Tokens and accounting objects

Tokens and accounting objects examines the use of small, portable objects to represent quantities, categories, obligations and transactions before and alongside written accounting.

Its central claim is:

Tokens made information physically manipulable: people could represent goods and obligations with objects that could be counted, sorted, grouped, transferred, sealed and checked.

A tally preserves a quantity by accumulating repeated marks. A token system goes further. It can detach the representation from the surface and turn each information unit into a movable object.

A cone, sphere, disc, cylinder, pebble, bead or specially shaped clay piece may stand for:

• one item;
• a conventional bundle of items;
• a measure of grain or liquid;
• a category of commodity;
• labour owed or delivered;
• a ration;
• a debt;
• a transfer;
• an administrative claim.

Because the objects are movable, a user can operate on the representation itself.

Tokens can be:

• added;
• removed;
• sorted by type;
• grouped into sets;
• placed into containers;
• handed from one person to another;
• compared with another collection;
• enclosed and sealed;
• impressed into clay;
• reconciled against delivered goods.

This creates a major change in information processing.

The record is no longer only a static statement. It can become a physical model of a stock, flow or obligation.

If five tokens represent five sheep, moving one token from a “pasture” group to a “delivered” group can update the represented state without moving the animals themselves. If several token shapes represent different commodities, one container can hold a miniature administrative world: grain, animals, oil, labour or other categories represented by standardised objects.

The best-known archaeological case comes from Neolithic and Chalcolithic West Asia. Small geometric clay objects occur from the early Neolithic, beginning in the tenth or ninth millennium BCE at some sites, and continue for thousands of years [1][2]. Denise Schmandt-Besserat developed the most influential interpretation of these objects: plain and complex tokens formed a long-lived accounting system whose categories and shapes contributed to numerical impressions and early cuneiform signs [3][4].

The model explains several important observations:

• geometric clay objects long predate writing;
• their forms recur across sites;
• some late examples are found inside sealed clay envelopes;
• impressions on envelope surfaces can correspond to enclosed objects;
• early numerical tablets use impressions resembling counters;
• the earliest Mesopotamian writing is overwhelmingly administrative;
• signs and numerical systems emerge from earlier bookkeeping practices.

The classic transition is often summarised as:

loose tokens → sealed token envelopes → external token impressions → numerical tablets → proto-cuneiform writing

This sequence is powerful, but it should not be treated as an uncontested conveyor belt.

Recent scholarship has complicated several parts of the traditional narrative.

First, “token” is an interpretation, not a neutral description. Small geometric clay objects may have served multiple functions. Depending on context, some may have been:

• counters;
• mnemonic devices;
• tools;
• gaming pieces;
• teaching aids;
• ritual offerings;
• models;
• ornaments;
• objects associated with household production.

Large assemblage studies have argued that early Neolithic examples should not automatically be treated as parts of a region-wide accounting code [5][6]. Context, wear, manufacture, deposition and local variation matter.

Second, the story that writing replaced tokens is too neat. Archaeological evidence shows that token-like accounting objects continued to be used alongside tags, seals, sealings and written documents into the first millennium BCE [7][8]. Older tools did not disappear simply because scribes acquired styluses. A movable counter may remain useful even in a literate bureaucracy, just as a modern warehouse can use databases, printed labels and physical bins simultaneously.

Third, the relationship between token shapes and written signs is strongest for some plain numerical counters and numerical impressions, but weaker or disputed for many “complex tokens.” The claim that a single coherent token vocabulary directly generated the full sign inventory of proto-cuneiform is more ambitious than the evidence securely supports [9][10].

Fourth, accounting did not emerge only because people suddenly invented clever shapes. It depended on a social world containing:

• countable and categorised goods;
• recurring production and distribution;
• delayed obligations;
• specialists or household managers;
• accepted units;
• trusted procedures;
• containers, seals and custodians;
• organisations capable of enforcing records.

The token is therefore both a cognitive technology and a social contract.

Its meaning does not reside in clay alone. A clay cone represents something only because users share a convention about:

• what category it denotes;
• how much it represents;
• who may issue it;
• whether it can be transferred;
• what counts as settlement;
• how disputes are resolved.

