Midden

A midden (from the Middle English midding, derived from the Old Norse mykidyngja, meaning "muck heap" or "dung pile"), also known in coastal contexts as a shell midden or kitchen midden (from the Danish køkkenmødding), is a stratified anthropogenic deposit composed of accumulated domestic refuse, food waste, hearth ash, human and animal excreta, and broken or discarded tools. Within archaeology, middens represent some of the richest high-resolution archives of past human diet, seasonal mobility, resource intensification, and paleoclimate. Whereas everyday habitation floors were frequently swept clean during occupation, adjacent refuse dumps and coastal shell mounds preserve densely packed concentrations of animal bones (zooarchaeology), charred seeds and phytoliths (archaeobotany), lithic debitage, ceramic sherds, and often formal human burials. Crucially, in shell middens dominated by marine or freshwater mollusk valves, the dissolution of biogenic calcium carbonate (CaCO₃) neutralizes acidic groundwater, creating an alkaline geochemical microenvironment (pH 7.5–8.5) that prevents the chemical leaching of hydroxyapatite (Ca₁₀(PO₄)₆(OH)₂) in delicate fish bones, bird elements, osseous harpoons, and human skeletons.

The modern interdisciplinary scientific study of middens originated in Denmark during the late 1840s. Along the raised Littorina Sea shorelines of northern Jutland and Zealand, enormous banks of oyster (Ostrea edulis), cockle (Cerastoderma edule), mussel (Mytilus edulis), and periwinkle (Littorina littorea) shells stretched for hundreds of meters. Geologists initially interpreted these shell banks as natural post-glacial marine beach ridges.

In 1848, the Royal Danish Academy of Sciences and Letters established the First Kitchen Midden Commission, uniting three scholars across distinct disciplines: the archaeologist Jens Jacob Asmussen Worsaae, the zoologist Japetus Steenstrup, and the geologist Johan Georg Forchhammer. Excavating at Meilgaard and Ertebølle, the commission proved that the shell banks were wholly anthropogenic kitchen middens (køkkenmøddinger) accumulated by Mesolithic hunter-fisher-foragers:

1. The deposits lacked juvenile mollusks and natural beach-sorted sand or gravel laminations, consisting exclusively of adult edible shells pried open by humans.
2. Stratified throughout the shells were wood-ash hearths, flint debitage, tranchet axes, antler tools, and pointed-base ceramic vessels.
3. Steenstrup noted that mammalian long bones within the middens had been split systematically for marrow extraction and exhibited gnawing marks on their spongy epiphyses produced by domestic dogs (Canis lupus familiaris), establishing the foundation of modern taphonomy and zooarchaeology.

In Michael Brian Schiffer's behavioral site-formation framework, domestic middens are classified according to how refuse was deposited relative to the locus of activity:

• Primary Refuse: Material discarded at the exact spot of manufacture or consumption (such as microdebitage falls at a knapping station or fish scales around a hearth).
• Secondary Refuse: Waste swept up from domestic house floors, workshops, or butchery areas and dumped in a designated exterior zone—such as an abandoned storage pit, a defensive ditch, a tell slope (as seen between mudbrick houses at Çatalhöyük), or a specialized coastal midden.

Calcium Carbonate Buffering and Micromorphology

In humid temperate and tropical environments where podzolic or lateritic soils exhibit acidic pH values below 5.5, uncalcined bone and shell dissolve within decades to centuries. Inside a shell midden, however, meteoric rainwater charged with dissolved carbonic acid (H₂CO₃) reacts first with the abundant aragonite and calcite (CaCO₃) of discarded mollusk shells (CaCO₃ + H₂CO₃ → Ca²⁺ + 2HCO₃⁻). This bicarbonate buffering stabilizes interstitial porewater at pH 7.8 to 8.4, below the solubility threshold of bone bioapatite. As geoarchaeology and thin-section soil micromorphology demonstrate, wood ash from domestic hearths contributes additional calcite (pseudomorphs after plant calcium oxalate crystals) and siliceous phytoliths (SiO₂·nH₂O), cementing midden lenses into durable breccias that preserve intact human skeletons, juvenile fish otoliths, and ancient DNA.

Methodological Challenges: The Marine Reservoir Effect in ¹⁴C Dating

When dating shell middens via radiocarbon dating (¹⁴C), archaeologists must correct for the Marine Reservoir Effect (MRE). Whereas terrestrial plants fix atmospheric CO₂ in real time, surface ocean waters mix with upwelled deep ocean water depleted in ¹⁴C over centuries of thermohaline circulation. Consequently, marine shells and fish bones yield apparent radiocarbon ages that are 300 to 800 years older than contemporaneous terrestrial charcoal (quantified globally via the Marine20 calibration curve plus local reservoir offsets, ΔR). Reliable midden chronologies therefore pair terrestrial short-lived plant seeds or ungulate bones directly with adjacent marine shells to calibrate local ΔR values and establish shell accumulation rates.

