Mediterranean civilizations; (2) the correspondence between these processes and the genetic differentiation of B. oleae; (3) the Near Eastern birth and westward spread of olive cultivation in early history; (4) the spatio-temporal coherence of the most ancient written and iconographic references; (5) the archaeological record of olive oil extraction technologies; and (6) the destinations and uses of olives and olive oil.

2 Origins

Wood fossils would reveal the presence of the genus Olea in India since the Early Tertiary, c.a. 66 million years BCE (Srivastava, 1991). Paleobotanical and pollen analysis converge in placing the occurrence of Olea sp. in Europe during the transition from Oligocene to Miocene (about 23 million years ago; Mya) (Terral et al., 2004; Besnard et al., 2018; Suc et al., 2018). Analogously, pollen grains of Olea have been surprisingly identified through combined light microscopy (LM) and scanning electron microscopy (SEM) from the Late Oligocene/Early Miocene sediments of Altmittweida in Germany (Kmenta and Zetter, 2013). The evidence of Olea pollen in the Early/Middle Miocene in Sicily and Spain (Suc and Bessais, 1990; Roiron et al., 1999) and of fossil leaves in the Late Miocene clayey marls of Gabbro, Italy (Therios, 2009) might likely be attributed to ancestors of the wild olive tree, or oleaster (O. europaea L. subsp. europaea var. sylvestris), from which the cultivated one would derive (Barazani et al., 2023), insofar as fossil European putative species of Olea like Olea moldavica Negru and Olea oleastroides have been described (Macaluso et al., 2017).

The time intervals reported so far are somehow consistent with the whole-genome duplication (WGD) events identified through the availability of annotated genomes of the subspecies europaea and cuspidata. WGD events represent a key point in plant speciation as most of the genetic material generated by the complete duplication of the genome is subsequently lost, generating mutations of genes that can also take on new functions, contributing to the evolution of a species (Kaltenegger et al., 2018). Three polyploidization events have been recognized for the subsp. europaea var. europaea and cuspidata, occurring 69.38-81.88 Mya, 34.69-40.94 Mya, and 4.34-5.12 Mya (Wu et al., 2022). According to Julca et al. (2018) three events in O. europaea var. europaea occurred 33–72 Mya (separation of the Oleaceae family), 14–33 Mya (separation of the Oleae tribe) and ~10 Mya specific for the olive tree and its relatives. Only two peaks have been found in O. europaea subsp. europaea var. sylvestris (Unver et al., 2017; Wu et al., 2022). Wu and collaborators also estimated the divergence time of olive subspecies sharing the last common ancestor 2.58-6.23 Mya, as a result of the most recent WGD event, which in accordance with the phylogenetic analysis (Besnard et al., 2009) that places the common ancestral origin of the O. europaea subspecies (cerasiformis, cuspidata, guanchica, laperrinei, maroccana and europaea) in the Late Miocene (between 4 and 8.3 million years ago).

The wild olive underwent a further differentiation into three different haplotype lineages in the Late Pliocene/Early Pleistocene (E1, E2, E3) corresponding to well-defined geographical areas: the eastern Mediterranean area (E1) and the central-western Mediterranean area (E2, E3) (Besnard et al., 2011; Julca et al., 2020; El Bakkali et al., 2025).

Fossil leaves (1.6 to 0.8 million years BCE) attributed to O. europaea L. (var. sylvestris) have been found in the Early Pliocene deposits of Mongardino, in Italy (Cavara, 1887). Palynological, carpological and anthracological evidences support the existence of wild forms of olive trees in the paleolithic Acheulian site at Gesher Benot Ya’aqov (Israel), dating to ~0.8 million years ago with a continual 100–150 kyears¹ hominin occupation (Goren-Inbar et al., 2004; Melamed et al., 2016; Van Zeist and Bottema, 2009). Subsequent finds date back to the Middle Paleolithic in glacial refugia in Morocco (~100,000 years old; Marquer et al., 2022), in Greece (~60,000 years old; Kuhn et al., 2010), in Thera -Santorini island- (~50,000 years old; Velitzelos, 1990), in Israel (43,000 -Liphschitz et al., 1991 and 19,000 -Kislev et al., 1992- years old), and in Spain (~42,000 years old; Jennings et al., 2009; Carrión et al., 2025), and from the Neolithic to Chalcolithic (10,000–4,500 BCE) in the rest of the Mediterranean basin including Lebanon (Kaniewski et al., 2023; Cheddadi and Khater, 2016), Ionian and Aegean Greek islands (Ntinou and Valamoti, 2025) and northern Sporades (Ntinou, 2011), Turkey (Fiorentino et al., 2014; Gurbuz-Veral et al., 2018), Portugal (Figuieral and Terral, 2002), Cyprus (Brun, 2004), Israel (Kislev et al., 1992; Langgut and Garfinkel, 2022), Jordan (Dighton et al., 2017), Syria (Sorrel and Mathis, 2016), Spain (Pantaleon-Cano et al., 2003; López Sáez et al., 2005; García-Rivero et al., 2019), Corsica (Vella et al., 2019) and mainland France (Solari and Vernet, 1992; Terral, 2000), Iraq (Al-Ameri et al., 2011), Croatia (imported drupes?; Parica, 2024) and Italy (Costantini, 1981; Tinner et al., 2009; Mercuri et al., 2013; Dodd, 2026). The expansion of the olive tree’s distribution range from the end of the last Glacial Period could be somehow linked, especially in the central-western Mediterranean, to the establishment of present-day flyways (Bruderer and Salewski, 2008) for frugivorous migrant birds, acting as olive seed dispersers.

