Blog

Jul 05, 2023

Un multi

Rapports scientifiques volume 12,

Scientific Reports volume 12, Article number: 12203 (2022) Cite this article

3313 Accesses

5 Citations

31 Altmetric

Metrics details

This study investigates changes in dietary practices and subsistence strategies in Bronze Age Italy integrating isotopic analyses with archaeobotanical and archaeozoological data. By investigating food habits, we contribute to reconstructing human lifestyles and highlighting possible links with the economic/social organization when the rise of stratified societies and new economic activities affected subsistence practices. Stable isotopes analyses in humans and animals were performed on 6 Italian sites dating from 2300 to 900 cal. BC, followed by a complete review of additional 19 sites, which forms the basis of a diachronic and geographic comparison for Bronze Age Italy. The geographic analysis shows a more varied diet in northern and central Italy, compared to the south. The diachronic analysis highlights the homogeneity of food habits during the Early Bronze Age, contrary to the later phases when an increase in dietary variability and a higher animal protein consumption are revealed. The Middle Bronze Age appears as a pivotal moment in protohistoric societies, a phase of transition. The consumption of different foodstuffs highlights the importance of cultural exchanges, resulting in a sort of "food globalization", although environmental and climatic fluctuations could also have affected dietary patterns, favoring some crops over others.

The Bronze Age in Europe is a cultural conundrum. According to Heyd1, it lies between a supra-regional cultural phenomenon and very local declinations of the same phenomenon. With regional pottery styles and metallurgy dominating the archaeological debate, less attention has been paid to other cultural practices such as subsistence strategies, with the general agreement that subsistence in Bronze Age Europe relied on neolithic plants and animals which would have been exploited in a almost unaltered combination2,3. Any speculation on the Bronze Age diet would be made as part of a reflection of social complexity in the food regime, in association either with age/gender or, more generally, social inequalities4. For the Bronze Age, more often than for earlier periods, food has been investigated for its ability to mark distinctions or to convey political/social agreement5.

Many of the studies on Italian subsistence strategies for late prehistoric contexts have shown considerable regional variations. However, any attempt to explore differences among social categories has only revealed scattered results. Furthermore, one of the main problems is that most of these studies focus mainly on a single proxy, e.g. archaeobotanical evidence, providing only a partial view of dietary trends.

The aim of this research is to reconstruct Bronze Age dietary behavior as well as detect possible changes in the human subsistence economy putting together the growing archaeozoological, archaeobotanical and isotopic literature on prehistoric Italy. This is possible through an integrated approach examining new and existing data from a broader perspective. Specifically, this paper aims to present a comprehensive overview of Italian Bronze Age human diets, bringing together a set of published isotopic records and new data, in order to reconstruct geographical variation within the Italian Peninsula and diachronic changes during the Bronze Age. The possibility that environmental constraints would have driven local variation and regional differences and the link between cultural traditions and local ecosystems, will also be explored and evaluated.

The most common plant species exploited during the Italian Bronze Age are barley (Hordeum vulgare), einkorn wheat (Triticum monococcum) and emmer wheat (Triticum dicoccum), which belong to the so-called northeast Neolithic package6,7,8. Significant changes occurred between the Neolithic and the Bronze Age: the spread of "minor" plant species has been recorded between the end of the 3rd and the beginning of the 2nd millennium BC9. Although the chronology of these changes is still not clear, the origin of this phenomenon seems to be a consequence of the necessity to overcome climatic fluctuations, as well as new social practices as attested from the second half of the 3rd millennium, as suggested in other Eurasia areas10. Likely linked to an increase in demand due to the raise of the population size, new technologies, like irrigation, crop rotation, manuring and specific agronomic strategies, were introduced in continuity with regional agricultural systems both in the Alpine and in the southern regions like Apulia, where most of the archaeobotanical southern Italy investigations took place11,12,13,14,15. It is from this point on that the spectrum of cereal and common edible plants becomes more diverse. The northeast traditional cereals were still cultivated, but in addition to these, species such as broomcorn millet (Panicum miliaceum), foxtail millet (Setaria italica), rye (Secale cereale), spelt (Triticum spelta), Avena sp. and legumes (e.g., Pisum sativum, Vicia ervilia, Lens culinaris) spread over the peninsula6,16,17.

Crops such as oats and rye are documented as cultivated only in more recent periods (Iron Age)6. Hence, such findings during the Bronze Age have been identified as weeds6,18,19. The limited botanical record does not shed light on the importance of these plants in people's livelihoods. The major findings are from the south of the peninsula, whereas the finds from the northern areas suggest these plants were mainly weeds. As for spelt, consistent evidence for its cultivation has only been recorded from the beginning of the Bronze Age. Spelt has been found at different sites in northern Italy, but its distribution is not yet well defined. Its presence in Trentino Alto Adige is probably due not only to its hardiness but also to the influence of Nordic countries, as this cereal seems to be much more widespread in central Europe than the Italian Peninsula20.

Particular attention must be devoted to millets, namely broomcorn and foxtail millet. The growing importance of these species during Later Prehistory calls for a critical reappraisal. An ever increasing number of botanical and isotopic discoveries all over Europe underscore the importance of these plants in human life21. Studies also highlight the spread of millets in Italy, supporting its substantial role in the human diet from prehistory to historical periods22,23,24,25,26.

Despite these findings, the spread of broomcorn and foxtail millet in Italy is still unclear. Identifying seeds from Neolithic sites is not always straightforward (e.g., Valgrana, Levata di Curtatone)9,27 and findings are often limited to a few seeds (e.g., Castello d’Annone, Capo Rondinella, San Domenico, Gr. della Tartaruga)27,28. Nevertheless, the archaeobotanical data from northern Italy suggest that broomcorn and foxtail millet were an important complement to cereal production during the Bronze Age9,15. The evidence is different for the center and the south of the peninsula, where botanical remains for these crops are rare7,16. Furthermore, archeobotanical studies are unable to determine whether a certain crop was for human or for domestic animal consumption. Conversely, isotopic analyses of human and animal samples are a useful tool in identifying millet intake, as plants with C4 photosynthesis, like millets, can be clearly distinguished from other cereals that belong to the C3 group. Indeed, stable isotope analysis provides relevant information on the distribution and consumption of millets and at the same time paves the way for new questions about the dynamics of the evolution of farming practices. It is in this context that research conducted in Italy during such a critical period is essential to explore the introduction of these species into the Mediterranean area.

Another C4 plant that occasionally played a significant role in Italy is the cockspur grass (Echinochloa crus-galli). Less widespread than broomcorn and foxtail millet in archaeological contexts, its ecological characteristics are similar to those of broomcorn and foxtail millet, except that it is better suited to wet environments. Substantial amounts of Echinochloa have been recorded for Bronze Age sites in northern Italy, Solarolo and Castellaro di Vho9,19. At present, there is no evidence of its presence in central and southern Italy16.

Research on animal husbandry in Bronze Age Italy reveals the existence of diversified settings28,29,31. The different distribution of domestic species depends on geographical and environmental physiognomy and traditional local models responsible for diversified fauna management strategies29,30,31,32,33,34,36.

At present, it is possible to draw a diachronical picture of animal husbandry for the different phases of the Bronze Age only for a few sites32. However, some general trends can be outlined. First, from the MBA there was a general increase in the exploitation of secondary products from domestic mammals30 leading to more sedentary animal breeding33. Secondly, hunting became increasingly marginal and cattle were predominantly slaughtered in adulthood during the Bronze Age37.

Animal husbandry practices in northern Italy during the Bronze Age were varied. Within this marked heterogeneity, cattle seem to experience the greatest variations in percentage ratios. They range from 10% in Cattolica38 to 50–60% in Sonnenburg39, compared to the slightly more stable percentages for sheep, goats and pigs34. In the Alpine area, there was a general abundance of caprines (sheep and goats) and a shortage of pigs36. In the South Tyrolean and Trentino sites, sheep were preferentially selected as livestock because of the geography, e.g. Sotćiastel35,40. Nevertheless, cattle represented a significant component of the local husbandry economy as well, e.g. Val Pusteria29,32,33,35,39,41. In the Garda regions, there appears to have been greater variability in term of subsistence strategies (agriculture, livestock, harvesting of wild fruits, fishing, hunting) thanks to the heterogeneity of the available resources. Indeed, this area appears to be the only one for which caprines, cattle and pigs are equally represented41. In the area north of the Po River, and particularly that occupied by pile-dwellings, all the domestic species are well represented. To a limited extent, cattle or sheep and goats breeding dominated in some areas where pigs were less prominent than in the subalpine and Garda areas29,33,42. Moreover, some settlements suggest that cattle breeding increased during the RBA31,32,34. South of the Po, where Terramare settlements spread, the economy was substantially pastoral as caprines were the main resource, such as in the pile-dwelling villages of Trentino29,33,34,43.

As for elsewhere in the peninsula, husbandry practices are less well-known, due to the limited number of studies.

In central Italy (Toscany and Lazio), husbandry was more focused on cattle, sheep and goats and less on pigs, e.g. Petrosa44,45. In this area at the beginning of the MBA, the animal economy was focused on the breeding of cattle, however from the late MBA husbandry strategies changed and caprines became predominant. This trend continued throughout the RBA and FBA up to the early Iron Age, when pigs become more significant46. During the last phases of the Bronze Age, pastoral activities seem to increase and caprines become the main species at the sites of Marche and Abruzzo47. Research in the Marche region during the MBA suggests a mixed type economy. Nevertheless, some settlements may have had a more pastoral economy, while others had subsistence strategies based on agricultural practices associated with cattle and pig husbandry48. For this region too, it seems that there was a marked increase in caprines mainly used for meat. Moreover, it seems that the surplus of sheep could be interpreted as a prestigious goods of the emerging elites43.

In southern Italy, studies on sites such as Broglio di Trebisacce49, Torre del Mordillo50 and Coppa Nevigata51 allow a diachronic analysis of of changes in the animal economy. However, the lack of information for the EBA constitutes a substantial gap52. Higher percentages of caprines compared to cattle and pigs have also been observed in southern Italy for the MBA and RBA where sheep were predominant43,52. Even though they were mainly exploited for meat, secondary products were important for the local economy, as recorded in Puglia and Sibaritide49,50. Notwithstanding, the sites with high percentages of cattle are numerous, e.g. Torre dei Passeri53, Monopoli54, and Bari S. Maria del Buon Consiglio55. These animals were no longer being reared exclusively for meat and other species such as sheep or pig also provided good quality meat. Occasionally game, i.e. deer, may also have purposefully supplemented meat production37. The increase of the number of calves in the animal records suggests an evolution of the dairy industry and the general prosperity of settlements43. In some of the southern sites, hunting was still significant for the local economy54,56 and seems to have intensified at the end of the Bronze Age. This phenomenon could be linked to increased demands for meat following a rise in the exploitation of secondary products43.

Evidence for aquatic species is rare for the Bronze Age sites of the Peninsula. When present, mainly freshwater and euryaline species are found in the north and in central Italy57. Pike and turtle have been found in the Lagazzi pile-dwellings and fishing is attested by tench, pike and rudd remains at Canàr57,58. Dating to the FBA and early IA, abundant remains of tench and pike have been found in Frattesina57. In the South of the Peninsula, the aquatic remains are mainly of saltwater species56 and the main discoveries are from the Apulia coast (Monopoli, Roca, Punta Le Terrare)56.

The cemeteries of Pertuso, Buco del Diavolo, Ostiglia La Vallona, Gr. Vittorio Vecchi, Trinitapoli and Castiglione presented here for the first time are shown in Fig. 1 and Table 1. The selection of the sites was driven by the aim at increasing the representability of each chronological period and geographical area. It takes into account the limitations due to lesser explored areas, i.e. the center and south of the Peninsula, and the changes observed in funerary practices with the passage from inhumation to cremation during the MBA/FBA, which curb the analysis of bone collagen. The paucity and heterogeneity of the information provided by some of the newly discovered sites restricts this study, nevertheless they contribute substantially to the reconstruction of a dietary overview.

Location of the Italian Bronze Age cemeteries for which stable isotope analyses on humans are available (the sites from the literature are in black and the sites analysed here are in red). The map is elaborated using "Natural Earth. Free vector and raster map data @naturalearthdata.com" available at https://www.naturalearthdata.com/downloads/10m-raster-data).

