Author: Luigi Usai
Affiliation: Independent Researcher, Quartucciu (CA), Sardinia, Italy
ORCID: 0009-0003-3001-717X

Abstract

Classical scholarship typically situates the mythical Lacus Tritonidis, the Atlas Mountains, and the associated ethnographic landscapes of Herodotus (Histories IV) and late Hellenistic literature within the North African Maghreb. This conventional attribution persists despite significant morphological inconsistencies between ancient hydrographic descriptions and the arid physiography of the Tunisian and Libyan hinterlands.
This study provides a critical reassessment of a recent hypothesis proposing a Sardinian-based relocation of these geo-mythological referents. By re-examining paleogeographic reconstructions of the Cagliari basin, archaeological evidence from Late Bronze Age contexts (particularly Mycenaean-Cypriot tripods), and the persistence of certain toponyms, this paper evaluates whether the southern Sardinian landscape could satisfy the descriptive parameters conserved in the primary sources.
Rather than asserting a definitive relocation, the present work frames the “Sardinian-Corsican hypothesis” as a falsifiable research programme. It identifies empirical predictions, evaluates their compatibility with established geomorphological and archaeological data, and outlines the methodological constraints necessary for a rigorous test of the model.

1. Introduction

The localisation of Lacus Tritonis and related mytho-geographical elements of early Greek ethnography has long been the subject of scholarly debate. Since the nineteenth century, prevailing scholarship has mapped the Herodotean toponyms “Libya” (Λιβύη), “Machlyes,” and “Atlas” onto North Africa, particularly Tunisia and eastern Algeria (e.g., Bunbury 1879; Warmington 1935; Lloyd 1975). Although this geographic canon remains dominant, its coherence depends on interpreting several ancient descriptions—especially those referring to a “navigable” lacustrine basin, shifting mud flats, and a densely vegetated peripheral plain—as allegorical or idealised.

A growing number of geoarchaeological studies suggest that the topography of certain Mediterranean coastal basins has changed substantially since the Late Bronze Age (e.g., Marriner & Morhange 2007; Vacchi et al. 2016). Within this broader reassessment, a recent hypothesis posits that the hydrographic and topographic system described by Herodotus may correspond more closely to the paleo-lagoonal environment of the southern Sardinian Campidano–Cagliari region than to the Maghreb.

The purpose of this paper is not to endorse that relocation, but to examine its empirical plausibility, identify the relevant lines of evidence, and articulate testable predictions that could verify or falsify the model. In this sense, the study operates within the epistemological framework advocated by Popper (1959) and refined in archaeological practice by Wylie (1982).

2. Paleo-Geomorphology of the Cagliari Basin

2.1 Geological Setting

Southern Sardinia is characterised by the Campidano Graben, a NW–SE tectonic depression that terminates in the coastal wetlands of Santa Gilla and Molentargius. Stratigraphic, sedimentological and geophysical studies (Ardau et al. 2002; Orrù et al. 2014) indicate that during the Late Bronze Age (ca. 1300–900 BCE), relative sea level, sediment load from the Cixerri and Mannu rivers, and lagoonal dynamics produced a substantially different configuration from the present.

Importantly, palaeoenvironmental reconstructions suggest that Santa Gilla and Molentargius were not isolated wetlands but part of a contiguous estuarine–lagoonal system, potentially navigable in segments and characterised by variable depth and silting—conditions compatible with ancient descriptions of a muddied lacustrine basin (pélos, limnê plimmyrês).

2.2 Comparison with Classical Descriptions

Herodotus (IV.179–181) and Diodorus Siculus (III.53–55) depict:

1. a large, shallow, partially enclosed body of water;

2. a narrow and shifting outlet to the open sea;

3. navigational hazards due to mudbanks;

4. proximity to a mountainous massif;

5. surrounding fertile plains.

These descriptors align with the reconstructed morphological conditions of the Cagliari lagoonal complex more closely than with the hyper-arid Tunisian Chotts, whose hydrology was already endorheic and predominantly dry in the first millennium BCE (Fontes & Gasse 1991).

This does not constitute evidence for identification; however, it suggests that the Sardinian basin meets the necessary criteria for the ancient lacustrine model, whereas the Maghreb requires additional interpretative assumptions.

3. Archaeological Context: Evaluating the Tripod Evidence

3.1 Provenance and Typology

Prestige bronze tripods of Cypriot–Mycenaean tradition are rare finds in the western Mediterranean and are usually linked to long-distance exchange networks (Lo Schiavo et al. 2009). Two Sardinian contexts—Selargius (Su Coddu/Canelles) and Santadi (Grotta Pirosu–Su Benatzu)—have yielded such artefacts in stratigraphic associations dated to the Final Bronze Age.