2.1 Correspondences with the genetic differentiation of Bactrocera oleae

Studies on the olive fruit fly (Bactrocera oleae, Rossi) populations based on mitochondrial DNA and microsatellite loci identified the existence of three distinct major groups from Africa, Pakistan and the Mediterranean Basin (Nardi et al., 2005, Nardi et al., 2010; Dogaç et al., 2013) and other two from China and Reunion Islands (Bon et al., 2016). This division followed the diffusion of the Olea europaea complex with particular regard to the subspecies europaea and cuspidata. Specifically, Nardi et al. (2010) suggested an African origin of the olive fruit fly, with the subsequent separation into the Mediterranean and Pakistani populations occurring between 0.57–0.83 (confidence interval 0.35–1.5) Mya. The timing of this differentiation would fit with the proposed expansion of wild olive populations after isolation in glacial refugia, in accordance with Besnard et al. (2013a), placing the most recent common ancestor of the Mediterranean lineages in the Middle/Late Pleistocene. A confirmation of the African origin of this dipterous would come from the close phylogenetic relationship between B. oleae and the exclusively African B. biguttula (da Costa et al., 2019; da Costa et al., 2024) diverging from a common ancestor ~10 Mya, as even reported by Choo et al. (2019). In time (but not spatial) agreement, by studying the divergence among the three alleles (I, F and S) of the polymorphism of the alcohol dehydrogenase-2 (Adh-2) gene in B. oleae, Goulielmos et al. (2003) speculated that the split between B. oleae and B. dorsalis would have happened 11–22 Mya, possibly in the Indian subcontinent -the center of origin of most of the species of the Bactrocera genus (Hoelmer et al., 2011)-, hence, olive fly would have followed the westward diffusion of olive. In turn, the appearance of the F allele from the I allozyme (considered as the ancestral allele as naturally present in less than 1% of the tested B. oleae population from Greece, with its frequency increasing in a few generations after artificial rearing/feeding of larvae) would have happened 1-2.4 Mya as a likely result of B. oleae specialization on O. europaea (oleaster)? fruit in the Mediterranean area (where B oleae is the only species of the genus). Once again there would be a general correlation with the expansion of wild olive populations. A westward diffusion of B. oleae in the Mediterranean was proposed following the evidence of a gradual decline of heterozygosity from east to west (Augustinos et al., 2005; Dogaç et al., 2013; Eti et al., 2017), possibly occurring after the spread of the domesticated olive from its Levantine origin.

Regardless of the species’ westward or north-eastward original migration direction, the Mediterranean population of B. oleae has been further subdivided in genetic groups showing a certain level of gene flow. However, although studies analyzing mitochondrial DNA converge in identifying the long-established presence of B. oleae populations in the Mediterranean Basin, the question of division into genetic groups is not yet resolved. Specifically, Nardi et al. (2005) and Segura et al. (2008) described no clear geographic breakdown; Nardi et al. (2010) identified an eastern and a central-western group, while other authors (Augustinos et al., 2005; Zygouridis et al., 2009; Dogaç et al., 2013; Van Asch et al., 2012; Van Asch et al., 2015; Ramezani et al., 2015) reported the distinction of an eastern (subclade P; Cyprus, Israel and eastern Turkey -plus California-), a central (subclade M; Greece, Italy, France and western Turkey) and a western (subclade O; Spain, Portugal) genetic group. This division would corroborate the differentiation into the E1, E2, E3 haplotype lineages of wild olives during the Late Pliocene/Early Pleistocene. More, genetic polymorphism at the cytochrome oxidase I (mtCOI) gene, highlighted the separation of both three Matallanas et al., 2013) and two (Lantero et al., 2022) different B. oleae genetic clusters. On the split into two groups, the geographic distribution of the different haplotypes of the B. oleae obligate coevolved symbiotic bacterium Candidatus Erwinia dacicola would reveal an eastern and a western lineage, with different degrees of mixing in the central Mediterranean. In a 2012 work based on 16S rRNA gene sequences, Savio et al. (2012) first described two Ca. E. dacicola haplotypes (htA and htB) from B. oleae specimens collected in several Italian Regions, finding that flies from Sardinia only harbored the htA, while only the htB was detected in Sicily, with intermixed populations in the continental areas. Only one haplotype was found in each fly. The authors suggested that the two haplotypes might have diverged about 1.3-1.6 Mya (recurring time). Subsequently, this research group expanded the survey area, covering most of the Mediterranean basin (plus California), South Africa and Pakistan (Viale, 2014; Martinez-Sañudo et al., 2024). A third haplotype (htC) was found exclusively in Pakistan, while the western Mediterranean B. oleae samples from Spain, Portugal, France -including Corsica-, Sardinia, Algeria and Marocco mostly harbored the htA, the eastern Mediterranean ones from Turkey, Israel, Cyprus, and Egypt -plus Tunisia- only harbored the htB, and the ones from Italy, Slovenia, Croatia and Greece (central Mediterranean) presented different degrees of mixing of the two types. An exception was Crete where only the haplotype A was found. As for B. oleae, an east-to-west Ca. E. dacicola haplotypes distribution was proposed. In partial confirmation, Nobre (2021), collecting B. oleae samples from Spain, Portugal, France and Tunisia, found no clear differentiation between populations in the European samples, but she discovered two new haplotypes (called htC -different from the one in the Italian study- and htD) in Ca. E. dacicola from two different localities in Tunisia. This last aspect seems particularly important in the light of a Tunisian genetic subgroup of B. oleae as described by Segura et al. (2008) and Matallanas et al. (2013).