The sites of Pertuso, Buco del Diavolo, Ostiglia La Vallona, Gr. Vittorio Vecchi, Trinitapoli and Castiglione, located in the north, center and south of the peninsula, span from EBA to the FBA/IA. These cemeteries include 164 humans and 51 domestic and wild animals recovered in association with the human remains, then they are all coeval. Each archaeological site context is detailed in Supplementary Text A.1.

These sites are compared with 19 sites previously investigated (Fig. 1, Table 1): Ballabio, Arano di Cellore, Mereto, Sedegliano, Gradisca di Codroipo, Dossetto di Nogara, Olmo di Nogara, Bovolone, Fondo Paviani, Gr. Misa, Gr. Scoglietto, Gr. Regina Margherita, Felcetone, Lavello, Toppo Daguzzo, Murgia Timone, Gr. della Monaca, P. di Zambrone and Gr. dell’Antenato25,26,58,59,60,61,62,64.

The carbon and nitrogen isotopic ratios from Pertuso, Buco del Diavolo, Ostiglia La Vallona, Gr. Vittorio Vecchi, Trinitapoli and Castiglione are presented in Supplementary Table A.1 and summarized in Supplementary Tables A.2 and A.3. The state of preservation is detailed in Supplementary Text A.2 and Table A.4.

To minimize the environmental differences, the humans have been compared considering the offset between the median of the humans and that of the associated domestic animals for each site to evaluate differences in the diets of human communities (Δ13Ch-f and Δ15Nh-f). Indeed, considering that the diet–tissue trophic enrichment factor is estimated to be around 3–5‰ for the nitrogen and 0–1‰ for the carbon per trophic level66,67, the offsets between the human and fauna are good indicators of a community diet (protein intake).

The sites in Liguria, Pertuso and Buco del Diavolo, are dated to the EBA and FBA/EIA respectively. The two animal groups present typical ranges for a temperate environment predominantly dominated by C3 plants. When considering the herbivores of both sites, no significant differences occur for both δ15N ​​and δ13C values (δ15N, p = 0.361; δ13C, p = 0.854). This suggests that the two sites were placed in a similar environment and that no great environmental and/or paleoclimatic changes occurred during the Bronze Age in this area and that herbivores consumed similar plants. Since the two animal assemblages show the same food pattern, the two human groups are directly comparable. If any difference in dietary patterns appears between the two groups, it will likely be the result of food preferences rather than of a "different environmental offer".

The δ15N human median is 7.2‰ (min = 6.5‰, max = 7.9 ‰, n = 14) and the δ13C median is − 20.6‰ (min = − 21.0‰, max = − 20.2‰, n = 14). The isotopic values show the consumption of a mixed terrestrial diet with resources typical of C3 plants environment. According to the Mediterranean marine resources isotopic data68,69, humans did not routinely consume seafoods. The human δ15N and δ13C results are spread in narrow ranges, suggesting a homogenous diet. Considering the δ15N and the δ13C human and domestic animal medians, the Δ15Nh-f and Δ13Ch-f are 3.5‰ and 0.2‰ respectively, values that are included in the recurring offsets between two different trophic levels. The Δ15Nh-f result indicates that the animal proteins played an important role in the diet of these individuals. When considering adults (n = 8) and juveniles (n = 4), significant statistical differences emerge for both δ15N (p = 0.048) and δ13C (p = 0.016). The juveniles show lower δ15N and δ13C values compared to the adults. A lower animal protein intake, possibly associated with vitamin and mineral deficiencies, may have contributed to weakening the youngest individuals, making them more vulnerable and inclined to disease and death, as supported by pathological evidence as well (Harris lines)70. Paleopathological analyses have also shown that adults present stress marks linked to their growth, possibly related to food deficiencies as well (dental hypoplasia, cribra cranii and orbitalia)70. All evidence seems to support that this community was exposed to repeated stress, especially during childhood. Similar cases in which the δ15N and δ13C results of juveniles are lower than for the rest of the population have been recorded in other sites in Liguria and in southeastern France for the Neolithic71,72.

Studies conducted on western Ligurian attest that Bronze Age communities were composed of small family groups73. Osteological studies on Pertuso humans seem to confirm this hypothesis, as olecranon perforations have been recorded in a number of individuals, supporting a certain degree of genetic links. The main settlements were in all likelihood scattered in naturally defended positions or consisted of fortified sites which controlled the surrounding area, e.g., Castellaro di Uscio74. As such, the archaeological evidence indicates an economy based on livestock and on the available local resources, as highlighted by the hunting, i.e. deer and wild boar, and gathering, i.e., apples, horns, acorns, that certainly enhances the evidence for a undifferentiated diet73.

The δ15N and δ13C human results show the presence of two groups with different dietary patterns. The first group, represented by BDA, BD716, BD1320, BD1092 and BD2453, has a median of 7.2‰ for δ15N (min = 6.6‰, max = 7.5 ‰, n = 5) and − 19.8‰ for δ13C (min = − 20.0‰, max = − 18.5‰, n = 5). Taking into account the local domestic animals, humans have an enrichment of 3.7‰ for nitrogen and 0.8‰ for carbon. These offsets indicate a varied mixed terrestrial diet, which includes both plant—mainly C3, and animal resources, with the exception of BD1092, whose δ13C value (− 18.5‰) suggests a mixed C3–C4 plants consumption. It is likely that meat and dairy products originate principally from sheep and goat, the most represented at the site65. Nevertheless, the contribution of other domestic species, like cattle and pigs, should not be underestimated. The similar isotopic results of these species and caprines makes it difficult to identify which species were predominantly consumed by the humans.

The second group includes BD876, BD1132, BD882 and BD794. The median values are 8.7‰ for δ15N (min = 8.5‰, max = 9.1‰, n = 4) and − 15.6‰ (min = − 16.5‰, max = − 15.0 ‰, n = 4) for δ13C. For both isotopes, significant offsets are evident between the humans and the domesticated animals (∆15Nh-f = 5.2‰ and ∆13Ch-f = 5.0‰). The relatively high δ15N values suggest a substantial animal protein intake and aquatic foodstuffs could have been part of the diet even though fish remains have not been found in the site. The δ13C values, enriched in 13C, support an important C4 plant consumption. Considering the carbon and nitrogen ratios, marine foodstuffs could have been part of the diet as well, but the altitude and the geographical position of the site make this option quite improbable. This evidence confirms that millets played an essential role in the diet, showing important differences in food choices between the two groups. The first archaeobotanical evidence for millets in Liguria comes from the MBA (Bric Tana)75, however the most consistent finds are dated to the FBA-early IA (Monte Trabocchetto)17. Consequently, considering the radiocarbon dates of the second group65, the δ13C values enriched in 13C support an important consumption of millets from the FBA-IA transition, consistent with the dates of the archeobotanical results.

The presence of two groups with different diets at Buco del Diavolo is not dependent on sex or age of the individuals because both groups consist of men and women of different age. Therefore, the first explanation to justify this variability could be related to social reasons. Food choices can be a sign of individuals’ differential access to resources, making these choices indirect evidence for the stratification of the community. Lightfoot et al.76 highlighted the consumption of millets for some individuals during the Bronze Age (Nadin-Gradina, Croatia) and proposed millet as a low status foodstuff after considering the relationship between δ13C and burial type. At Buco del Diavolo, some bracelets and a bronze torque are indicative of people of high social status. Unfortunately, as skeletal remains were not articulated, it was not possible to associate the grave goods to the individuals, but it is possible that such differences may also be reflected in diet.

The second explanation could be related to cultural reasons. A viable alternative is that the two groups had different origins and in accordance different dietary habits. Millet was already consumed since the MBA in northern Italy ​​(Verona area)59. Therefore, it is likely that some individuals from this area had moved to Liguria. Some archaeological elements typical of the southwestern alpine region are present in the Po basin, in the Provencal areas, in Switzerland, Germany and in southeastern France (RSFO Group of the Urnfield Culture), as attested by bronze grave goods77. Thus, the growing exchanges led to an increasing mobility, which could be reflected in different diets.

A third explanation could be associated to the different chronology. The first group is dated to the FBA, while the second group to the beginning of the IA65. It is therefore possible that the diets could be the result of different agricultural practices during the two periods: cultivation of C3 plants during the FBA and more C4 plants during the IA. Moreover, a rapid climate change between 1550–550 cal. BC across the Northern Hemisphere has been recorded78 and the paleoclimatic fluctuations responsible for the increase in drier conditions could have led to the diffusion of new crops, like millets. In addition, the spread of human activity affected the ecosystems leading to a decline in woodland cover, the expansion of shrubland and grassland and an increase in soil erosion. There is also evidence for burning from the Late Bronze Age onwards79. At present, none of these hypotheses can be excluded, further research on additional isotopes, such as Sr to study the mobility, should provide a clearer picture.

The δ15N human median is 10.7‰ (min = 10.1‰, max = 13.3‰, n = 5) and the δ13C median is − 13.0‰ (min = − 13.3‰, max = − 12.6‰, n = 5). The offsets between humans and animals are 4.4‰ for nitrogen and 8.0‰ for carbon. These results indicate an important consumption of enriched 15N and 13C resources. The high carbon ratios suggest C4 plants consumption like millets, as confirmed by coeval and nearby communities (Olmo di Nogara, Bovolone)25, whereas the enrichment in 15N could highlight an important animal proteins intake as well as aquatic foodstuffs. Indeed, local communities have exploited the aquatic environment since the EBA-MBA transition, as confirmed by the freshwater animals recovered in the pile-dwelling and Terramare settlements57. The isotopic values of freshwater fish from Fondo Paviani25 are further evidence of the possible contribution of these resources to the human diet.

Moreover, plants enriched in 15N, like the modern cattail (Typha latifolia), a species typical of marshy environments (δ15N \(\overline{{\text{x}}}\) = 7.0‰, n = 79)80 could have been part of the diet, and its consumption is attested since the Mesolithic81. Off-site palynological studies in the lower Veneto plain and in Lombardy show a higher rate of afforestation (30–50%), with extensive presence of wooded or marshy areas in the Bronze Age81,82,84. Consequently, wetlands with their typical fauna and vegetation, like bulrush12, were potentially exploited for food resources.

According to the archaeological fish isotopic data from the Mediterranean68,69, marine resources could have been consumed as well. If marine foodstuffs were consumed, this could lead to the enrichment of both human δ15N and δ13C values, consequently limiting the importance of C4 plants in the diet. However, when considering the local archeozoological evidence and the geographical position of the site, a freshwater food intake is more likely alongside an important consumption of C4 plants.

Notwithstanding the limited number of individuals, no dietary differences were detected according to age groups. The only exception is OLV371sub, who shows an enrichment in 15N compared to rest of the group but the young age of this individual (1–1.5 years old) suggests that this could be associated to breastfeeding. The only individual whose sex has been estimated is OLV183, determined as male. This individual has the lowest δ13C and δ15N values. Given the lack of sex data for the other individuals, it is not possible to compare diet between the two sexes.

The δ15N human median is 6.4‰ (min = 5.7‰, max = 6.7‰, n = 4), and the δ13C median is − 19.6‰ (min = − 20.1‰, max = − 19.2‰, n = 4). Concerning the ∆15Nh-f, an enrichment of 1.8‰ is recorded. This value is lower than expected, suggesting that the animal protein intake was rather limited. In addition to this, even though the assemblage is based on a low number of records, it seems that legume consumption was likely significant. This hypothesis would be supported by the botanical findings. Vicia faba seeds represent 57.3% of the total seeds (2182/3811)85. Notwithstanding, it is difficult to assess whether the legume intake was direct or indirect, this because some herbivores show ​​low nitrogen values too and they were likely fed with legumes.

Concerning the ∆13Ch-f, an enrichment of 0.9‰ is recorded. This result suggests that animal proteins were part of the diet. However, the small offset recorded for nitrogen and the likely important legume intake highlight that meat and dairy products were probably not a significant component of the diet.

The carbon and nitrogen data indicate that the individuals we analysed show a homogeneous terrestrial diet as the δ13C and δ15N are within a narrow range. The plant component was significant and mainly represented by cereals and legumes, as archaeobotanical studies confirm. Based on the available data, the δ13C and δ15N values ​​do not seem to support the consumption of aquatic resources. Further isotopic investigations conducted on coeval individuals of this area will be necessary to confirm this main dietary pattern where a low animal protein intake and a likely important legume consumption is highlighted. Interesting, some new data from the area seem to confirm it86.