This produces the topic’s most important systems insight:

An accounting object is a material sign embedded in a procedure. Remove the procedure and the object becomes an oddly confident pebble.

Token systems reduce several constraints:

• they reduce reliance on memory;
• they separate quantity from the physical goods counted;
• they preserve state across time;
• they make heterogeneous inventories sortable;
• they support reconciliation and audit;
• they enable delayed exchange;
• they make obligations visible;
• they support administration beyond face-to-face trust.

They also introduce limitations.

A loose token can be lost, substituted or misunderstood. Its referent may be ambiguous. A container can conceal its contents. A seal can authenticate an enclosure but also centralise authority. A token collection may represent quantities and categories but struggle to record names, motives, conditions, exceptions or narrative explanation. The system depends heavily on a shared codebook and institutional context.

These limitations encourage several successor technologies:

• sealings and authenticated containers;
• impressed clay envelopes;
• numerical tablets;
• written accounting;
• numerical notation;
• ledgers;
• receipts;
• coinage and transferable financial instruments;
• abaci and counting boards;
• databases and inventory systems.

The research notes recommends keeping Tokens and accounting objects in the map but changing its analytical name to:

Manipulable Accounting Tokens and Administrative Objects

The public-facing title Tokens and Accounting Objects remains clear.

The topic should be upgraded from Supporting to Core because it represents a decisive shift:

Information becomes not merely visible, but operable.

The big idea

| Question | Verdict | |---|---| | Does the topic belong in the map? | Yes, as a Core transition | | Is the register name adequate? | Yes for public use, although “token” must be treated as a functional interpretation | | Recommended analytical name | Manipulable accounting tokens and administrative objects | | Primary contribution | Represents quantities, categories and obligations as movable physical objects | | Main successor advantage | Numerical and written records compress more information onto portable surfaces and can identify actors, commodities and conditions more explicitly | | Main caution | Not every geometric clay object was an accounting token, and the pathway from tokens to writing was neither universal nor a simple replacement sequence | | Taxonomy pressure | The project needs a formal field for manipulability and a clearer distinction between storage objects, processing tools and authenticated records |

| Field | Provisional value | |---|---| Tokens and accounting objects | | Register name | Tokens and accounting objects | | Recommended analytical name | Manipulable accounting tokens and administrative objects | | Topic type | Material accounting system, external-memory tool and manipulable representation | | Primary category | Storage and persistence | | Secondary categories | Processing; encoding; governance; authentication; distribution; economic coordination | | Approximate emergence | Small geometric clay objects occur from the tenth or ninth millennium BCE in parts of West Asia; secure complex accounting use is especially visible by the fourth millennium BCE | | Main problem addressed | Dependence on memory and direct inspection for tracking quantities, categories, stocks, transfers and obligations | | Key predecessors | Counting; one-to-one correspondence; tallying; categorisation; exchange; biological memory | | Key successors | Numerical impressions; sealings; clay envelopes; numerical tablets; proto-cuneiform; written accounting; ledgers; abaci; inventory systems | | Current status | Researched and provisionally synthesised |

For this project, an accounting token system is defined as:

A socially governed method in which portable material objects represent quantities, categories, transactions or obligations and can be physically manipulated to create, update, transfer, secure or verify an informational state.

This definition has eight parts.

3.1 Socially governed

The object does not carry an accounting meaning by nature.

A sphere is not born representing grain. A cone does not wake up owing somebody three goats.

Users must share rules concerning:

• what the object represents;
• the unit or bundle value;
• who may create or issue it;
• where it may be used;
• whether it is transferable;
• how the record is settled;
• how mistakes or fraud are detected.

3.2 Portable

Unlike a notch fixed to a wall, stick or bone, a token can be moved independently.

Portability enables:

• rearrangement;
• grouping;
• transfer;
• containment;
• physical calculation;
• state change.

3.3 Material object

The representation has a physical form.

Possible materials include:

• clay;
• stone;
• bone;
• shell;
• wood;
• metal;
• ceramic;
• glass;
• seeds or pebbles used conventionally.