3 Early history

3.1 The near east birth of olive cultivation and its Mediterranean westward diffusion

The first forms of olive cultivation would date back to the Late Neolithic/Early Chalcolithic (VI-V millennium BCE) in the geographical region extending from Mesopotamia (Iraq and part of Syria) to Lebanon, Jordan and ancient Palestine (Neef, 1990; Galili et al., 1997; Galili et al., 2021; Besnard and Bervillé, 2000; Lovell et al., 2010; Kaniewski et al., 2023; Dighton et al., 2017; Langgut et al., 2019), with a possible pre-dating at 10,000 years ago in northern Iraq (Al-Ameri et al., 2011). Moreover, the study of the shapes and dimensions of the well-preserved kernels of the submerged site of Hishuley Carmel (Israel), mid-fifth millennium BCE, revealed the presence of olives most likely belonging to both wild olive trees and domesticated forms (Terral et al., 2021). On the same matter, morphometric analysis of fossil olive pits would place the cultivation of domesticated forms in the Early Pottery Neolithic village of Pella in Jordan (Dighton et al., 2017) and in the 3000 years old site of Ugarit (Newton et al., 2014).

The near east birth of olive cultivation together with the very strong eastern-Mediterranean E1 oleaster lineage highlighted by modern genetic data (Besnard et al., 2013b; Besnard et al., 2018) might suggest the introduction of Levantine plant material (and, to some extent, of oliviculture) by Semitic speaking populations in southern Anatolia and in the Greek islands of the Aegean, where the first proposed forms of olive cultivation dated to the V-IV millennium BCE (Asouti, 2003; Gurbuz-Veral et al., 2018; Jouffroy-Bapicot et al., 2021; Ntinou and Valamoti, 2025). A possible alternative involves avian dispersal of Levantine seeds (Spennemann and Allen, 2000), creating populations subsequently exploited and selected by humans. Accordingly, pollen analysis would place olive appearance in Crete around the end of the V millennium BCE, with a considerable expansion concurrently with the beginning of the Minoan period (ca. 3,500 BCE), indicating a plausible certain degree of human intervention (Cañellas-Boltà et al., 2018; Livarda et al., 2021), so that a an independent start of cultivation from native or “bird-introduced” wild forms cannot be excluded in this area, possibly leading to independent primary domestication events². This could somehow justify the different roots of the terms used to define the olive tree in the Mycenaean Greek (e-ra-wa^³) and in the ancient Semitic languages (the Akkadic/Sumerian/Babilonian/Hittite zi-ir-tum/serdum, the Ugaritic/Phoenician zt, the Syriac Aramaic zaytā, the Persian za-a-da-um’ etc.; Watson, 2004; Blažek, 2014), and the fact that these roots spread throughout the rest of the Mediterranean Basin via an Aegean and a “Semitic” (Figure 1) route. Independent forms of relevant olive exploitation may also have occurred in Italy. Palynological data from Lago Alimini Piccolo (Apulia; Di Rita and Magri, 2009), charcoal data collected in the archaeological site of Punta di Zambrone (Calabria; D’Auria et al., 2017) and olive oil residues in potteries found at the coeval settlements of Castelluccio di Noto (Sicily; Tanasi et al., 2018) and Coppa Nevigata (Apulia; Evans and Recchia, 2003), would suggest O. europaea (likely var. sylvestris) cultivation in southern Italy already in the lower limit of Early Bronze Age -EBA- (XXI-XVIII century BCE), well before the Middle to Late Bronze Age (XV-XIII century BCE) when the connection between southern Italy and the Aegean started to leave traces of a consolidated commercial relationship (Astour, 1985)⁴. Consistent with this, olive pits morphometrically attributable to domestic forms have been found in the Appennine and Sub-Apennine levels (1500–1200 BCE) of the archaeological site of Piazza Palmieri, in Apulia (Terral et al., 2004) and in the Final Bronze Age (1150–950 BCE) settlement of Fonte Tasca di Archi, Abruzzo (Di Fraia, 1998).

In the transition from the III to the II millennium BCE, traces of early olive cultivation have been identified in the peninsular and continental Greece (Valamoti et al., 2018) and almost simultaneously to this “second wave”, Hamito-Semitic populations introduced olive cultivation in a southwestern direction up to Egypt (Asensi Amoròs, 2003), were olive leaves⁵ made up the floral collars found in the burial chamber of the tomb of the pharaoh Tutankhamun (ca. 1336–1327 BCE; Supplementary Figure S1). Eventually, under the pressure of -basically- the Phoenicians, from the XIII century BCE, oliviculture reached the central and westernmost shores of North Africa -but with a great intensification only from the IV century BCE (Wolff, 1996; van Zeist et al., 2001)- and Spain (Manzano-Agugliaro et al., 2025). It was probably subsequently introduced in the Massalian territories (Marseille, France) by the Phocaeans from the VI century BCE (Strabo, Geography, IV, 1.5; Justin, Epitome XLIII, 4). In accordance, data based on olive pits and pollen analysis (Alonso et al., 2016; Pérez-Jordà et al., 2017; Bourgeon et al., 2018; Langgut et al., 2019) agree in framing the birth of a structured olive growing in southern Spain at the beginning of the first millennium BCE⁶, while olive archaeobotanical remains have been reported to occurr more frequently in the VI century BCE levels of Eastern Languedoc and the Rhône Valley (Pérez-Jordà et al., 2021). Based on this reconstruction, the expansion routes of olive growing in the Mediterranean basin are resumed in Figure 1.