A comparison of the adults with the only adolescent shows the latter had the lowest nitrogen value (5.7‰) and the maximum carbon value (− 19.2 ‰). It is likely that this individual had a smaller animal protein intake than the others, but since the contribution of legumes to this group's diet is important, the picture could be much more complex.

The data obtained by the present study are considered together with the seven individuals analysed by Arena et al.63. The δ15N human median is 8.3‰ and δ13C median is − 19.5‰. The δ15N results range between 6.5‰ and 10‰ (n = 14) and δ13C values ​​between − 20.0‰ and − 18.7‰ (n = 14). Considering the human and animal offsets, ∆13Ch-f is 1.0‰ and ∆15Nh-f is 1.6‰. The carbon enrichment fits into the range between two consequent trophic levels, while the nitrogen one is lower than expected. These results suggest that, even if terrestrial foodstuffs from a C3 environment were mainly consumed, animal proteins were not a major source, indicating a diet mainly based on plant proteins, including legumes, as attested by the archaeobotanical records of the area7.

Despite the consistency of the results, TRIAB20 and TRIAB22, differ from the main group, showing lower δ15N values and highlighting a lower animal protein intake. Tafuri et al.59 identified a male with similarly low values at Toppo Daguzzo, suggesting the consumption of similar resources. Several hypotheses can explain these data: (I) a possible common origin for the individuals, explaining analogous dietary patterns; (II) a similar social status—the grave goods associated with the Toppo Daguzzo human indicate this individual was probably a warrior as he was buried with weapons59. Regarding the two women of Trinitapoli, it is difficult to associate the grave goods with the individuals because of the collective burials, thus it not possible to define social status except in specific cases87. Further investigations to reconstruct their origins could provide additional elements helping to explain these dietary differences.

No significant differences occur between the dietary patterns of the humans from the two sectors of the necropolis, AB and C (δ15N, p = 0.805; δ13C, p = 0.736). Similarly, no significant differences occur between individuals of different sex (δ15N, p = 0.28; δ13C, p = 0.97).

The δ15N human median is 7.5‰ (min = 6.3‰, max = 9.3‰, n = 61) and the δ13C one is − 19.2‰ (min = − 19.7‰, max = − 18.9‰, n = 61). These values are indicative of a terrestrial diet and, even though no fauna is available for this site nor for the whole island for Later Prehistory, results show that most of the protein intake comes from plants from a C3 environment. Nevertheless, humans show a large δ15N range suggesting a diversified intake of animal protein. As already highlighted for Trinitapoli, the individuals with the lower nitrogen values could indicate they had a significant legume consumption. The importance of agricultural foodstuffs is supported by studies on caries and other oral observations on the teeth of these individuals88. Statistical tests between individuals of different burials show no significant differences for δ13C (p = 0.355, H = 6.683) but differences emerged for δ15N (p = 0.003, H = 17.756). Significant differences have been detected between the following burials: TB96-TB93 (p = 0.035), TB95-TB93 (p = 0.046), TB119-TB96 (p = 0.041) and TB119-TB95 (p = 0.028) (p-values corrected by FDR). It is likely that the tombs reflect social organization, highlighted by a differential access to resources. For instance, the presence of a bossed bone plaque (ossi a globuli), interpreted as a prestigious object, in TB93, could imply that the individuals in this burial were of higher social status than those in the other tombs. Another hypothesis suggests a different origin for some of the individuals. Archaeological data support contacts between Sicily and the Aegean area89,90. Moreover, the general heterogeneity of the metric traits and, at the same time, similarities of some of these traits between Castiglione and groups from Greece and other Mediterranean countries, support migration events91, as also highlighted by other of southern Italy sites92.

Sex and age were estimated only for some individuals. Nevertheless, these parameters do not seem to have influenced food choice. Due to the small number of assessments, statistical tests are only possible for age categories (adults vs. juveniles) and no significant differences occur (δ15N p = 0.569; δ13C p = 0.56), advising that age did not affect food habits.

The present Bronze Age Italian dataset includes 25 sites, which have provided 355 human and 148 terrestrial and aquatic animal samples. The diachronic overview is presented below, while the geographical comparison, only partially presented in other papers61,63, is described in detail in Supplementary Text A.3, Table A.5–A.7, Fig. A.1–A.3.

The individuals of different ages and both sexes are from diverse funerary sites (caves, cemeteries, dwellings) and burial types (collective and individual). Domestic and wild animals establish a dietary baseline for the humans. However, in order to ensure consistency, the offsets between human and animal medians have been based only on the domestic herbivores and omnivors (Supplementary Table A.8). All the published data from Bronze Age sites included in this study are summarized in Supplementary Tables A.2 and A.3.

The Early Bronze Age sites, Arano, Ballabio, Dossetto, Mereto, Pertuso, Sedegliano, Gr. Scoglietto and Castiglione, have provided a total of 165 human and 47 animal samples, animal data being provided by six sites (Fig. 2).

The δ15N and δ13C human and animal values for the Italian Early Bronze Age sites.

Before comparing the humans, the domestic and wild herbivores and omnivores of each site were statistically compared. Significant differences occur among these groups (Supplementary Table A.5), likely linked to the geography and the landscape where the cemeteries are situated. Nevertheless, the animals mainly present a C3 terrestrial diet with only small variations, since most of the animal ranges of the sites overlap. The nitrogen values span from 2.8 to 6.1‰ (excluding the fox from Gr. Scoglietto), indicative of the consumption of mostly unmanured resources93. The δ13C values are widespread, but the 89% of them (41/46) are included in a rather narrow range, spanning from − 22 to − 19.2‰. The five outliers, four cattle (two from Arano, one from Mereto and one from Gr. Scoglietto) and one deer (from Arano) have high δ13C values, indicating a C4 plant contribution. Concerning Gr. Scoglietto, this evidence is in line with human dental calculus studies, where the presence of millet starch grains and phytoliths have been recorded94. The recovered phytoliths may be attributed to different cultivated crops like millets, but also to barnyard grass. In the latter case, grazers like cattle could have eaten C4 plants in an anthropogenic environment. For Arano, the fact that a wild animal shows values enriched in 13C suggests that C4 non-cultivated plants where present in the local environment and wild animals could easily access them.

To minimize the environmental variability, we have compared the offset between the human and domestic animal medians to evaluate differences in the food choices (Supplementary Table A.8) and some patterns are evident. Excluding Dossetto and Sedegliano for δ13C values and Gr. Scoglietto for δ15N values for which greater offsets are recorded, most of the human groups show a homogeneous mixed terrestrial diet based on animal and C3 plant proteins. In fact, the Δ15Nh-f spans from 2.9 to 3.7‰ and the Δ13Ch-f from 0.0 and 0.7‰, ranges that fit within the known offsets between two consecutive trophic levels.

A more detailed analysis reveals a uniform distribution of the δ15N data for Pertuso, Ballabio, Arano, Mereto, Castiglione and Sedegliano. All the sites show a quite wide range, spanning from 6.3 and 10‰, likely due to a diverse meat and/or dairy products consumption within the communities. The only site which significantly differs from the others is Gr. Scoglietto, where aquatic resources intake has already been suggested26. The location of Gr. Scoglietto in a coastal zone and in a lagoon-like area may have led the population to exploit the aquatic environment at the expense of agricultural products. Furthermore, the dental calculus results support the absence of any leguminous species94. The apparent absence of these plants in the diet associated to an important intake of animal foodstuffs can explain the enrichment in 15N.

In terms of the carbon results, most sites (Arano, Pertuso, Ballabio, Mereto and Gr. Scoglietto) show δ13C values falling within a relatively narrow range, from − 21.0 to − 19.7‰. This distribution suggests a diet based on C3 terrestrial resources, such as einkorn, emmer and bread wheat and barley associated to wild species differently available according to the regions (e.g., Quercus sp, Fucus carica, Juglans regia). Sedegliano, Dossetto and Castiglione show higher δ13C values compared to the previous ones. If for Castiglione the slightly higher δ13C values could be related to its southern geographical position and to a drier environment95, regarding Sedegliano and Dossetto, the higher enriched δ13C values are probably due to millets intake. These are likely the first isotopic evidence for millet consumption in Italy and this result is essential to trace the introduction of C4 plants to the human diet.

The Middle Bronze Age includes 13 sites: Olmo di Nogara, Bovolone, Fondo Paviani, Gr. Misa, Felcetone, Gr. Vittorio Vecchi, Gr. Regina Margherita, Lavello, Toppo Daguzzo, Trinitapoli, Gr. della Monaca, Gr. dell’Antenato and Murgia Timone, for a total of 157 humans and 61 animals. This period seems to be a phase of change according to the varied results for all the humans and animals (Fig. 3).

The δ15N and δ13C human and animal values of the Italian Middle Bronze Age sites.

Both carbon and nitrogen animal group data are widespread and statistical differences occur (Supplementary Table A.5). The animals from northern Italy (Olmo di Nogara and Fondo Paviani) show higher δ15N and δ13C values compared with the sites from the center and the south (Fig. 3). In addition to the environmental and ecological differences which affect the values from the bottom of the trophic chain, the 15N enrichment in the northern animals can be interpreted as them grazing close to lands with manured crops or being fed manured plants. The carbon data fit within the typical range of values from a C3 environment, with the exceptions of Fondo Paviani, that show enriched δ13C values suggesting a C4 plants intake, being likely fed food waste25.

Given the variability in the animal results, the human diets are compared using the offsets between the humans and their associated fauna. The Δ15Nh-f span from 0.6‰ at Gr. dell’Antenato to 2.9‰ at Gr. Misa, below what is normally estimated between two trophic levels. As for the northern Italy, Olmo di Nogara and Bovolone, the Δ15Nh-f are 2.0‰ and 2.4‰ respectively, indicating that the animal protein intake was not very important in this area. The lowest offsets are found in the center and the south of the peninsula, suggesting a limited consumption of animal foodstuffs and supporting a diet mainly based on cereals and legumes. The consumption of legumes seems to have been considerable in central Italy during the Bronze Age and the variety of species is greater than in the Neolithic16, as attested by the remnants of a soup of barley, oat, einkorn/emmer wheat and broad beans found at Gorgo del Ciliegio96. Therefore, if during the EBA the site of Scoglietto does not show evidence for legume consumption neither from an archaeobotanical nor an isotopic perspective, the δ15N values from Felcetone, Gr. Vittorio Vecchi and Gr. Misa confirm the consumption of pulses from the MBA. Additional data from La Pastena cave support the importance of pulses86. Based on these results, the MBA represents an important moment of change in terms of the adoption of new foodstuffs for central Italy. For the south of the peninsula, archaeobotanical research supports an increase in the pulse varieties, although broad beans still continue to represent the main taxon7. Consequently, according to the nitrogen data for the MBA sites of the entire peninsula, animal protein was not highly consumed during this period, revealing an economy more oriented towards vegetal products rather than animal products.

Regarding the δ13C data, interesting patterns appear. The δ13C human values are widespread, from − 20.7 to − 10.6‰. A sligh variability has already been identified at the end of the EBA, but at the MBA this pheonomenon is accentatued. Three main groups with specific dietary patterns can be identified. The first includes sites from central and southern Italy: Trinitapoli, Lavello, Gr. dell’Antenato, Gr. della Monaca, Toppo Daguzzo, Gr. Regina Margherita and Gr. Vittorio Vecchi. Here, the Δ13Ch-f are between 0.1 and 1.0‰, the characteristic enrichment between two trophic levels. Furthermore, the δ13C values of these humans overlap and fall within a limited range, from − 20.7 to − 18.7‰, suggesting C3 plant consumption. The second group is composed by Gr. Misa and Felcetone. These two communities have common characteristics. First, both sites show wide Δh-f13C offsets, (GM = 2.7‰ and FEL = 1.8‰) and, second, they show a wide intragroup δ13C range (GM = 2.9‰; FEL = 3‰). The large spread of δ13C values, associated with their intermediate position (− 20.3‰ < δ13C < − 16.5‰), suggests a mixed C3–C4 plant intake for both sites. The third group is represented by Olmo di Nogara and Bovolone. Olmo di Nogara's Δh-f13C offset is 5.5‰ and Bovolone's is 3.5‰. These two groups also show the greatest intragroup variability: the Olmo di Nogara values span from − 17.8 to − 12.8‰ and the Bovolone ones from − 18.5 to − 10.6‰. Thus, millet represented a significant component of the diet of all the individuals. Furthermore, the important Δh-f13C enrichment implies that these crops were mostly ingested directly.

