The bones that survive are not a transparent inventory
An assemblage can contain many fragments from few animals, few fragments from many animals, or an uneven set of skeletal parts shaped by transport, processing, recovery and decay. This guide shows how abundance measures and density checks work together without turning preservation into a single-cause explanation.
Choose a reading route
- Start from first principles — define the unit and the threat.
- Follow the worked example — make the hidden arithmetic visible.
- Test transfer — change the setting and preserve the rule.
- Audit the evidence — inspect records, assumptions and failure points.
- Use the practical checklist — apply the method to new work.
Chapter contents
Foundations: Chapters 1–3
Method: Chapters 4–6
Transfer and limits: Chapters 7–9
Connections and use: Chapters 10–12
1. Abundance begins with a counting rule
Taxonomic abundance describes how much evidence for each animal group is present, but the answer depends on what is counted. Number of identified specimens (NISP), minimum number of elements (MNE), minimum number of individuals (MNI) and minimum animal units (MAU) answer different questions.
Record the identification threshold, anatomical zone, side, age, fusion and context before calculating summary measures. A count without its protocol cannot be reproduced.
First principle
Choose a unit that matches the inference; never treat one abundance number as the animals themselves.
2. Skeletal parts carry behavioural and taphonomic signals
Heads, axial bones, upper limbs, lower limbs and feet differ in food value, transport cost, identifiability and durability. Human transport, butchery, craft working, carnivore access and recovery methods can all create anatomical profiles.
| Pattern | Possible process | Alternative |
|---|---|---|
| Many heads and feet | Primary processing | Dense parts survive better |
| Few small bones | Recovery loss | Small animals absent |
| Upper limbs common | Transport choice | Identification preference |
| Burnt fragments | Cooking or disposal | Later fire exposure |
Interpret representation with stratigraphy, cut marks, burning, gnawing, fragmentation and recovery records.
← Previous · Contents · Next →
3. Density-mediated attrition is a filter, not a verdict
Bone portions differ in mineral density and resistance to destruction. If low-density portions disappear preferentially, the surviving profile can resemble selective transport even when deposition was broader.
Compare an abundance measure such as %MAU with an appropriate density reference, while checking taxon, age and measurement compatibility. A positive association can support density-mediated attrition; it does not prove that density was the only cause.
- Use the same anatomical units in density and abundance data.
- Check small sample instability and tied ranks.
- Separate contexts before pooling.
- Report portions that could not be scored.
← Previous · Contents · Next →
4. Worked example: two taxa, three measures
A fictional deposit contains 120 identified cattle fragments and 60 identified sheep/goat fragments. After refitting and zoning, cattle have MNE 24 and MNI 6; sheep/goat have MNE 30 and MNI 10.
| Taxon | NISP | MNE | MNI |
|---|---|---|---|
| Cattle | 120 | 24 | 6 |
| Sheep/goat | 60 | 30 | 10 |
NISP makes cattle appear twice as abundant. MNE and MNI reverse the ranking. The result does not tell us which metric is “correct”; it tells us cattle fragments are more numerous per inferred element and individual.
A defensible report presents the counts together and explains which one supports each inference.
← Previous · Contents · Next →
5. Standardise element zones and aggregation
MNE depends on how fragments are assigned to diagnostic zones and how left/right, proximal/distal, age and size are reconciled. MAU then divides MNE by the number of that element in one skeleton. Standardisation helps compare body regions but does not erase identification uncertainty.
| Decision | Effect on profile | Required record |
|---|---|---|
| Broad zones | More fragments merged | Zone definitions |
| Fine zones | More sparse cells | Illustrated landmarks |
| Pool phases | Higher minima | Aggregation rule |
| Split contexts | Lower counts per cell | Context integrity |
Freeze the zoning and aggregation plan before comparing taxa or phases.
← Previous · Contents · Next →
6. Separate preservation, recovery and behaviour
Use multiple evidence streams. Density association addresses survival; mesh and collection strategy address recovery; cut marks and articulation address processing; spatial and stratigraphic patterning address deposition and mixing.
A model with one cause is rarely sufficient. State which observations each hypothesis predicts, then look for contradictions as well as supporting patterns.
Practical check
Build a table with skeletal region, NISP, MNE, %MAU, density, fragmentation, burning and recovery fraction. Inspect the rows together.
← Previous · Contents · Next →
7. Transfer test: burnt ritual deposit
Move the method to a deposit containing highly burnt and calcined bone. Heat changes fragmentation, shrinkage and identifiability, while collection and flotation may recover different size fractions. Ordinary density references may not represent the altered material.
Transfer succeeds when the analysis treats burning as a new preservation regime, reports indeterminate fragments and compares like with like. A missing element may reflect combustion and recovery rather than deliberate exclusion.
The same reasoning applies to waterlogged, weathered and carnivore-accumulated assemblages: recalibrate the filter before interpreting behaviour.
← Previous · Contents · Next →
8. Misconceptions and limits
- “More fragments mean more animals.” One element can break into many identifiable pieces.
- “MNI solves fragmentation.” MNI is sensitive to aggregation, siding and age.
- “A density correlation proves attrition.” Behaviour can also favour dense parts.
- “No correlation means no destruction.” Small samples and poor reference matching can hide it.
- “Skeletal profiles directly show diet.” Transport, processing, disposal and recovery intervene.
Preservation analyses narrow plausible stories; they do not reconstruct every lost bone.
← Previous · Contents · Next →
9. Evidence audit
Audit context integrity, collection strategy, mesh size, reference collection, identification rules, zoning, refits, taxonomic resolution, age and side criteria, density source and statistical code. Preserve specimen-level data so summaries can be regenerated.
| Audit question | Evidence | Warning sign |
|---|---|---|
| Was recovery consistent? | Field and sieve logs | Unknown fraction |
| Are taxa comparable? | Identification criteria | Different resolution |
| Does density match? | Reference and element zones | Mixed species or ages |
| Can counts be rebuilt? | Specimen table | Totals only |
← Previous · Contents · Next →
10. Questions readers often ask
Should NISP or MNI be preferred?
Neither universally. Present multiple measures when they illuminate different biases.
What does %MAU add?
It standardises element abundance by the number of each element in a skeleton, helping compare body regions.
Can density tests identify transport decisions?
Only indirectly. Combine them with butchery, articulation, spatial context and independent economic evidence.
← Previous · Contents · Next →
11. Reading routes and ecosystem connections
Begin with the Gold Standard of Education: learning that transfers into life, The Gold Standard and the How X Works hub. The subject foundation is the Gold Standard of Archaeology.
Continue with guides on assemblage integrity and refitting, recovery fractions and mesh bias, bulk samples and rare-find detection, and MNI, fragmentation and aggregation bias.
← Previous · Contents · Next →
12. Sources and practical checklist
Authoritative and primary references include Historic England’s Animal Bones and Archaeology: Recovery to Archive and rapid mammal-bone recording method, plus the open archaeological case study At the Feet of the Fortress, which reports NISP, MNE, MNI, MAU and a density-attrition test.
- State the analytical unit.
- Keep contexts and recovery fractions visible.
- Report NISP, MNE and MNI where useful.
- Define zones and aggregation rules.
- Match density references to the material.
- Test behavioural and taphonomic alternatives.
- Archive specimen-level data.