However, further localized independent events of olive cultivation preceding this narrative have been proposed during the Neolithic for Spain (Terral and Arnold-Simard, 1996; Carrión et al., 2024), mainland France (Terral, 2000) and Corsica (Vella et al., 2019), as well as -with a certain caution- for Morocco (Leroy et al., 2025) and Tunisia (Lebreton et al., 2015). On this argument, a summary of the proposed dates for the beginning of olive cultivation throughout the Mediterranean basin is reported in Table 1.

Table 1. Proposed starting dates of olive cultivation throughout the Mediterranean basin.

3.2 Spatio-temporal coherence of the earliest written references

One of the driving forces that led to the creation of written representations for the fixation of linguistic signs in a lasting way, was the need for recording transactions and sales. As olives and olive oil were among the most commonly traded goods, they already appeared in the most ancestral forms of writing, in accordance with the same temporal and spatial pattern of olive diffusion. The first reference to olive could be found in the Protodynastic Libyan Palette (or Tehenu Palette; Supplementary Figure S2) dated to the Naqada III Period of Egypt (ca. 3100 BCE), one of the earliest written documents, whereas the first example of writing system -the cuneiform script, evolved from a much older clay token system- date back to about 3200 BCE (Schmandt-Besserat, 2014). The fourth register on the front side of the stone shows eight trees, possibly olive trees, to the left of which appears the hieroglyph indicating the word tehenw (transcribed tehenu), probably the name of the region corresponding to -part of- modern Libya. Newberry (1915) translated this toponym into “Olive-land”, even for the occurrence of a Tehenu-oil in several lists of offerings from the Old Kingdom. Nevertheless, this reconstruction, as well as the identification with Libya, have been contested (Cooney, 2011); furthermore, there do not appear to be sufficient biological evidences to confirm this hypothesis. However, the first document concerning the production and sale of olive oil is represented by the cuneiform clay tablets in Eblaite (a peculiar Semitic language) found in the city-state of Ebla (modern Tell Mardikh), in northern Syria, dating back to the mid-third millennium BCE. In particular, the text TM.03.G.1000 reports the transport of a tribute of 1,081 olive oil jars (ca 32,430 liters) from Ebla to the city of Mari (Archi and Biga, 2003) during the reign of the king Nizi, confirming extensive olive growing and oil production in the area in the XXIV century BCE. Likewise, strong evidence of olive cultivation in the territories controlled by Ebla comes from the administrative tablets MEE 7, 45, enlisting a total of 2600 olive trees, and TM.75.G.1700 leading to the estimated production of 169,050 (Benati and Bonechi, 2020) to 330,000 (Archi, 1991) liters of olive oil, from consequently 241–483 to 600–1200 ha of land occupied by olive groves. Production and trade of olive oil in Ebla was confirmed by the findings in the Royal Palace G; in particular by the presence in one room (L.4420) of the remains of what could likely be an olive-press, and of pattern-combed jars suggested for oil transport (Peyronel et al., 2014). In fact, proofs of the employment of metallic and pattern-combed jars for oil storage have been found in the coeval EBA city of Khirbet al-Batrawy, in Jordan, containing olive oil from the surrounding producing areas (Nigro, 2011). In the same period olive cultivation must have been particularly important also for the city of Tunip (western Syria; Bonechi, 2016). On the Iraqui side of Mesopotamia, the logograms for serdum (olive tree) and šaman serdī (olive oil) -and their variants- can be found in different Akkadian (end of III-II millennium BCE) and in Sumerian texts dating to the Third Dynasty of Ur (XXI century BCE) (Heimpel, 2011) and to the coeval Second Lagaš Dynasty (Colonna d'Istria, 2022). In this latter case, the tablets BM 29719 and RTC 216 found in Girsu describe the storage of tens if not hundreds of sila (=liters) of olive oil in jars of various sizes (Maiocchi and Visicato, 2020; Colonna d'Istria, 2022). Moreover, the presence of olive trees of different ages in the Sumerian royal gardens of Puzriš-Dagān, Garšana and Lagaš (over 100 trees) for ornamental purposes (i₃.MesopOil Project, 2025), in an area (Southern Mesopotamia) beyond the limits of olive cultivation, highlights consolidated knowledge on olive tree propagation. Eventually in the Code of Hammurabi, the king of Babylon who lived between 1792 and 1750 BCE, there are two laws dedicated to the trade of goods including olive oil, concerning the issuance of receipts (law 104^th) and the penal liability of sea carriers (law 237^th).

Ideogram for “olive” and “olive oil” have been deciphered for both Linear A and Linear B (ideogram AB 122 for olive and A302, B130 for olive oil; Everson and Younger, 2010). Linear A is the still encrypted Minoan (1800–1450 BCE) writing system precursor of Linear B, the system based on syllabic signs with phonetic value and on ideograms used by the Mycenaeans. Regarding the latter, several written references on olive are found in the tablets of Knossos (XV-XIII century BCE) found in Crete. In the tablets of the Fh, Fp and Fs series (classified according to origin), the ideogram 130 is often combined with the symbol of other spices, indicating the production of ointments or perfumes (Chadwick, 1966). Moreover, five clay cylinders discovered near large storage pithoi in the “Pithos Hall”, in the Late Bronze Age (ca. 1500–1200 BCE) Building X (a monumental structure intended for feasting and for the storage large amount of olive oil) at Kalavasos-Ayios Dhimitrios in Cyprus feature comprehensive Cypro-Minoan inscriptions, possibly relating to the management of large-scale olive oil production, according to researches by South et al. (1989) and Smith (2002). These cylinders, associated with the site’s central administration, are a further proof of how olive oil was a key component of Minoan culture.