According to these data, different trends are evident for the MBA. Some of them are present in the whole peninsula and others are only regional. A general low animal protein intake is detected in all the human groups of this period. In fact, most of the sites do not reach the Δ15Nh-f offset of 2‰. The greatest inter- and intragroup dietary differences are recorded in the north. Here, diet is mainly based on C4 plants (Olmo di Nogara and Bovolone), with evidence for millet consumption also seen among the animals (Fondo Paviani). In central Italy, C4 plants started to play an important role in the diet from this period. Nevertheless, C3 plants remained the staple food (Felcetone, Gr. Misa, Gr. Regina Margherita and Gr. Vittorio Vecchi) with only some individuals apparently consuming C4 plants (Felcetone and Gr. Misa). Furthermore, it seems that pulses were an important part of the diet in this area (Gr. Vittorio Vecchi, Felcetone), as recently supported by other studies as well86 In the south, there is a lower intra- and intergroup dietary variability and C4 plants were not routinely consumed (Toppo Daguzzo, Lavello, Trinitapoli, Gr. Della Monaca and Gr. dell’Antenato). It is, therefore, probable that millets started to be routinely consumed from the north to the south of the peninsula, with a likely gradual replacement of the more common wheat and barley.

There are four Recent and Final Bronze Age funerary sites: Ostiglia, Buco del Diavolo, P. di Zambrone and Gradisca di Codroipo, for a total of 16 humans and 40 animals. These later Bronze Age phases are not well-represented because of the limited number of funerary sites with inhumations, following the introduction of cremation in most of Italy.

The animal nitrogen and carbon values are widespread and significant differences occur among the groups (Supplementary Table A.5). This wide variability is particularly evident for carbon values because the animals from P. di Zambrone, at the south of the Peninsula, are enriched in 13C compared to the ones from the north. Even though differences linked to the diverse environmental conditions could not be excluded because of the different latitudes, the results are more likely related to different diets. The specimens from the north had a diet mainly based on C3 plants, whereas the ones from the south, at least some of them, grazed upon seaweeds or, more likely, had a mixed C3–C4 plants consumption, in accordance with the archaeobotanical remains of P. di Zambrone97 (Fig. 4).

The δ15N and δ13C human and animal values for the Italian Recent and Final Bronze Age sites.

According to the human and animal offsets, different human dietary habits are evident. If the Δ15Nh-f and Δ13Ch-f at P. di Zambrone are not remarkable, suggesting that both humans and animals had a C3–C4 mixed terrestrial diet, the ones from Ostiglia and Buco del Diavolo are considerable, 8.0‰ and 2.1‰ respectively, indicating a human inter- and intragroup food variability and the significant consumption of 13C enriched resources. As mentioned, millets started to be consumed from the MBA and these results demonstrate that this trend continues later. Furthermore, humans mainly consumed C4 plants directly, since animals do not show any C4 signal. At Ostiglia and Buco del Diavolo, the Δ15Nh-f are 4.4‰ and 4.0‰ respectively, suggesting that animal protein and/or aquatic resources had an important role in the diet. Taking into account the geographical area of the two sites, Buco del Diavolo in Liguria, over 1000 m. a.s.l., and Ostiglia, in the Po Basin, in the center of one of the most economically developed areas during the Bronze Age, it is not surprising that such a wide variety of resources was consumed. In addition, studies on animal management during the FBA in the Po Basin show that even if the subsistence economy was based on agricultural activities, pigs were slaughtered before reaching adulthood to exploit the higher quality of meat and cattle and caprines were reared until adulthood to exploit their secondary resources42,98. These results, which are consistent with animal protein intake, indicate that specific attention was paid to animals depending on their resources and their different purposes.

The great diversity in the dietary practices for the RBA-FBA compared to the previous phases is present at the intra-group level as well, particularly at Buco del Diavolo, as previously described. This is the first time that such different dietary strategies are recorded between individuals from the same community in the Italian Peninsula.

Stable isotope data highlight drastic changes in food habits during the Bronze Age. The new results combined with published data have revealed the emergence of three phenomena: (I) the introduction of C4 plants in the diet; (II) the increasing importance of legumes; (III) the increased consumption of animal products.

The consumption of C4 plants represents the first major change in terms of new crops since the introduction of agriculture, and it has also been identified as an example of "food globalization"21. In order to trace the origins of this new practice, botanical research together with isotopic data from all of Italy have been evaluated. Some of the archaeobotanical evidence supports the spread of millets starting in the Neolithic and throughout the Bronze Age. First, there is a significant increase in the number of sites where millet can be found (Fig. 5 and Supplementary Information 2, Table B). Second, the distribution area is wider, including sites in southern Italy. Third, the findings are varied, representing both seeds and chaff. The greatest amount of millets has been found in the Po Basin, highlighting how crucial this zone was for the spread of millet throughout the peninsula (Fig. 5 and Supplementary Fig. A.4).

Map of Neolithic, Eneolithic and Bronze Age sites where archeobotanical remains of common and foxtail millet have been found (the list of the sites and the relative references are in Supplementary Information 2, Table B).

Isotopic studies in Italy show no evidence of C4 plant consumption for the Neolithic65,71,72,98,99,100,101,102,103,105, even if doubtful consumption seems to have been recorded in some individuals at Mora Cavorso106. Neither were millets commonly consumed during the Eneolithic106,107,108,110. Even if an enrichment in 13C could be induced by several causes, i.e. marine foodstuffs, from this study it is evident that increasing higher δ13C values are mainly due to C4 plants. The first evidence for C4 plants intake is detected in the EBA and EBA-MBA in the northeast of the peninsula—Dossetto and Sedegliano. All other evidence dates to the MBA in northern Italy—Olmo di Nogara and Bovolone and to a lesser extent in the central regions—Gr. Misa and Felcetone. The C4 signal is well established later on in the Bronze Age, the RBA-FBA, mainly in the north—Ostiglia, Buco del Diavolo, with some evidence for the south at P. di Zambrone (Figs. 6 and 7).

Distribution of δ13C (left) and δ15N (right) human group medians according to their chronology. Horizontal bars with blurred strokes indicate the absence of absolute dates. Horizontal well-defined lines indicate chronological periods determined through radiocarbon dates. Regression lines are calculated and weighted according to the number of individuals for each site: n = 1, w = 0.25, n = 2–3, w = 0.5, n > 3, w = 1.

The "Italian millet road". Map of Bronze Age Italian sites for which isotopic analyses have been carried out. The color intensities indicate differential consumption of C4 plants according to chronological period (The map is elaborated using "Natural Earth. Free vector and raster map data @naturalearthdata.com" available at https://www.naturalearthdata.com/downloads/10m-raster-data).

When combining isotopic and archaeobotanical data, a more detailed scenario emerges for the identification of the "Italian millet road". Where botanical studies infer the presence/absence of C4 plant species, stable isotopes analyses quantify the direct or indirect intake of these plants. In fact, archaeobotanical remains reveal the spread of millets from northern Italy since the Neolithic (Fig. 5), but the isotopic data indicates humans only started to regularly consume millets from the EBA-MBA transition, beginning in the northeast, then the center and finally the south of the peninsula (Fig. 7). Radiocarbon dates performed on charred broomcorn millet grains confirm this trend. Filipović et al.21 estimate the start of millet cultivation in the Po Basin between 1570 and 1410 cal. BC (95.4% probability) at the beginning of the full development of the Terramare culture111. Furthermore, biomolecular analysis performed on on-site archaeological sediments (i.e. Fondo Paviani) confirms the presence of broomcorn millet by the biomarker miliacin112. Millets were probably consumed in previous phases, like the dental calculus from Gr. Scoglietto suggests94, but in such a small quantity that it cannot be isotopically detected.

In southern Europe, the first evidence for millet consumption comes from Greece at Perachora113 in the Early Helladic (3100–2100/2050 BC) and at Rymnio and Spathes114, Pineiada, Aghia Triada and Almyri114,115,117 dated to the Late Helladic (1700/1675–1075/1050 BC). Croatia, Spain and France, as well as countries beyond the Alps, like Switzerland, Germany and Austria, show sporadic C4 signals, but only from the Late Bronze Age onwards. Conversely, during the IA, results are more homogeneous exhibiting the typical values of millet consumers76,117,118,119,120,121,122,123,124,125,127.

It seems that the Caucasus area represents a key zone for the spread of millets into Europe because millet intake is recorded from the first half of the 2nd millennium BC128,129. The identification of an ‘Isotopic Millet Road’ from China to Central Europe and the Mediterranean reveals that C4 dietary signatures reflect early links, in the form of migration and/or resource transfer, between the Bronze Age inhabitants of China and Europe130. These results show that the westward expansion of millet as a foodstuff into Europe was a rather fast and global phenomenon instead of a gradual one. This picture is confirmed by Filipović et al.21, whose research indicates the earliest occurrence of millet at the middle of the 2nd millennium BC, supporting its spread "at only slightly different times in different parts of Europe". Their Bayesian model shows that it was cultivated in southeastern Europe (Ukraine) at the earliest during the sixteenth century BC, promptly spreading during the fifteenth-fourteenth centuries BC to most of southeast and Central Europe and only in the late 13th/early twelfth centuries BC into northwest Europe. Consequently, the radiocarbon-dated caryopses indicate that millet did not arrive in Europe during the Neolithic and that most of the Neolithic attributions should therefore be doubted. Furthermore, considering regional variability, the first millet consumption in Europe, including Italy, and the earliest radiocarbon dates on broomcorn millets seems to overlap. This evidence suggests that the adoption of millet as part of a human diet was heterogeneous and was contingent on the area and the site.

Why are new crops like millets introduced? An important demographic growth associated to an increasing socio-political and economic connectivity between the Mediterranean area and continental Europe may provide the answers. This connectivity has been crucial not only for the exchange of ideas, lifestyle and technologies but also for objects, products (crops and animals) and resource transfer (e.g., raw materials, prestigious goods), as demonstrated by archaeological evidence and confirmed by bioarchaeological and genetic data130,131,133. Contacts between the Carpathian basin and northern Italy, through Austria and Slovenia, have been well identified, particularly from the Bronze Age134, and millet was probably part of this "package exchange". Millets were likely introduced to diversify the agricultural production and to ensure harvests when annual winter wheat and barley crops were not sufficient.

Pivotal agricultural system changes have been recorded since the Eneolithic: these mainly consisted in the transition from a self-sufficient horticultural subsistence strategy to a more extensive agricultural system, where intensive production was managed by a more complex society and when a gradual demographic growth is recorded. The ever increasing areas dedicated to agricultural and pastoral activities caused heavy deforestation with the over-exploitation of land and caused the soil to dry up. This aridity led to the introduction of innovative strategies, including a network of irrigation ditches as well as the cultivation of less demanding plants, such as millets14,135.

In addition, episodes of dryness are recorded in the EBA, as different proxies in the Italian sub-alpine lakes show84,136, they are also observed as a prominent warm spell in the residual ∆14C curve. These short events could have affected variations in crop cultivation. Therefore, the increasing aridity caused by climatic fluctuations associated to heavy deforestation and the over-exploitation of soils12,14,137, supports the introduction of new, more tolerant species, less sensitive to stressful conditions.

However, some questions arise: why was millet mainly consumed only in the north of Italy? Why was millet differently consumed within groups (e.g., Buco del Diavolo)? According to the evidence from other European sites, variable intake could be related to social, cultural and geographical factors118,120,121,123,127. Indeed, millet consumption is attributed to lower social status individuals in Croatia and Germany76,123. Nevertheless, the C4 signal, when detected in the Italian sites, is usually recorded for all the community's individuals, like at Olmo di Nogara, Bovolone and Ostiglia, even if this last site is just represented by few individuals. It is, therefore, likely that millet was easily accessible in Italy.