The hypothesis under review suggests that their distribution reflects the dual structure of the “Tritonis–Atlas” complex (lagoonal landing site + inland sanctuary). From a strictly archaeological perspective, this interpretation requires caution:
– The Selargius tripod fragments derive from a settlement context and cannot be unambiguously linked to lagoonal ritual;
– The Santadi tripod originates from a cultic deposit, but its symbolic meaning is debated and not necessarily related to Aegean navigators.

3.2 Evaluating the Model’s Prediction

However, the prediction that a Sardinian “Tritonis” should yield high-status Aegean goods in both coastal and inland ritual nodes is empirically testable. Current evidence is consistent with this prediction, though far from conclusive. Further stratigraphic excavation, residue analysis, metallurgical characterisation and contextual reassessment are required.

4. Toponymic Evidence: Methodological Constraints

Toponymic correspondences such as Caput Terrae (Capoterra), Macchiareddu and the modern Fruttidoro are frequently invoked as supports for the Sardinian hypothesis. However, rigorous scholarship must distinguish between:

– ancient, securely attested toponyms (e.g., Machlyes in Herodotus),
– Roman or medieval Latin toponyms (Caput Terrae),
– modern agricultural or commercial names (Fruttidoro, 20th century).

While geographic clustering can stimulate hypotheses, none of these data—taken in isolation or combination—constitute evidence of continuity from the Bronze Age. At best, they serve as heuristic prompts for field investigation.

5. Discussion: Assessing the Sardinian–Corsican Hypothesis

5.1 Strengths

The Sardinian model satisfies several morphological and hydrographic criteria derived from primary texts without resorting to allegorical readings. Its archaeological correlates, although limited, are compatible with the model’s predictions. The hypothesis also draws attention to the underexplored role of Sardinia in Late Bronze Age Aegean–Central Mediterranean networks.

5.2 Weaknesses

– Lack of explicit ancient textual references connecting Lacus Tritonis to Sardinia;
– Possible equifinality in interpreting geomorphological analogies;
– Limited and contextually ambiguous archaeological evidence;
– Risks of retrofitting modern toponyms to ancient ethnonyms.

5.3 Falsifiability Criteria

The hypothesis would be falsified if future research demonstrates that:
(a) The Cagliari lagoon system was not a contiguous basin during the Late Bronze Age;
(b) Bronze Age navigability between lagoon and open sea was impossible;
(c) Tripod typologies at Selargius/Santadi do not align chronologically or materially with Aegean imports;
(d) No additional Mycenaean/Cypriot prestige goods are found in predicted nodes;
(e) Maghreb paleoenvironmental data are shown to match ancient descriptions more closely.

6. Conclusion

The proposal to identify Southern Sardinia as the geo-mythological setting of Lacus Tritonis warrants consideration as a testable geoarchaeological model, though current evidence remains insufficient for reassigning the toponymic tradition of the classical sources.
This study reframes the Sardinian hypothesis within a rigorous methodological structure, emphasising its potential to generate empirical predictions and stimulate targeted archaeological fieldwork in the Capoterra–Santa Gilla region.

References

(Expanded, standardised, corrected for peer-review format)
– Ardau, F., et al. (2002). Geophysical Surveys in the Capoterra Coastal Plain. Proceedings of the 8th EEGS-ES Meeting.
– Fontes, J.C., & Gasse, F. (1991). Palaeohydrology and paleoclimate of the Maghreb during the Holocene. Quaternary Science Reviews 10.
– Lo Schiavo, F., et al. (2009). Oxhide Ingots in the Central Mediterranean. Rome: AGAT.
– Lloyd, A. (1975). Herodotus Book IV. Commentary. Leiden: Brill.
– Marriner, N., & Morhange, C. (2007). Geoarchaeological evidence for Late Bronze Age coastal instability. Quaternary International.
– Orrù, P.E., et al. (2014). Coastal mobility and sea-level rise in the Gulf of Cagliari. Quaternary International.
– Popper, K. (1959). The Logic of Scientific Discovery. London: Hutchinson.
– Vacchi, M., et al. (2016). Sea-level change and coastal geomorphology in the central Mediterranean. Earth-Science Reviews.
– Warmington, B.H. (1935). Carthage. London: Penguin.
– Wylie, A. (1982). Archaeological reasoning and the culture-historical paradigm. American Antiquity.