Ugaritic (the original semitic language of the ancient city of Ugarit) clay documents datable to the XIII century BCE, record the olive oil trade with Cyprus (RS 18.42) and Egypt (RS 20.168), confirming the key role of this port city of ancient-time Syria, while XVIII century BCE documents from the old Babylonian city of Mari, revealed an extensive olive trade networks with Yamhad (ARM 7.238), Kutkutu (ARM 9.6) and Imār (ARM 24.278). Likewise, in the Amarna Letters EA 34 and EA 35 (XIV century BCE) a trade in flavoured olive oil between Alashiya and Egypt is attested. Remaining in the Levantine area, a reference to olive cultivation is likely contained in the enigmatic Gezer agricultural almanac from the X century BCE, a 11 centimeters long limestone tablet found in the archaeological site of Tell el-Jezari, the ancient Gezer, bearing the earliest inscriptions in archaic/pre-/proto- Hebrew (or Phoenician) describing the division of the months according to agricultural activities (Pardee, 2013), whereas olive oil is mentioned as part of the military food ration in Tel Arad (Israel) in the so called Arad ostraca, a collection of over 200 inscribed pottery shards dating from VII century BCE (van W Scheepers, 2006). Moving northward, olive diffusion in Anatolia during the Hittite imperial period is confirmed by the cuneiform tablet KUB 40.2 (obv. 36), quoting a property featuring 300 olive trees in the territoty of Kizzuwatna (Hoffner, 1974).

Regarding the olive healing properties, in the Ebers papyrus, a treatise on Egyptian medicine dated between 1550 and 1530 BCE, olive oil is an ingredient in ointments and remedies indicated for the most diverse pathologies (diseases of the eyes, ear, nose and mouth; Bryan, 1930), while in the Hearst Medical Papyrus (ca. 1500 BCE) olive oil is contemplated in preparation of decoctions and in the adjuration against the Asiatic disease (Leake, 1952). In the surviving fragments of the cuneiform Nineveh Medical Encyclopaedia, the Assyrian medical literature compendium found in the Library of the king Ashurbanipal (VII century BCE) and dated to the Neo-Assyrian period (ca. X-VII century BCE), olive oil and olive leaves appear among the treatments for fever and cranial (chapter 1⁷) or neck (chapter 2) pathologies⁸. A less clear reference is found instead in the medical tablet BAM II 159, datable to the 7th century BCE and belonging to the Assur Medical Catalogue, interpreted as a reciprocal protective rubbing with olive oil for the patient and the healer/caregiver (Parys, 2014). Considering the practice of anointing, a means of conferring both physical and spiritual benefits, it is paradigmatic that in the Old Testament (starting from XIII century BCE) this word and its derivatives are used over 130 times, with particular reference to olive oil, while in Homer’s Odyssey (VIII-VII century BCE) 10 out of the 11 total references to olive oil concern the anointing. Homer also describes the ritual of oiling with perfumed oil the dead bodies of Patroclus (Iliad, XVIII, 349-351) and Achilles (Odyssey, XXIV, 44-45)⁹ destined for the pyre. It is furthermore specified that Achilles places two amphorae filled with oil tilted towards Patroclo’s deathbed, as a ritual part of the funeral sacrifices. The recurrent presence of stirrup jars¹⁰, the typical olive oil/balm-bearing vessels of the Minoan Proto- to Post-palatial period (1925–1200 BCE), in Cretan cremation burials would place this practice since the Late Helladic IIIB period (1250–1200 BC) (Cultraro, 2011).

It took the Etruscan culture to find the first written references to olive oil outside the eastern Mediterranean regions. In fact, the term elaivana (an adjective meaning “of oil”) occurs on an aryballos (oil flasks of Greek stylistic derivation) dating to the second half of the VII century BCE ¹¹ (Poetto and Facchetti, 2009), and (in the variant eleivana) on an almost coeval inscription (TLE 762) on a bucchero (an Etruscan ceramic) aryballos bearing a similar inscription¹² (Maggiani, 1972). These inscriptions served as “speaking objects”, where the vessel itself identifies its contents or owner to the user. The word eleiva (oil) itself is a direct loan from the Greek élaion (ἔλαιoν). Its appearance in the VII century BCE represents one of the oldest documented Greek loanwords in the Etruscan language, highlighting the early trade of olive oil and perfume between Greece and Etruria.

3.3 Ancient archaeological evidences of olive oil extraction process

Although it is believed that the production of oil began to become a widespread practice starting from the Bronze Age (III millennium BCE), findings in the Jordan site of Pella led to propose a first form of -likely- oil extraction from ~5200 BCE (Dighton et al., 2017). Contemporarily, olive oil turned out to be one of the most common organic residues in artifacts (vases of pottery, large quantities of flint blades and various tools) from the Late Neolithic period (Wadi Rabah culture or pre-Ghassulian phase, from the VI to the V millennium BCE) found in the archaeological site of Ein Zippori in Israel (Namdar et al., 2015). Accordingly, the radiocarbon dating of thousands of crushed olive pits with olive pulp -thus indicating the process of oil extraction- found in excavations of the prehistoric settlement of Kfar Samir off the coast of Carmel, south of Haifa (Israel) places the development of this process in the mid-sixth millennium BCE (Galili et al., 1997). Back to Jordan, two large installations found in the Early Bronze Age IB (late IV millennium BCE) site of Jebel al-Mutawwaq have been cautiously suggested as possible bases of some kind of olive press system (Caselli et al., 2024), coeval to the olive oil press installations discovered at the site of Tall Rakan (Banning, 2007). A proto form of specialization is deduced from the structures and the remains of olive stones in the architectural complexes at the late III millennium BCE site of Khirbat Umm al-Ghuzlān (Fraser et al., 2022).