The second phenomenon concerns the growing importance of legumes both in diversifying the human diet and for improving the fertility of the soil through crop rotation9. Pulse cultivation was mainly based on broad bean all across Italy and, depending on the site, significant quantities of pea, grass pea, red pea, lentil, common vetch and ervil have been found6. Indeed, in addition to a greater abundance of legume seeds, there is also a wider variety of species, especially compared to the Neolithic138. If this picture is quite clear for the north, it is less evident for the center and the south of Italy. Despite the increasing number of taxa, the presence of legumes is often controversial. Sometimes it is not possible to ascertain whether these plants were weeds or were cultivated for their physiological characteristics6,7. Isotopic analyses contribute to discriminating between legumes as part of the diet or not and determining whether they were cultivated or not. Indeed, legume consumption can be detected when associated with a low animal protein intake. This is the case of Gr. Vittorio Vecchi. Furthermore, the archaeobotanical data from this site show that bean seeds represent more than 50% of the total remains16,139, revealing the significance of legumes for this community. Consequently, these new data, associated with other coeval sites of the region and additional unpublished data86,110, provide evidence for the increasing importance of legume consumption during the Bronze Age for central Italy. This site clearly showcases how multidisciplinary studies that combine isotopic data and archaeobotanical research are essential for the reconstruction of subsistence practices and for defining the significance of these plants in human diet.

Finally, the third evidence is a rise in animal protein consumption. Indeed, a gradual enrichment in 15N in the individuals of the northeast, northwest and in the south throughout the Bronze Age is recorded (Fig. 6). The higher nitrogen values could be attributed to different causes, like the consumption aquatic resources as well as the increase in aridity recorded from the EBA onwards cannot be underestimated. However, the global increase of dryness could have negatively affected crop cultivations, explaining the greater interest in meat and secondary products highlighted for the end of the Bronze Age. In northern Italy, pollinic analyses indicate that only 30–40% of the total deforested area was used for agriculture and the rest was for livestock breeding140,141. Indeed, RBA Terramare sites, like S. Rosa di Poviglio and Montale, show an increasing number of caprines at the expense of pigs and cattle. Goats prefer open spaces and can tolerate drier pastures. Therefore variations in subsistence practices can be due to readjustments of economic strategies in reply to changes in the ecosystem. Towards the end of the Bronze Age, a worsening of the agricultural yield is recorded at Montale and Rosa14. Cereal cultivation significantly declines, while the uncultivated areas spread142. It is therefore possible that several years of dryness, in conjunction with economic and social factors, contributed to precipitating a situation that was already beyond the thresholds of sustainability, pushing the population to invest in new resources, like in a pastoral economy for meat and secondary products.

Even though the species exploited as well as the animal size varied according to geographical area143, research into livestock farming in Liguria, for instance, supports an increased interest in the pastoral economy. During the Neolithic, during times of important demographic growth, pastoral practices, which included the use of caves as stables and suckering to provide foliar fodder, were introduced144,73. From the end of the Neolithic, transhumant pastoralism spread and contributed to the thinning out of coniferous forests in the mountains to obtain, through the practice of controlled fires, new highland pastures17. During the Eneolithic and Bronze Age, when the climate deteriorated, the practice of transhumance towards highland pastures was consolidated145.

This research presents the first global overview of human and animal isotopic data used to reconstruct dietary patterns for Bronze Age Italy. As some areas are still isotopically unexplored, the present scenario may change with new studies. However, based on the present data, some trends have been highlighted.

Results suggest that the MBA was a period of major change and that the north of Italy, particularly the northeast, seemed to have been at the forefront of this dynamic, where new trends were introduced that spread into other areas of the peninsula. Changes in dietary patterns are clearly detected, resulting in either global or more regional patterns. Indeed, we highlighted (I) an increased consumption of broomcorn and foxtail millet, mainly for humans from the EBA/MBA transition, (II) a greater legume consumption, particularly for the populations of the center and south of Italy from the MBA onward and (III) a likely increase in meat and/or dairy product intake at the end of the Bronze Age. Several factors could have been responsible for these changes, including the climatic fluctuations that resulted in more arid soils, alongside the strong impact of human activities. Indeed, anthropic activities like deforestation, the over-exploitation of soils for agriculture and livestock breeding and the increase of metallurgic production could have reduced soil fertility. Consequently, the need for a more diverse diet arose, leading to the adoption of more tolerant crops and further resources to exploit. In addition, the increase of trade and the related economical activities with other Mediterranean and continental European areas points to exchanges at different levels, not only in terms of prestigious goods, i.e. ivory, amber and silk, but also in terms of local agricultural practices, ways of life, traditions and, consequently, socio-economic systems.

This study is a first step towards a more in-depth understanding of the Italian dietary patterns during the Bronze Age, which is still a largely unexplored period in terms of biomolecular investigations, and it represents also a new piece of the gigsaw of the global understanding of Bronze Age societal developments. Further isotopic data on botanical, animal and human remains are necessary to improve the resolution of our analyses and to gain a more fine-grained depiction of subsistence strategies during a time of rapid socio-economic change.

Protocols proposed by Longin et al.146 and modified by Bocherens147 were followed for bone collagen extraction for Pertuso, Buco del Diavolo, Ostiglia La Vallona, Gr. Vittorio Vecchi, Trinitapoli and Castiglione. The collagen extraction was performed at the LAMPEA Biochemistry Unit (Aix-en-Provence, France) and the isotopic ratios were measured at Iso-Analytical Ltd (Europa Scientific 20–20, Crewe, UK). The isotope ratios are reported as delta that has been defined according to IUPAC (International Union of Pure and Applied Chemistry):

where Rs and Rst are the isotope ratios 15N/14N and 13C/12C of the isotopic abundances of 15N, 14N, 13C, and 12C, and ‰ = 10–3. The analytical precision of both samples and standards is 0.1‰ for both δ15N ​​and δ13C.

As for the published data, the samples from Ballabio, Felcetone, Gr. Misa, Gr. Scoglietto were analyzed following the above cited protocol as part of the same research26,61. The sites of Arano, Olmo di Nogara, Mereto, Sedegliano, Gradisca di Codroipo, Lavello, Toppo Daguzzo, Dossetto di Nogara, Bovolone, Fondo Paviani, Gr. Regina Margherita and Punta di Zambrone followed a similar collagen extraction protocol applied to a fragment of cortical bone25,59,60,62,64. The samples from Gr. della Monaca, Gr. dell’Antenato, Toppo Daguzzo, Trinitapoli and Murgia Timone, analyzed by Arena et al.63 followed a similar protocol but they were also ultrafiltered.

0.5 g of powdered bone was soaked in HCl (1 M, 20’) for demineralization and in NaOH (0.125 M, 18 h) to eliminate contaminants and finally dissolved in HCl (pH = 2, 17 h, 100 °C). The residue was then frozen (> 5 h, − 60°) and freeze dried for 48 h. Isotopic ratios and elemental contents were provided by Elemental Analyzer coupled to Isotope Ratio Mass Spectrometry at Iso-Analytical Ltd (Europa Scientific 20–20, Crewe, UK).

Isotopic accuracy was monitored through routine analyses calibrated against international standards: bovine liver (IA-R042), a mixture of ammonium sulfate (IA-R045) and beet sugar (IA-R005), and a mixture of sugar cane (IA-R006) and ammonium sulfate (IA-R046). The analytical precision of both samples and standards is 0.1‰ for both C and N. Several preservation criteria have been considered to check for collagen integrity: collagen yields greater than 1%, %C ≥ 30% and %N ≥ 10%, and C/N ratios ranging between 2.9 and 3.6146,147,150.

Statistical analysis used R 3.6.1151 and the coin package v. 1.3–1152. Considering the small sample size, the non-parametric tests of Kruskal–Wallis (Monte Carlo approximation, 10,000 replicates) and exact Wilcoxon Mann–Whitney are used to assess for significant differences in δ13C and δ15N. In the case of multiple tests, in order to identify as many significant comparisons as possible while still maintaining a low false positive rate, the False Discovery Rate (FDR) is applied on the results obtained by the Wilcoxon Mann–Whitney tests153. Only corrected p-values are presented. A 0.05 probability (p < 0.05) is considered significant.

All data needed to evaluate the conclusions in the paper are present in the paper and in the Supplementary Information 1 and 2.

Heyd, V. Europe 2500 to 2200 BC: between expiring ideologies and emerging complexity. In The Oxford Handbook of the European Bronze Age (eds Fokkens, H. & Harding, A.) (Oxford University Press, 2013).

Google Scholar

Nisbet, R. I cereali, le leguminose e i problemi della loro conservazione nell’Italia nord-occidentale dalla preistoria al XIII secolo. in Il seme, l’aratro la messe (eds. Comba, R. & Panero, F.) 13–22 (Soc. Studi Stor. Archeologici, Rocca dè Baldi, 1996).

Nisbet, R. Agricoltura ed economia vegetale. In I liguri Un antico popolo europeo tra Alpi e Mediterraneo (eds De Marinis, R. C. & Spadea, G.) 65–67 (Skira, 2004).

Google Scholar

Knipper, C. et al. A distinct section of the early bronze age society? Stable isotope investigations of burials in settlement pits and multiple inhumations of the Únětice culture in central germany. Am. J. Phys. Anthropol. https://doi.org/10.1002/ajpa.22892 (2015).

Article PubMed Google Scholar

Hamilakis, Y. Wine, oil and the dialectics of power in Bronze Age Crete: A review of the evidence. Oxf. J. Archaeol. 15, 1–32 (1996).

Article Google Scholar

Fiorentino, G., Castiglioni, E., Rottoli, M. & Nisbet, R. Le colture agricole in Italia nel corso dell’età del Bronzo: sintesi dei dati e linee di tendenza. In L’età del Bronzo Recente in Italia (ed. Cocchi Genick, D.) 219–226 (Baroni, 2004).

Google Scholar

Primavera, M., D’Oronzo, C., Muntoni, I. M., Radina, F. & Fiorentino, G. Environment, crops and harvesting strategies during the II millennium BC: Resilience and adaptation in socio-economic systems of Bronze Age communities in Apulia (SE Italy). Quatern. Int. 436, 83–95 (2017).

Article Google Scholar

Nisbet, R. & Rottoli, M. Le analisi dei macroresti vegetali dei siti dell’eta‘ del bronzo, in Le Terramare. La più antica civiltà padana. (eds. Bernabò Brea, M., Cardarelli, A. & Cremaschi, M.) 469–474 (1997).

Carra, M. L. Per una storia della cerealicoltura in Italia settentrionale dal Neolitico all’Età del Ferro: strategie adattive e condizionamenti ambientali (Università degli Studi di Bologna, 2012).

Google Scholar

Miller, N. F., Spengler, R. N. & Frachetti, M. Millet cultivation across Eurasia: Origins, spread, and the influence of seasonal climate. Holocene 26, 1566–1575 (2016).

Article ADS Google Scholar

Mercuri, A. et al. Pollen and macroremains from Holocene archaeological sites: A dataset for the understanding of the bio-cultural diversity of the Italian landscape. Rev. Palaeobot. Palynol. 218, 250–266 (2015).

Article Google Scholar

Cremaschi, M., Pizzi, C. & Valsecchi, V. Water management and land use in the terramare and a possible climatic co-factor in their abandonment: The case study of the terramara of Poviglio Santa Rosa (northern Italy). Quatern. Int. 151, 87–98 (2006).

Article Google Scholar

Cremaschi, M. Ambiente, clima ed uso del suolo nella crisi della cultura delle Terramare. in Le Ragioni del Cambiamento - Reasons for change. Nascita, declino, crollo delle società fra IV e I millennio a.C. (eds. Cazzella, A., Cardarelli, A., Frangipane, M. & Peroni, R.) 521–534 (Scienze dell’Antichità 15, 2009).

Cremaschi, M. et al. Climate change versus land management in the Po Plain (Northern Italy) during the Bronze Age: New insights from the VP/VG sequence of the Terramara Santa Rosa di Poviglio. Quatern. Sci. Rev. 136, 153–172 (2016).

Article ADS Google Scholar

Perego, R. Contribution to the development of the Bronze Age plant economy in the surrounding of the Alps : an archaeobotanical case study of two Early and Middle Bronze Age sites in northern Italy (Lake Garda region) - edoc. (University of Basel, 2017).

Bellini, C., Mariotti-Lippi, M., Mori Secci, M., Aranguren, B. & Perazzi, P. Plant gathering and cultivation in prehistoric Tuscany (Italy). Veg. Hist. Archaeobot. 17, 103–112 (2008).

Article Google Scholar

Arobba, D. & Caramiello, R. Rassegna dei ritrovamenti paleobotanici d’interesse alimentare in Liguria tra Neolitico ed età del Ferro e variazioni d’uso del territorio. In Archeobotanica e alimentazione (eds Mercuri, A. M. et al.) 229–247 (Società dei Naturalisti e Matematici di Modena, 2007).

