Reading the Knossos Storeroom Sealings as Evidence for Administration
A student is analysing 1,240 clay sealings recovered from a single storeroom at Knossos. Of these, 38% carry a single impression from one seal, 54% carry two or more impressions from different seals, and 8% show no impression at all but retain cord holes. The sealings were found in a destruction deposit dated by pottery to Late Minoan IB. No Linear B tablets were recovered from this room, though 47 tablets came from an adjacent corridor. The student's draft claims: 'The multiple impressions prove that two officials jointly authorised every transaction, and the absence of tablets in the storeroom shows the room had no administrative function.' A supervisor flags both sentences as overreaching.
1. Identify the two inferential errors in the student's draft and explain what the sealing evidence can and cannot support.
Read the worked answer
The first error is treating a repeated association as a rule. Multiple impressions on a sealing are consistent with several procedures: multiple seals may be applied by one official using more than one seal, by different parties at different stages (sender, carrier, recipient), or as tamper-evident closure rather than joint authorisation. The 54% figure establishes that multi-seal closure was common; it does not establish joint authorisation, still less that 'every transaction' was jointly approved. The second error is an argument from absence in a disturbed context. The deposit is a destruction deposit, and the adjacent corridor yielded 47 tablets, so the absence of tablets in the storeroom is weak evidence. Absence only supports a negative claim when preservation and recovery are shown to be comparable. The defensible claim is narrower: the sealings indicate that goods or containers in this storeroom were routinely secured and that sealing was a normal administrative act; the distribution of seal types can support inferences about how many sealing authorities were active and how impressions cluster. It cannot by itself identify offices, authorisation procedures, or the room's function.
2. The student wants to strengthen the argument by comparing the 38% single-impression sealings with the 54% multi-impression sealings. Design a comparison that would actually test an administrative hypothesis, and state what result would count against it.
Read the worked answer
A testable hypothesis: if multiple impressions mark transactions involving more than one party, then multi-impression sealings should differ systematically from single-impression ones in associated variables such as commodity type, container form, find location within the storeroom, and the number of distinct seal devices represented. The comparison should therefore be stratified: group sealings by impression count, then cross-tabulate against (a) the seal devices identified, (b) the shape or capacity of the container implied by the sealing, and (c) horizontal position in the storeroom. A result that counts against the hypothesis would be multi-impression sealings showing no association with commodity, container, or location, and drawing on the same small set of seal devices as single-impression sealings. That pattern would suggest multi-sealing is a routine mechanical practice rather than a marker of multi-party transactions. A result that supports it would be multi-impression sealings concentrated on particular commodities or container types and drawing on seal devices not found on single-impression sealings, which would indicate distinct sealing authorities tied to particular goods.