Moving westward in the Mediterranean, structures interpreted as a rudimental pressing system and charred olive stones dated to the V millennium BCE have been discovered in a cave of Scaffa Piana (Saint-Florent) in Northern Corsica (Magdeleine and Ottaviani, 1984). Chromatographic analysis of the organic residue in a clay bowl dated to 4500–3800 BCE, found in the Gerani cave (Crete), highlighted the presence of oleic acid (Martlew, 1999). Nevertheless, it is debated whether it was from olive oil or another vegetable oil. Gas chromatographic analysis of sherds of nine big pithoi (capable of holding up to 165 liters) found at the Early Minoan I (3200–2700 BCE) fortified site of Aphrodite’s Kephali in Crete, has revealed their intended use for holding olive oil and wine (Koh and Betancourt, 2010). These significant volumes confirm the high level of specialization already achieved on the island in the IV millennium BCE. In the Cyclades, charcoal analysis from the ancient city of Akrotiri, destroyed and buried by the eruption of the Santorini volcano in 1628 BCE would demonstrate olive tree exploitation as firewood and for oil production since EBA (3000–2100 BCE), but still with an indefinable level of domestication (Asouti, 2003). In mainland Italy, apart from the already mentioned olive oil-bearing jars found at the EBA settlement of Castelluccio and Coppa Nevigata belonging to the Protoappennie period (1800–1600 BCE), large pithoi dating from the Middle Bronze Age to the Early Iron Age, were found in several sites of Sicily, Calabria, and Apulia where they were commonly used for the storage and possibly the transport of foodstuff including olive oil, as confirmed by chimical analysis (Schiappelli, 2015). They were locally manufactured, inspired by Minoan (and subsequently Mycenaean and Cypriot) models transmitted through the close contacts established between Southern Italy and the Aegean world during this period, and could hold up to a thousand liters of oil, as was the case with the dolia from Broglio di Trebisacce (Calabria) dated to the XIV-XII century BCE (Schiappelli, 2015). All of this points to substantial olive-oil production at the time (Caracuta, 2020).

From here on, the presence of olive oil presses spread extensively throughout the Mediterranean Basin, with the largest -to date- industrial archaeological site for olive oil production, dating back to the VII century BCE, found in Tel Miqne (identified with the biblical Ekron) in Israel, which was to contain over 100 presses (Eitam, 1996).

3.3.1 On olive oil destination

The first phases of olive exploitation/cultivation, and thus of oil extraction, obviously had a localized and limited dimension. Paleo-botanic evidences from EBA (3200–2100 BCE) sites occurring in domestic contexts led to the conclusion that olive oil production and consumption were mostly circumscribed to a family level, but already starting from Middle Bronze Age, olive oil probably became an elite product (Sabatini, 2019). Notwithstanding, the discovery of Palestinian pottery vessels in Early First Dynasty tombs led to suggest that these containers served for the transport of olive oil (and wine) traded from Canaan to Egypt since Early Bronze Age I (EBI; Stager, 1985; Ben-Tor, 1986; Lev-Yadu and Gophna, 1992). This hypothesis is supported by the fact that olive was an established crop in EBI Canaan (Liphschitz et al., 1991).

Moreover, archaeological excavations on the island of Cyprus (site of Pyrgos-Mavroraki) put forward the hypothesis, supported by experimental data, that 4000 years ago (age of the finds) olive oil was used as liquid fuel (to increase temperatures) in furnaces for melting copper and for the manufacture of metal alloys (Belgiorno et al., 2009). In the same regard, several studies (revised in Rowan, 2015) show that also pomace (or jift) was used as a fuel both in domestic and industrial contexts starting from the third millennium BCE as deduced from the multiple findings of large quantities of carbonized olive stones. On this point, in the archaeological site of Tell Tweini (Syria), a marked increase of oil production was reconstructed during the Early Iron Age as a possible consequence of a higher employment of olive oil or pomace as fuel (Linseele et al., 2019), while in Tell Hammeh (Jordan), the site of the earliest bloomery smelting of iron, the remains of olive wood charcoal suggest its metallurgical use during the X century BCE (Yahalom-Mack and Eliyahu-Behar, 2015). These evidences together with the fact that the first systematic smelting of iron from ores began in the early II millennium BCE in the Anatolian area¹³ (Erb-Satullo, 2019), and the proven olive cultivation by -especially- the Hittites¹⁴ (XXI-XIII century BCE; Güterbock, 1968; Oflaz et al., 2019; Deckers and Gates, 2026) and the Phrygians¹⁵ -even at extraordinary altitudes over 1000 m asl- (XII- VII century BCE; Strabo, Geographia XII, 8)¹⁶ could direct towards a possibly important role of olive oil in the transition to the Iron Age, especially considering the need to reach higher temperatures inside the furnaces for the processing of iron tools and weapons. This usage may have contributed to the increase (or at least the non-decrease) of oil production occurred in this transition phase in the Southern Levant, although it was traced to the introduction of round and rectangular oil presses allowing a higher and more efficient processing of the olives, needed to reach self-sufficiency for the increased internal olive oil and oil rose consumption and to support a greater share exported to Egypt (Onozuka, 2012).