Google Scholar

Castiglioni, E., Motella De Carlo, S. & Nisbet, R. Indagini sui resti vegetali macroscopici Canàr. Padusa, Quad. 2, 115–130 (1998).

Google Scholar

Rottoli, M. Castellaro del Vho’. Campagne di scavo 1996–1999. Scavi delle Civiche Raccolte Archeologiche di Milano. in Analisi archeobotaniche: i macroresti vegetali (ed. Frontini, P.) 175–195 (2001).

Rottoli, M. Italia settentrionale. in Storia dell’agricoltura italiana, L’età Antica, 1° Preistoria (eds. Forni, G. & Marcone, A.) 235–245 (Edizioni Polistampa, 2002).

Filipović, D. et al. New AMS 14C dates track the arrival and spread of broomcorn millet cultivation and agricultural change in prehistoric Europe. Sci. Rep. 10, 13698 (2020).

Article ADS PubMed PubMed Central CAS Google Scholar

Iacumin, P., Galli, E., Cavalli, F. & Cecere, L. C4-consumers in southern Europe: The case of Friuli V.G. (NE-Italy) during Early and Central Middle Ages. Am. J. Phys. Anthropol. 154(4), 561–574. https://doi.org/10.1002/ajpa.22553 (2014)

Article CAS PubMed Google Scholar

Laffranchi, Z., Delgado Huertas, A., Jiménez Brobeil, S. A., Torres, A. G. & Riquelme Cantal, J. A. Stable C & N isotopes in 2100 Year-B.P. human bone collagen indicate rare dietary dominance of C4 plants in NE-Italy. Sci. Rep. 6, 38817 (2016).

Article ADS CAS PubMed PubMed Central Google Scholar

Killgrove, K. & Tykot, R. H. Food for Rome: A stable isotope investigation of diet in the Imperial period (1st–3rd centuries AD). J. Anthropol. Archaeol. 32, 28–38 (2013).

Article Google Scholar

Tafuri, M. A. et al. Estimating C4 plant consumption in Bronze Age Northeastern Italy through stable carbon and nitrogen isotopes in bone collagen. Int. J. Osteoarchaeol. 28, 131–142 (2018).

Article Google Scholar

Varalli, A., Moggi-Cecchi, J., Moroni, A. & Goude, G. Dietary variability during bronze age in Central Italy: First results. Int. J. Osteoarchaeol. 26, 431–446 (2016).

Article Google Scholar

Motella De Carlo, S. & Venturino Gambari, M. Dalle foreste ai campi, ambiente, risorse e economia nel Neolitico dell’Italia nord-occidentale. in Bull. Etudes Prehistoriques et Archaeologiques Alpines 125–142 (2004).

Fiorentino, G. et al. Climate changes and human-environment interactions in the Apulia region of southeastern Italy during the Neolithic period. Holocene 23, 1297–1316 (2013).

Article ADS Google Scholar

Riedel, A. L’età del Bronzo in Italia nei secoli dal XVI al XIV A.C. Rassegna di Archeologia 10, 173–175 (1992).

Google Scholar

Riedel, A. Archaeozoological investigations in North-eastern Italy: The exploitation of animals since the Neolithic. Preistoria alpina 30, 43–94 (1996).

Google Scholar

Battisti, M. & Marconi, S. L. fauna dell’insediamento dei Pizzini di Castellano (TN) e l’allevamento nell’Italia Nord Orientale nel corso dell’antica età del Bronzo. Padusa. Nuova Serie 39, 45–59 (2003).

Google Scholar

De Grossi Mazzorin, J. & Riedel, A. La fauna delle terramare. in Le Terramare. La più antica civiltà padana (eds. Bernabò Brea, M., Cardarelli, A. & Cremaschi, M.) 87–89 (Electa, 1997).

De Marinis, R. Il Museo Civico Archeologico Giovanni Rambotti. Una introduzione alla preistoria del lago di Garda (Comune di Desenzano del Garda, Garda, 2000).

Google Scholar

Maini, E. & Curci, A. Considerazioni sull’economia di allevamento nella Romagna durante l’età del Bronzo. in Economia e ambiente nell’Italia padana dell’età del bronzo. Le indagini bioarcheologiche (eds. De Grossi Mazzorin, J., Curci, A. & Giacobini, G.) 357–376 (Edilpuglia, 2013).

Riedel, A. & Tecchiati, U. Settelements and economy inthe Bronze and Iron Age in Trentino and South Tyrol. Notes for an archaeozoological model. Preistoria Alpina 35, 105–113 (2001).

Google Scholar

Tecchiati, U. et al. Zooarchaeological evidence of functional and social differentiation in northern Italy between the Neolithic and Bronze ages. Quatern. Int. 539, 105–121 (2020).

Article Google Scholar

Wilkens, B. Archeozoologia Il Mediterraneo, la storia la Sardegna (Editrice Democratica Sarda, 2012).

Google Scholar

Maini, E. Le analisi archeozoologiche nel sito di Cattolica - centro VGS (RN). in Economia e ambiente nell’Italia padana dell’età del bronzo. Le indagini bioarcheologiche (eds. De Grossi Mazzorin, J., Curci, A. & Giacobini, G.) 271–291 (Edilpuglia, 2013).

Riedel, A. Die Fauna der Sonnenburger Ausgrabungen. Preistoria alpina 20, 261–280 (1984).

Google Scholar

Salvagno, L. & Tecchiati, U. I resti faunistici del villaggio dell’età del Bronzo di Sotciastel. Economia e vita di una comunità protostorica alpina (ca. XVII-XIV sec. a.C.). Ladinia Monografica 3, Istitut Ladin «Micurà de Rü», S. Martino in Badia, Bolzano. 3, (2011).

Riedel, A. L’archeozoologia del veronese: cenni sul suo sviluppo e sui risultati delle ricerche dell’ultimo decennio. in Dalla Terra al Museo. Mostra di reperti preistorici e protostorici degli ultimi dieci anni di ricerca dal territorio veronese (eds. Belluzzo, G. & Salzani, L.) 169–178 (Fondazione Fioroni, 1996).

De Grossi Mazzorin, J. Fondo Paviani e Frattesina: economia animale di due central places della tarda Età del Bronzo veneta. Preistoria e Protostoria del Veneto, Studi di Preistoria e Protostoria 2, 389–400 (2015).

Google Scholar

De Grossi Mazzorin, J., Riedel, A. & Tagliacozzo, A. L’evoluzione delle popolazioni animali e dell’economia nell’età del Bronzo Recente. in L’età del Bronzo Recente in Italia. (ed. Cocchi Genick, D.) 227–232 (Baroni, 2004).

Corridi, C. La Fauna. in Un insediamento dell’età del Bronzo a Sesto Fiorentino (ed. Sarti, L.) 35–40 (Garlatti e Razzai, 1994).

Maini, E. Lo sviluppo dell’allevamento in Emilia-Romagna. Aspetti economici e implicazioni sociali nella gestione della risorsa animale durante l’età del Bronzo. (Università degli Studi di Bologna, 2012).

Cosentino, S. & De Grossi Mazzorin, J. Archeozoologia e paleobotanica dell’insediamento protostorico del Borgo di Vejano (VT). in Atti Preistoria e Protostoria in Etruria IV (Manciano-Montalto di Castro-Valentano, 12–14 settembre 1997) 447–460. (Centro Studi di Preistoria e Archeologia, 2000).

De Grossi Mazzorin, J. Economie di Allevamento in Italia centrale dalla media Età del Bronzo alla fine dell’Età del Ferro. in Settlement and Economy in Italy 150 BC to AD 1500. (ed. Christie, N.) 167–177 (Oxbow Books, 1995).

Wilkens, B. La fauna dell’età del Bronzo nelle Marche. in Atti del II Convegno Nazionale di Archeozoologia (Asti, 14–16 novembre 1997) 251–255 (Abaco, 2000).

Tagliacozzo, A. I dati archeozoologici: economia di allevamento e caccia a Broglio di Trebisacce. in Enotri e Micenei nella Sibaritide (eds. Peroni, R. & Trucco, F.) 587–652 (Istituto per la storia e l’archeologia della Magna Grecia, 1994).

Tagliacozzo, A. & Curci, A. I dati archeozoologici: allevamento e caccia nell’età del Bronzo. in Torre Mordillo 1987–1990. Le relazioni Egee di una comunità protostorica della Sibaritide (eds. Trucco, F. & Vagnetti, L.) 347–418 (2001).

Siracusano, G. La fauna degli scavi in estensione. Le scelte nella dieta carnea degli abitanti del villaggio dell’età del Bronzo di Coppa Nevigata (FG). in Coppa Nevigata e l’area umida alla foce del Candelaro durante l’età del Bronzo (eds. Cazzella, A., Moscoloni, M. & Recchia, G.) 235–246 (Grenzi, 2012).

De Grossi Mazzorin, J. L’utilizzazione degli animali nella documentazione archeozoologica dell’Età del Bronzo in Puglia. in Ambra per Agamennone: Indigeni e Micenei tra Adriatico Ionio ed Egeo. (eds. Radina, F. & Recchia, G.) 69–72 (Adda, 2010).

Wilkens, B. I resti faunistici di alcuni insediamenti dell’età del Bronzo nell’Italia centro-meridionale. Rassegna di Archeol. 10, 463–469 (1992).

Google Scholar

Wilkens, B. Le risorse animali. in Documenti dell’età del Bronzo. Ricerche lungo il versante adriatico pugliese (eds. Cinquepalmi, A. & Radina, F.) 223–247 (Schena editore, 1998).

Wilkens, B. I resti faunistici dell’abitato dell’età del Bronzo di Bari (S.Maria del Buon Consiglio). Taras. Riv. Archeol. 11, 85–93 (1991).

Google Scholar

Wilkens, B. Gli animali nell’economia degli insediamenti pugliesi dell’età del Bronzo. Taras 15, 491–499 (1995).

Google Scholar

De Grossi Mazzorin, J. lo Sfruttamento delle risrose ittiche in alcuni insediamenti dell’età del Bronzo. in Atti Preistoria e Protostoria in Etruria V 257–267 (Centro Studi di Preistoria e Archeologia, 2002).

De Grossi Mazzorin, J. & Frezza, A. M. Lo sfruttamento delle risorse fluviali di due insediamenti veneti dell’Età del Bronzo: Canàr e Frattesina. Atti del II Convegno Nazionale di Archeozoologia (Asti, 14–16 Novembre 1997). 241–250 (2000).

Tafuri, M. A., Craig, O. E. & Canci, A. Stable isotope evidence for the consumption of millet and other plants in Bronze Age Italy. Am. J. Phys. Anthropol. 139, 146–153 (2009).

Article PubMed Google Scholar

Varalli, et al. Dietary continuity vs discontinuity in Bronze age Italy. The isotopic evidence from Arano di Cellore (Illasi, Verona, Italy). J. Archaeol. Sci. Rep. 7, 104–113 (2016).

Google Scholar

Masotti, S., Varalli, A., Goude, G., Moggi-Cecchi, J. & Gualdi-Russo, E. A combined analysis of dietary habits in the Bronze Age site of Ballabio (northern Italy). Archaeol. Anthropol. Sci. 11, 1029–1047 (2019).

Article Google Scholar

Rumolo, A., Forstenpointner, G., Rumolo, P. & Jung, R. Palaeodiet reconstruction inferred by stable isotopes analysis of faunal and human remains at Bronze Age Punta di Zambrone (Calabria, Italy). Int. J. Osteoarchaeol. 30, 90–98 (2020).

Article Google Scholar

Arena, F., Gualdi-Russo, E., Olsen, J., Philippsen, B. & Mannino, M. A. New data on agro-pastoral diets in southern Italy from the Neolithic to the Bronze Age. Archaeol. Anthropol. Sci. 12, 245 (2020).

Article Google Scholar

Skeates, R. et al. Rethinking Collective Burial in Mediterranean Caves: Middle Bronze Age Grotta Regina Margherita Central Italy. J. Field Archaeol. 2, 1–17 (2021).

Google Scholar

Varalli, A., Goude, G. & Dal Del Lucchese, A. Neolitico all’Età del Ferro: indagine paleonutrizionale su alcune popolazioni della Liguria attraverso un approccio isotopico. Archeol. Liguria 5, 11–19 (2015).

Google Scholar

DeNiro, M. J. & Epstein, S. Influence of diet on the distribution of carbon isotopes in animals. Geochim. Cosmochim. Acta 42, 495–506 (1978).