Beyond cataloging species lists, midden mollusks and fish otoliths (ear stones composed of CaCO₃) function as sub-seasonal biological thermometers through sclerochronology and stable oxygen isotope geochemistry:

1. Incremental Growth-Line Sclerochronology: Bivalves such as the hard clam (Mercenaria mercenaria) and cockle (Cerastoderma edule) precipitate daily and tidal microscopic growth increments within their outer prismatic or cross-lamellar shell layers. During rapid spring and summer growth, increments are wide and translucent; during winter cold stress or spawning, growth slows and deposits a dense, dark translational check ring. By thin-sectioning the shell margin at the ventral lip (the final increment precipitated immediately before the shellfish was harvested and killed by the forager), analysts determine the exact season of collection.
2. Carbonate Oxygen Isotope Thermometry (δ¹⁸O): The ratio of heavy oxygen-18 to light oxygen-16 (δ¹⁸O_shell) incorporated into mollusk shell carbonate is governed by thermodynamic isotopic fractionation between water (δ¹⁸O_water) and CaCO₃ as an inverse function of ambient sea surface temperature (T), decreasing by approximately −0.22‰ per +1 °C increase in water temperature. By micro-milling sequential carbonate powder samples along a shell's ontogenetic growth axis from the umbo to the terminal ventral edge and measuring δ¹⁸O via isotope-ratio mass spectrometry (IRMS), archaeologists reconstruct the annual sinusoidal sea-temperature curve and pinpoint whether the midden was occupied solely during winter scarcity, summer aggregation, or year-round.

Cross-cultural research over the past three decades has overturned the utilitarian assumption that all large shell middens were passive trash heaps. Across multiple continents, coastal societies intentionally engineered shell mounds as monumental architecture, territorial landmarks, and ancestral mortuary landscapes:

• The Earliest Marine Shell Middens (South Africa, ~164 ka): At Pinnacle Point Cave 13B on the southern coast of South Africa, Curtis Marean and colleagues documented cemented shell midden lenses dated by OSL to 164,000 years ago (during glacial Marine Isotope Stage 6), proving that early Homo sapiens tracked lunar spring-tide cycles to harvest brown mussels and limpets while processing red ochre and bladelets.
• Brazilian Sambaquis (~6000 cal BCE – 1000 CE): Along the southern Atlantic coast of Brazil (Santa Catarina and Rio de Janeiro), pre-colonial fisher-foragers constructed thousands of monumental shell mounds (sambaquis, from the Tupi tamba'kī). The largest sambaquis—such as Garopaba do Sul and Figueirinha—rise 30 to 50 meters high (the height of a 15-story building) and contain millions of tons of layered shells, whale bones, and clay platforms built over 1,000 years to house hundreds of structured human burials accompanied by polished stone zoomorphic sculptures (zooliths).
• Japanese Jōmon Kaizuka and Florida Calusa Shell Islands: In Japan's Tokyo Bay basin, Over 2,500 Jōmon shell mounds (kaizuka, ~9500–300 cal BCE) form giant circular and horseshoe-shaped rings up to 170 meters across surrounding swept village plazas and dog burials. Along the Gulf Coast of Florida, the non-agricultural Calusa Kingdom (~500–1600 CE) engineered Mound Key—a 50-hectare artificial island metropolis built out of 500,000 cubic meters of oyster and lightning-whelk shells, bisected by a 30-meter-wide grand canal, engineered fish-holding water courts, and flat-topped shell temple mounds rising 10 meters above Estero Bay.

1. Stein, J. K. (Ed.). (1992). Deciphering a Shell Midden. Academic Press. https://www.elsevier.com/books/deciphering-a-shell-midden/stein/978-0-12-664730-3
2. Andersen, S. H. (2000). "'Køkkenmøddinger' (shell middens) in Denmark: a survey." Proceedings of the Prehistoric Society, 66, 361–384. https://doi.org/10.1017/S0079497X00001857
3. Marean, C. W., Bar-Matthews, M., Bernatchez, J., Fisher, E., Goldberg, P., Herries, A. I., ... & Williams, H. M. (2007). "Early human use of marine resources and pigment in South Africa during the Middle Pleistocene." Nature, 449(7164), 905–908. https://doi.org/10.1038/nature06204
4. Álvarez, M., Briz Godino, I., Balbo, A., & Madella, M. (2011). "Shell middens as archives of past environments, human dispersal and specialized resource management." Quaternary International, 239(1–2), 1–7. https://doi.org/10.1016/j.quaint.2010.10.025
5. Thompson, V. D., Marquardt, W. H., Cherkinsky, A., Roberts Thompson, A. D., Walker, L. J., Newsom, L. A., & Savarese, M. (2016). "From shell midden to midden-mound: The geoarchaeology of Mound Key, an anthropogenically engineered island in Southwest Florida, USA." PLOS ONE, 11(4), e0154611. https://doi.org/10.1371/journal.pone.0154611