In this historical phase, although it is not possible to make definitive statements, the primary use of olive oil was likely for the -on a larger scale- production of balms, perfumes and anointments employed for both cosmetic and ritualistic purposes, as attested in Eblaite (Viganò, 2000), Mesopotamian (Malul, 1996), Minoan/Mycenaean (Melena, 1983; Cuyler, 2012) and Hittite (Nowicki, 2009) sources. This demand for olive oil for the production of balms and perfumes created an articulated network of contacts in the Mediterranean. The Minoans exported olive oil to Troy and the Hittites and imported it from Cyprus (Rjabchikov, 2023), with Cretan stirrup jars that have been found at sites in Sardinia, Nubia, Syria and Cyprus (Leonard et al., 1993). Moreover, it has been argued that the olive oil demand from Egypt, intended for aesthetic and sacral aims, played an important role in the expansion of olive growing in the first two millennia BCE (Faust and Weiss, 2005), with trade exchanges involving the Minoans (Rjabchikov, 2023), Ugarit (Heltzer, 1987), Alashiya (Parpas, 2022), the Hittites (Onozuka, 2012) and the Phoenicians (Faust and Weiss, 2005). Moreover, in Hittite documents the occasionally appearing attribute “pittalwan” for olive oil (^GISSERDUM) has been interpreted as “plain, without additives” (Hoffner, 1995). This specification may perhaps indicate a more common use as a basis for other products. Another written clue suggesting its principal “external use” is found in the biblical passage¹⁷ Micah 6:15-16, dated to the VIII-VII century BCE (Thou shalt sow, but thou shalt not reap; thou shalt tread the olives, but thou shalt not anoint thee with oil; and sweet wine, but shalt not drink wine). Nevertheless, as the large quantities of olive oil produced during Iron Age II in Palestine have led to estimates of annual consumption of 20–25 liters per capita (Mataix and Barbancho, 2006; Katz, 2008), it is likely that edible oil represented a large share of these volumes.

3.4 Olive drupes exploitation as food

Regarding wild olive drupes exploitation as food, Marquer et al. (2022) suggest that the oldest evidences would come from the Aterian Middle Stone Age (ca. 100,000 years old) archaeological sites of El Mnasra and El Harhoura 2 (Morocco), where fossil olive stones would reveal the presence of prehistoric gatherers. With a notable temporary break, wild olive drupes would have been eaten in the hunter-gatherer archaeological site Ohalo II, in Israel, datable to approximately 19,000/23,000 years ago (Weiss et al., 2004). The gathering of drupes for food purposes has been also proposed in the 10,000 years old site of Zawi-Chemi Shanidar, Iraq (Al-Ameri et al., 2011), and in the Epipaleolithic and the Early Neolithic Ifri Oudadane, Morocco (Morales et al., 2013); while the first cultivated table olives would have been consumed at Hishuley Carmel (Galili et al., 2021). The cultivation and domestication of the olive tree inevitably increased the availability of its fruits so that Watson (2004) suggests that during Bronze Age olive drupes were a staple food in Ugarit, attesting the distinction between autumnal (hrpnt) and winter (mm; ripe) olives (zt).

Evidences of the increase in consumption and trade of olives as food in the Eastern Mediterranean are provided by two of the most famous and ancient wrecks in history: the Ulu Burun and the Cape Gelidonya shipwrecks, in modern Turkey. On the Ulu Burun Bronze Age shipwreck (ca. 1325–1300 BCE, the second most ancient after the Dokos shipwreck, ca. 2200 BCE) a Canaanite jar warding over 2500 olive stones was found. The stones, slightly larger than those found in coeval sites, led to the -extremely divergent- hypotheses that these drupes were part of the sailors’ food supplies or, alternatively, a commercial luxury product (Haldane, 1993). This latter would be in accordance with the elitist boat trade of olive oil reported in the Ebla and Mari documents (Chambon, 2020), and with the quarry inscription of Seti I (ca. 1294–1279 BCE) from the eastern side of Gebel Silsila (Egypt), in which rations of olives were given to the leaders (a royal messenger and a standard-bearer) of a workforce of 1000 men for the extraction of sandstone destined for the building of monuments (Kelder, 2009). These “luxury” and likely imported supplements were not enlisted among the rations intended to the common laborers. a royal messenger and standard-bearer. On the other hand, in support of the first hypothesis, olive stones have been found among what was left of the crew’s meal on the merchant ship that wrecked at Cape Gelidonya in ~1200 BCE (Bass, 1961). It seems reasonable to assume that drupes were somehow processed to face long sea journeys and transport all the way to the sandstone quarries of Upper Egypt. Concurrently, on the other side of the Mediterranean, western proto-olivicultures were still far from reaching this level of development so that olive drupes were typically consumed discontinuously (Alonso et al., 2016) and locally, like in the archaeological site of Sa Osa (Sardinia, Italy) dated to the Late Bronze Age (1350–1200 BCE), where olive stones have been found in a Nuragic age storage pit (classified as N) together with animal remains and plants seeds, as stored food (Usai et al., 2016). It should be noted, however, that the transport of wild olives fruits, perhaps for commercial purposes, has been proposed in the Neolithic Andalusia between coastal and inland (outside their natural distribution) settlements (Rodríguez-Ariza and Montes Moya, 2005).

4 Conclusions

The millennial cultivation of the olive tree, apart from representing an integral part of the territory and a characteristic element of the landscape, has inevitably permeated and sometimes shaped the culture and history of the Mediterranean countries (and beyond).