Article ADS CAS Google Scholar

Bocherens, H. & Drucker, D. Trophic level isotopic enrichment of carbon and nitrogen in bone collagen: Case studies from recent and ancient terrestrial ecosystems. Int. J. Osteoarchaeol. 13, 46–53 (2003).

Article Google Scholar

Vika, E. & Theodoropoulou, T. Re-investigating fish consumption in Greek antiquity: results from δ13C and δ15N analysis from fish bone collagen. J. Archaeol. Sci. 39, 1618–1627 (2012).

Article Google Scholar

Craig, O. E., Bondioli, L., Fattore, L., Higham, T. & Hedges, R. Evaluating marine diets through radiocarbon dating and stable isotope analysis of victims of the AD79 eruption of vesuvius. Am. J. Phys. Anthropol. 152, 345–352 (2013).

Article PubMed Google Scholar

Silvestri, S. Nuovi reperti dell’Eneolitico-Bronzo da grotte sepolcrali dell’area ligure-provenzale Antropologia e comportamento funerario (Università di Pisa, Pisa, 2010).

Google Scholar

Le Bras-Goude, G. et al. Stratégies de subsistance et analyse culturelle de populations néolithiques de Ligurie approche par l’étude isotopique δ13C et δ15N des restes osseux. Bull. Mémoires Soc. d’Anthropologie de Paris 18, 43–53 (2006).

Article Google Scholar

Goude, G., Dori, I., Sparacello, V. S., Starnini, E. & Varalli, A. Multi-proxy dentine microsections analyses reveal diachronic changes in life history adaptations, mobility, and tuberculosis-induced wasting in prehistoric Liguria (Finale Ligure, Italy, northwestern Mediterranean). Int. J. Paleopathol. 28, 99–111 (2020).

Article CAS PubMed Google Scholar

Dal diaspro al bronzo. L’età del Rame e l’età del Bronzo in Liguria. (Luna Editore, 1998).

Maggi, R. et al. Note sugli scavi del Castellaro di Uscio (GE), 1981–1985. Prehistoria Alpina 21, 59–83 (1985).

Google Scholar

Nisbet, R. & Scaife, R. Analisi archeobotaniche: storia forestale ed uso delle piante. Bullettino di Paletnologia Italiana 89, 273–280 (1998).

Lightfoot, E., Šlaus, M., Šikanjić, P. R. & O’Connell, T. C. Metals and millets: Bronze and Iron Age diet in inland and coastal Croatia seen through stable isotope analysis. Archaeol. Anthropol. Sci. 7, 375–386 (2015).

Article Google Scholar

Paltineri, S. & Rubat Borel, F. Le armille tipo Zerba: un riesame della questione. Preisotia e Protostoria in Etruria 9, 609–621 (2010).

Google Scholar

Mayewski, P. A. et al. Holocene climate variability. Quatern. Res. 62, 243–255 (2004).

Article ADS Google Scholar

Branch, N. P. & Marini, N. A. F. Mid-Late Holocene environmental change and human activities in the northern Apennines Italy. Quat. Int. 353, 34–51 (2014).

Article Google Scholar

Cloern, J., Canuel, E. & Harris, D. Stable carbon and nitrogen isotope composition of aquatic and terrestrial plants of the San Francisco Bay estuarine system. Limnol. Oceanogr. 47, 713–724 (2002).

Article ADS CAS Google Scholar

Schulting, R. J., Blockley, S. M., Bocherens, H., Drucker, D. & Richards, M. Stable carbon and nitrogen isotope analysis on human remains from the Early Mesolithic site of La Vergne (Charente-Maritime, France). J. Archaeol. Sci. 35, 763–772 (2008).

Article Google Scholar

Accorsi, C. A., Bandini Mazzanti, M., Mercuri, A. M., Rivalenti, C. & Torri, P. Analisi pollinica di saggio per l’insediamento palafitticolo di Canàr-Rovigo, 680–700 m s.l.m. (Antica Età del Bronzo). in Canàr di San Pietro Polesine. Ricerche archeo-ambientali sul sito palafitticolo (eds. Balista, C. & Bellintani, P.) 131–149 (1998).

Marchesini, M., Biagioni, S. & Marvelli, S. Indagine palinologia della serie stratigrafica di Ca’Emo (Adria, Rovigo) dall’età del Bronzo al XIX sec. d.C. La variabilità del clima nel Quaternario - La ricerca italiana. Atti del Convegno AIQUA 110–111 (2009).

Valsecchi, V., Tinner, W., Finsinger, W. & Ammann, B. Human impact during the Bronze Age on the vegetation at Lago Lucone (northern Italy). Veg. Hist. Archaeobotany 15, 99–113 (2006).

Article Google Scholar

Costantini, L. & Costantini Biasini, L. Economia agricola del Lazio a sud del Tevere tra Bronzo antico e Bronzo medio. in Atti della XL riunione scientifica Istituto Italiano di Preistoria e Protostoria 787–801 (Istituto Italiano di Preistoria e Protostoria, 2007).

Cortese, F. et al. Isotopic reconstruction of the subsistence strategy for a Central Italian Bronze Age community (Pastena cave, 2nd millennium BCE). bioRxiv 2022.04.21.488758 (2022). https://doi.org/10.1101/2022.04.21.488758.

Tunzi Sisto, A. M. Ipogei della Daunia. Culti e riti funerari della media età del Bronzo. (Grenzi, 1997).

Brasili, P. & Castrì, L. Aspetti nutrizionali nell’età del Bronzo e del ferro in un’antica comunità siciliana (Castiglione, Ragusa). in Science and technology for the safeguard of cultural heritage in the mediterranean basin. Atti 1st International Congress - 1491–1500 (1998).

Pellegrini, E. L’età dei metalli nell’Italia meridionale e in Sicilia. in Italia Preistorica 471–511 (Laterza, 1993).

Tusa, S. La Sicilia nella Preistoria. (Sellerio Editore, Nuovo Prisma, 1999).

Facchini, F. I reperti scheletrici della necropoli di Catiglione (Ragusa) (Età del Bronzo). Archivio per l’antropologia e l’etnologia 105, 80–153 (1975).

Google Scholar

Cavazzuti, C. & Arena, A. The bioarchaeology of social stratification in bronze age Italy. Arheo 37, 69–105 (2020).

Google Scholar

Fraser, R. A. et al. Assessing natural variation and the effects of charring, burial and pre-treatment on the stable carbon and nitrogen isotope values of archaeobotanical cereals and pulses. J. Archaeol. Sci. 40, 4754–4766 (2013).

Article CAS Google Scholar

Mariotti Lippi, M., Pisaneschi, L., Sarti, L., Lari, M. & Moggi-Cecchi, J. Insights into the Copper-Bronze Age diet in Central Italy: Plant microremains in dental calculus from Grotta dello Scoglietto (Southern Tuscany, Italy). J. Archaeol. Sci. Rep. 15, 30–39 (2017).

Google Scholar

Di Rita, F. & Magri, D. The 4.2 ka event in the vegetation record of the central Mediterranean. Clim. Past 15, 237–251 (2019).

Article Google Scholar

Arrighi, S. & Moroni Lanfredini, A. Sansepolcro (AR). Gorgo del Ciliegio. Notiziario della Soprintendenza per i Beni Archeologici della Toscana 3, 542–544 (2007).

Google Scholar

Jung, R., Pacciarelli, M., Zach, B., Klee, M. & Thanheiser, U. Punta di Zambrone (Calabria)—a Bronze Age Harbour Site. First Preliminary Report on the Recent Bronze Age (2011–2012 Campaigns). Archaeol. Austr. 1, 53–110 (2015).

Google Scholar

Bertolini, M., Zanini, S. & Thun Hohenstein, U. Nuovi dati sullo sfruttamento delle risorse animali tra il Bronzo Medio e Recente nei territori del Basso Veronese e Medio Polesine. Preistoria e Protostoria del Veneto 2, 321–326 (2015).

Google Scholar

Goude, G., Binder, D. & Del Lucchese, A. Alimentation et modes de vie néolithiques en Ligurie. Rivista Stud. Liguri 77, 371–383 (2014).

Google Scholar

Lelli, R. et al. Examining dietary variability of the earliest farmers of south-eastern Italy. Am. J. Phys. Anthropol. 149, 380–390 (2012).

Article PubMed Google Scholar

Tafuri, M. A. et al. Herding practices in the Ditched Villages of the Neolithic Tavoliere (Apulia, south-east Italy). Proc. Br. Acad. 198, 143–157 (2014).

Google Scholar

Mannino, A., Talamo, S., Goude, G. & Richards, M. Analisi isotopiche e datazioni sul collagene osseo degli inumati dell’Arma dell’Aquila (Finale Ligure, Savona). in Gli scavi dell’Arma dell’Aquila (Finale Ligure, Savona): le ricerche e i materiali degli scavi del Novecento (eds. Biagi, P. & Starnini, E.) 183–189 (Società per la Peristoria e Protostoria della regione Friuli-Venezia Giulia, 2018).

Scorrano, G. et al. Effect of Neolithic transition on an Italian community: Mora Cavorso (Jenne, Rome). Archaeol. Anthropol. Sci. 2, 1–17. https://doi.org/10.1007/s12520-018-0615-9 (2018).

Article Google Scholar

Giorgi, F., Bartoli, F., Iacumin, P. & Mallagni, F. Oligoelements and isotopic geochemistry: A multidisciplinary approach to the reconstruction of the palaeodiet. Hum. Evol. 20, 55–82 (2005).

Article Google Scholar

Parkinson, E. W. & McLaughlin, T. R. Lifeways at the acme of the south Italian Neolithic: New chronological and bioarchaeological data from Fonteviva, Apulia. J. Archaeol. Sci. Rep. 34, 102589 (2020).

Google Scholar

De Angelis, F. et al. Exploring mobility in Italian Neolithic and Copper Age communities. Sci. Rep. 11, 2697 (2021).

Article ADS PubMed PubMed Central CAS Google Scholar

Bernardini, S. et al. Social dynamics and resource management strategies in copper age Italy: Insights from archaeological and Isotopic Data. Environ. Archaeol. 2, 1–23 (2021).

Article Google Scholar

De Angelis, F. et al. Eneolithic subsistence economy in Central Italy: First dietary reconstructions through stable isotopes. Archaeol. Anthropol. Sci. 2, 1–16 (2019).

Google Scholar

Formicola, V. et al. An unusual surgical treatment of the skull following trauma during the Copper Age (IV millennium B.C.) in Italy. J. Anthropol. Sci. 90, 1–2 (2012).

Google Scholar

Romboni, M. et al. Isotopic evidence for population dynamics in the Central Italian copper age and bronze age. BioRxiv https://doi.org/10.1101/2021.09.30.462554 (2021).

Article Google Scholar

Cardarelli, A. The collapse of the Terramare culture and growth of new economic and social systems during the Late Bronze Age in Italy Scienze dell’Antichità. Storia Archeol. Antropol. 15, 449–520 (2009).

Google Scholar

Dal Corso, M. et al. Exploiting the feasibility of biomarker analysis in alluvial contexts: The first miliacin evidence from the northern Italian Bronze Age. In Millet and what else? The wider context of the adoption of millet cultivation in Europe (eds Kirleis, W. et al.) 231–256 (SIdestone Press, 2022).

Google Scholar

Petroutsa, E. I., Richards, M. P. & Manolis, S. K. Stable isotope analysis of human remains from the early Helladic site of Perachora, Korinth, Greece. In Cooking up the past : food and culinary practices in the Neolithic and Bronze Age Aegean (eds Mee, C. & Renard, J.) 290–296 (Oxbow Books, 2007).

Google Scholar

Triantaphyllou, S. A bioarchaeological approach to prehistoric cemetery populations from central and western Greek Macedonia. in British Archaeological Reports, International Series S976 (Archaeopress, 2001).

Triantaphyllou, S. A stable isotope analysis of skeletal assemblages from Greek Macedonia. In Were they what they ate? Dietary reconstruction in Greece from stable isotope analyses (eds Papathanasiou, A. et al.) 113–153 (Occasional Wiener Laboratory Series, 2014).

Google Scholar

Petroutsa, E. I. & Manolis, S. K. Reconstructing Late bronze age diet in mainland Greece using stable isotope analysis. J. Archaeol. Sci. 37, 614–620 (2010).

Article Google Scholar

Dotsika, E. et al. Establishment of a Greek food database for palaeodiet reconstruction: Case study of human and fauna remains from neolithic to late bronze age from Greece. Geosciences 9, 165 (2019).

Article ADS CAS Google Scholar

Varalli, A. et al. Bronze age innovations and impact on human diet: A multi-isotopic and multi-proxy study of western Switzerland. PLoS ONE 16, e0245726 (2021).

Article CAS PubMed PubMed Central Google Scholar

Goude, G., Rey, L., Toulemonde, F., Cervel, M. & Rottier, S. Dietary changes and millet consumption in northern France at the end of Prehistory: Evidence from archaeobotanical and stable isotope data. Environ. Archaeol. 22, 268–282 (2017).

Article Google Scholar

Lightfoot, E., Šlaus, M. & O’Connell, T. C. Changing cultures, changing cuisines: Cultural transitions and dietary change in iron age, roman, and early medieval Croatia. Am. J. Phys. Anthropol. 148, 543–556 (2012).

Article CAS PubMed Google Scholar

Le Huray, J. D. & Schutkowski, H. Diet and social status during the La Tène period in Bohemia: Carbon and nitrogen stable isotope analysis of bone collagen from Kutná Hora-Karlov and Radovesice. J. Anthropol. Archaeol. 24, 135–147 (2005).

Article Google Scholar

Murray, M. L. & Schoeninger, M. J. Diet, status, and complex social structure in Iron Age Central Europe: some contributions from bone chemistry. In Tribe and polity in late prehistoric Europe: Demography, production and exchange in the evolution of complex social systems (eds Gibson, D. B. I. & Geselowitz, M. N.) 155–176 (Plenum Press, 1988).

Chapter Google Scholar

Knipper, C. et al. Social differentiation and land use at an Early Iron Age "princely seat": Bioarchaeological investigations at the Glauberg (Germany). J. Archaeol. Sci. 41, 818–835 (2014).

Article Google Scholar

Papathanasiou, A., Panagiotopoulou, E., Beltsios, K., Papakonstantinou, M.-F. & Sipsi, M. Inferences from the human skeletal material of the early iron age cemetery at agios dimitrios, Fthiotis, Central Greece. J. Archaeol. Sci. 40, 2924–2933 (2013).

Article CAS Google Scholar

Moghaddam, N., Müller, F. & Lösch, S. A bioarchaeological approach to the Iron Age in Switzerland: Stable isotope analyses (δ13C, δ15N, δ34S) of human remains. Archaeol. Anthropol. Sci. 10, 1067–1085 (2018).

Article Google Scholar

López-Costas, O., Müldner, G. & Martínez Cortizas, A. Diet and lifestyle in bronze age northwest Spain: The collective burial of Cova do Santo. J. Archaeol. Sci. 55, 209–218 (2015).

Article Google Scholar

Goude, G., Balasescu, A., Réveillas, H., Thomas, Y. & Lefranc, P. Diet variability and stable isotope analyses: Looking for variables within the late neolithic and iron age human groups from gougenheim site and surrounding areas (Alsace, France). Int. J. Osteoarchaeol. 25, 988–996 (2015).

Article Google Scholar

Herrscher, E. et al. Dietary inferences through stable isotope analysis at the Neolithic and Bronze Age in the southern Caucasus (sixth to first millenium BC, Azerbaijan): From environmental adaptation to social impacts. Am. J. Phys. Anthropol. 167, 856–875 (2018).

Article PubMed Google Scholar

Martin, L. et al. The place of millet in food globalization during Late Prehistory as evidenced by new bioarchaeological data from the Caucasus. Sci. Rep. 11, 13124 (2021).

Article ADS CAS PubMed PubMed Central Google Scholar

Wang, T. et al. Tianshanbeilu and the Isotopic Millet Road: Reviewing the late Neolithic/Bronze Age radiation of human millet consumption from north China to Europe. Natl. Sci. Rev. 6, 1024–1039 (2019).

Article CAS PubMed Google Scholar

Allentoft, M. E., Sikora, M. & Sjögren, K. G. Population genomics of Bronze Age Eurasia. Nature 522, 167–172 (2015).

Article ADS CAS PubMed Google Scholar

Cavazzuti, C. et al. Flows of people in villages and large centres in Bronze Age Italy through strontium and oxygen isotopes. PLoS ONE 14, 1–43 (2019).

Article Google Scholar

Iacono, F. et al. Establishing the middle sea: The late Bronze Age of Mediterranean Europe (1700–900 BC). J. Archaeol. Res. https://doi.org/10.1007/s10814-021-09165-1 (2021).

Article Google Scholar

Kristiansen, K. & Larsson, T. B. The rise of Bronze Age society: travels, transmissions and transformations (Cambridge University Press, 2005).

Google Scholar

Dal Corso, M. et al. Bronze Age crop processing evidence in the phytolith assemblages from the ditch and fen around Fondo Paviani, northern Italy. Veg. Hist. Archaeobotany 26, 5–24 (2017).

Article Google Scholar

Baroni, C., Zanchetta, G., Fallick, A. E. & Longinelli, A. Mollusca stable isotope record of a core from Lake Frassino, northern Italy: Hydrological and climatic changes during the last 14 ka. Holocene 16, 827–837 (2006).

Article ADS Google Scholar

Mercuri, A. et al. A marine/terrestrial integration for mid-late Holocene vegetation history and the development of the cultural landscape in the Po valley as a result of human impact and climate change. Veg. Hist. Archaeobotany 21, 353–372 (2012).

Article Google Scholar

Rottoli, M. & Castiglioni, E. Prehistory of plant growing and collecting in northern Italy, based on seed remains from the early Neolithic to the Chalcolithic (c 5600–2100 cal bc). Veg. Hist. Archaeobotany 18, 91–103 (2009).

Article Google Scholar

Costantini, L. & Costantini Biasini, L. Economia agricola del Lazio a sud del Tevere tra Bronzo antico e Bronzo medio. In Atti della XL riunione scientifica Istituto Italiano di Preistoria e Protostoria 787–801 (Istituto Italiano di Preistoria e Protostoria, 2007).

Google Scholar

Ravazzi, C., Cremaschi, M. & Forlani, L. Studio archeopalinologico della terramara S Rosa Nuovi dati, analisi floristica e sintassonomica della vegetazione nell’età del Bronzo. In Il villaggio piccolo della Terramara di S Rosa di Poviglio Scavi 1987–1992 (eds Bernabò Brea, M. & Cremaschi, M.) 703–736 (Origines, 2004).

Google Scholar

Mercuri, A. M. et al. Cereal Fields from the Middle-Recent Bronze Age as found in the terramara di Montale, in the Po plain (Emilia Romagna, Northern Italy), based on pollen, seeds/fruits and microcharcoals. In The Archaeology of Crop fields and garden Archaeology (eds Morel, J.-P. et al.) 251–270 (Edilpuglia, 2006).

Google Scholar

Cremaschi, M. Acque, campi e boschi nella civiltà delle Terramare. Le ragioni di un collasso di civiltà nella Pianura Padana dell’età del bronzo. In La gestione dell’acqua per l’agricoltura nella Pianura Lombarda dal passato al futuro (ed. Cavallin, A.) 5–21 (Istituto Lombardo di Scienze e Lettere, 2017).

Google Scholar

Trentacoste, A., Nieto-Espinet, A. & Valenzuela-Lamas, S. Pre-Roman improvements to agricultural production: Evidence from livestock husbandry in late prehistoric Italy. PLoS ONE 13, e0208109 (2018).

Article PubMed PubMed Central Google Scholar

Varalli, A., Villotte, S., Dori, I. & Sparacello, V. S. New Insights into Activity-Related Functional Bone Adaptations and Alterations in Neolithic Liguria (Northwestern Italy). Bull. Mémoires Soc. d’Anthropologie de Paris 32, 34–58 (2020).

Morandi, L. F., Frémondeau, D., Müldner, G. & Maggi, R. Sequential analyses of bovid tooth enamel and dentine collagen (δ18O, δ13C, δ15N): new insights into animal husbandry between the Late Neolithic and the Early Bronze Age at Tana del Barletta (Ligurian Prealps). Archaeol. Anthropol. Sci. 13, 1–26 (2021).

Article Google Scholar

Longin, R. New method of collagen extraction for radiocarbon dating. Nature 230, 241–242 (1971).

Article ADS CAS PubMed Google Scholar

Bocherens, H. Biogéochimie isotopique (13C, 15N 18O) et paléontologie des vertébrés: Applications à l’étude des réseaux trophiques révolus et des paléoenvironnements (Université Pierre et Marie Curie, 1992).

Google Scholar

Ambrose, S. H. Preparation and characterisation of bone and tooth collagen for isotopic analysis. J. Archaeol. Sci. 17, 431–451 (1990).

Article Google Scholar

DeNiro, M. J. Postmortem preservation and alteration of in vivo bone collagen isotope ratios in relation to paleodietary reconstruction. Nature 317, 806–809 (1985).

Article ADS CAS Google Scholar

van Klinken, G. J. Bone collagen quality indicators for palaeodietary and radiocarbon measurements. J. Archaeol. Sci. 26, 687–695 (1999).

Article Google Scholar

R core team. R: a Language and Environment for Statistical Computing. R Foundation for Statistical Computing (2019).

Hothorn, T., Hornik, K., Van De Wiel, M. A. & Zeileis, A. A lego system for conditional inference. Am. Stat. 60, 257–263 (2006).

Article MathSciNet Google Scholar

Benjamini, Y. & Hochberg, Y. Controlling the false discovery rate: A practical and powerful approach to multiple testing. J. R. Stat. Soc. Ser. B 57, 289–300 (1995).

MathSciNet MATH Google Scholar

Download references

The authors thank M.A. Tafuri (University La Sapienza, Rome) for her essential advice and important discussions, L. Bachechi, G. Stefania (University of Florence), M. Zavattaro (Museum of Natural History, Anthropology and Ethnology Section), G. Belcastro (University of Bologna), F. Mallegni and F. Bartoli (University of Pisa), A. Cassieri (Soprintendenza Archeologica di Formia), E. Bruckner (Antiquarium Museum of Sezze), A. Del Lucchese (Soprintendenza archeologica della Liguria), L. Marchi, G. Ricci and L. Di Francescantonio (Museo Civico di Sanremo) for access to the material and authorizations. The authors are grateful to R. Nisbet, U. Tecchiati, E. Maini, A.M. Mercuri, A. Moroni, F. Sacchetti, S. Cabut, G. André, V. Ollivier, P. Fernandez, M. Bailly, R. Peake for useful discussions and technical support helping to improve this work.

AV benefitted from a scholarship from the Centre of Excellence LabexMed—Human and Social Sciences at the Interdisciplinary Center for the Mediterranean, reference 10-LABX-0090. Further funding was provided by JMC (Finanziamenti di Ateneo, University of Florence), GG and LAMPEA (CNRS, Aix-Marseille University) for the isotopic analysis and the English proofreading service (https://karinagerdau.com).

Aix Marseille Univ, CNRS, Minist Culture, LAMPEA, Aix-en-Provence, France

Alessandra Varalli & Gwenaëlle Goude

Department of Biology, Laboratory of Anthropology, University of Florence, Florence, Italy

Alessandra Varalli & Jacopo Moggi-Cecchi

CaSEs Research Group, Department of Humanities, Universitat Pompeu Fabra, Barcelona, Spain

Alessandra Varalli

You can also search for this author in PubMed Google Scholar

You can also search for this author in PubMed Google Scholar

You can also search for this author in PubMed Google Scholar

A.V.: conceptualization, formal analysis, investigation, writing—original draft, writing—review & editing, visualization, funding acquisition. J.M.C., G.G.: conceptualization, resources, funding acquisition, writing—review & editing.

Correspondence to Alessandra Varalli.

The authors declare no competing interests.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and Permissions

Varalli, A., Moggi-Cecchi, J. & Goude, G. A multi-proxy bioarchaeological approach reveals new trends in Bronze Age diet in Italy. Sci Rep 12, 12203 (2022). https://doi.org/10.1038/s41598-022-15581-0

Download citation

Received: 03 April 2022

Accepted: 27 June 2022

Published: 16 July 2022

DOI: https://doi.org/10.1038/s41598-022-15581-0

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

Archaeological and Anthropological Sciences (2023)

Archaeological and Anthropological Sciences (2022)

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.