Multiple, independent lines of evidence now support a Near Eastern origin of olive cultivation, with primary domestication in the southern and northern Levant around 4000–5000 BCE, embedded in older wild lineages that survived Late Quaternary climatic oscillations in several refugia. From these centers, olive cultivation spread westward to the Aegean, central Mediterranean, and ultimately western Mediterranean, in step with the emergence of complex societies, maritime trade networks, and colonial enterprises. Genetic and phylogenomic studies reveal a complex domestication trajectory shaped by recurrent admixture and the retention of relatively high diversity in cultivated olives. Three major wild lineages and multiple plastid haplotype groups contributed to regional cultivar pools, while two or a few “founder” lineages underlie a large proportion of modern varieties. The East–West genetic structuring observed in olive populations is mirrored by the phylogeography of Bactrocera oleae and its primary symbiont, whose haplotype distributions and genetic groupings follow patterns consistent with long-term association and dispersal alongside their host tree.

Archaeobotanical, architectural, and textual records, further anchoring this spatio-temporal narrative, collectively document the emergence of integrated olive-oil economies and cultures throughout the Mediterranean Basin. Early oil-pressing installations in the southern Levant and subsequent technological elaborations in the Aegean, Egypt, Anatolia, and Italy reveal an increasingly capital-intensive processing chain, with oil destined for food, lighting, ritual, cosmetics, and trade. Taken together, these results support a model in which the Near East-birth of olive cultivation, followed by its maritime-mediated westward diffusion, underpinned the rise of distinctive Mediterranean agrosystems and cultural practices.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fevo.2026.1967318/full#supplementary-material

Supplementary material Supplementary material at the source

Footnotes

Edited by: Antonio Garcia-Alix, University of Granada, Spain

Reviewed by: Claudia Moricca, Sapienza University of Rome, Italy

Jon Camuera, Spanish National Research Council (CSIC), Spain

Footnotes

Marine Isotope Stages 18-20

The question of whether some of the secondary centers of domestication might actually be independent primary ones has been raised in Terral et al., 2004; Hannachi et al., 2010; Newton et al., 2014; Díez et al., 2015, despite the evidence that almost all the cultivated varieties derive from the E1 gene pool of the wild olive, confirming the eastern domestication hypotheses (Besnard et al., 2013b).

The proto Greek elaíwā, the classical Greek élaion, the Etruscan eleivana and the Latin oleum will derive from e-ra-wo, the transliteration of the Linear B ideogram for olive oil, following the northwestern diffusion route of the olive tree. It should be noted, however, that élaion has been proposed to derive from the Akkadian sostantive ulû = finest, best oil, or from the adjective ellû = clean, pure (Cardini, 2018), just like ewł, the Armenian word for olive oil -despite being generally considered a Greek loanword- (Nielsen, 2023) or from the Akkadian sostantive ellu (ilu/ili in Sumerian) meaning "sesame oil"; possibly deriving from the Dravidian word for sesame (eḷḷ or eḷḷu), by meaning the prevalent type of oil (Bedigian, 2004). A less accredited theory that e-ra-wo would have been developed from the Pelasgian word *loi-ou-m/loi-wo- meaning greasy or *1ei- ‘drip’ (Panagl, 2004; Trumper, 2014).

Despite the first “non-structured” contacts should date to the beginning of Chalcolithic (Cazzella and Maniscalco, 2012).

Whose analysis showed that today’s olive leaves have about 33% fewer stomata as a likely response to the increase in atmospheric levels of carbon dioxide (Beerling and Chaloner, 1993).

An indirect confirmation can be found in Chapter 135 of Pseudo-Aristotle's De Mirabilibus Auscultationibus (On Marvelous Things Heard) describing the trade of olive oil in change of silver of the first Phoenicians who sailed to Tartessus (a city-state in southern Spain) in the IX century BCE, eventuality confirmed in Book V, Chapter 35, of Diodorus Siculus’ Bibliotheca Historica (Gill, 1988).

For attribution problems of tablet BAM 494 to the -so-called UGU- first chapter, see Bácskay and Simko (2017).

https://www.britishmuseum.org/collection/object/W_K-2354 ; https://www.britishmuseum.org/collection/object/W_Rm-116;

Body-oiling with olive oil was a common practice in ancient Greece. According to Amouretti (1986), in the VII-IV century BCE Greece, a free citizen attending the gymnasion might have been able to consume up to a maximum of 55.5 liters of olive oil, mostly destined for body care (30 liters) and food purposes (20 liters).

Stirrup jars have been commonly found in Minoan and Greek tombs (Leonard et al., 1993) so that the anointing of the corpses has been associated to sanitary / purification practices, but in cinerary contexts it has been proposed to be further linked to the speeding up of the cremation process (Cultraro, 2012).

mi mlaχ mlakas larθus elaivana araθia numasianas (“I am the fine oil jar of Araθ Numasiana, for the good Larθu”)

mlakas se la aska mi eleivana (“I am the oil-flask of the beautiful Sela”)

A recent publication reporting the radiocarbon dating of charcoal samples would pre-date the begining of iron smelting in Tamil Nadu, India, in the IV millennium BCE (Rajan and Sivanantham, 2025).

Although early iron production by the Hittites was particularly developed during the imperial period, no written records have survived on the technologies employed (Cordani, 2016)

Olive oil was found (McGovern et al.,1999) to be employed in the funeral banquet of king Midas (ca 700 CE).

and, subsequently, the Lydians (V-IV century BCE; Xenophon, Cyropaedia, VI, 2)

Obviously, references to food or lighting use are also present (e.g. Genesis 49:20; Exodus 25:6; Leviticus 2:1-2, Leviticus 11:34).