The West Asian archaeological tradition focuses heavily on clay because clay is abundant, shapeable and archaeologically durable.

3.4 Representation

The object stands for something beyond itself.

Its referent may be:

• a discrete item;
• a measure;
• a category;
• a person;
• a unit of labour;
• a transaction;
• a claim;
• an entitlement;
• an obligation.

3.5 Quantity and category

Tokens may encode more than quantity.

A collection can use:

• repeated identical objects for multiple units;
• different shapes for different commodities;
• different sizes for higher-order values;
• surface markings for subtypes or qualifications;
• containers or locations for account categories.

3.6 Manipulation

The user can act upon the representation.

Operations may include:

• addition;
• subtraction;
• sorting;
• grouping;
• matching;
• partitioning;
• transfer;
• enclosure;
• impression;
• reconciliation.

3.7 Informational state

The arrangement or collection represents a condition at a particular moment.

Examples:

• stock currently held;
• animals delivered;
• rations owed;
• labour completed;
• goods transferred;
• balance remaining.

3.8 Procedure

An accounting object is useful only within an operating process.

The process may require:

1. identifying the referent;
2. assigning the token value;
3. adding or moving the token;
4. storing or sealing the collection;
5. presenting it to another party;
6. reconciling it against goods or actions;
7. closing or updating the account.

The token is therefore not merely a miniature noun. It is a component inside an administrative algorithm.

4.1 It makes quantities manipulable

A tally preserves a count. A token allows the count to be rearranged.

This supports operations before formal arithmetic notation.

4.2 It separates the model from the goods

A herd is difficult to place on a table.

A set of counters can represent it in miniature.

The information can be examined without gathering the entire referent set.

4.3 It represents heterogeneous inventories

Different token classes can represent different goods.

One collection can therefore model a mixed inventory rather than only one repeated quantity.

4.4 It supports interrupted administration

A person can stop, store the tokens and resume later.

The record survives the interruption.

4.5 It supports transfer

A token or sealed collection can move between people and places.

This makes information about goods travel separately from the goods themselves.

4.6 It supports delayed obligations

A record can outlive the original conversation.

This matters when delivery, payment, labour or distribution occurs later.

4.7 It enables reconciliation

A represented quantity can be compared with:

• goods delivered;
• stock remaining;
• another record;
• an external impression;
• a sealed contents set.

4.8 It enables audit

Because the record is material and inspectable, another person can check it.

Audit is not guaranteed, but it becomes possible.

4.9 It supports authentication

Seals, envelopes and controlled procedures can make later alteration more detectable.

4.10 It externalises administrative categories

Different shapes force a system to decide what counts as the same kind of thing.

The record therefore does not merely reflect the economy. It helps organise it.

4.11 It prepares the ground for written accounting

Impressing tokens or token-like styli into clay moves representation from a collection of objects to a patterned surface.

4.12 It creates a path from objects to signs

The three-dimensional object can become a two-dimensional impression and eventually a written sign.

This transition is strongest where archaeological correspondence can be demonstrated, but it should not be generalised carelessly.

4.13 It reveals information processing before computers

Tokens are early data structures.

They have:

• types;
• values;
• states;
• containers;
• operations;
• access rules;
• validation procedures.

Calling them “ancient databases” would be too loose. Recognising their computational logic is not.

5.1 Token

A portable object conventionally standing for a referent, category, value, claim or action.

In archaeology, the term often refers to small shaped clay objects interpreted as counters or administrative devices.

5.2 Counter

An object used to represent a unit or value during counting or calculation.

5.3 Calculus

A Latin term meaning a small stone or pebble, later associated with counting and calculation.

The plural is calculi.

5.4 Accounting object

A broader analytical term for a material object used to record, model, authorise or verify economic information.

5.5 Plain token

In Schmandt-Besserat’s typology, a relatively simple geometric token without elaborate markings.

Examples may include spheres, cones, discs, cylinders and tetrahedra.

5.6 Complex token

A shaped object with incisions, punctures, appliqué or other elaboration.

Claims linking complex tokens directly to proto-cuneiform commodity signs are more debated than some links between plain tokens and numerical impressions.

5.7 Commodity

A good treated as belonging to an exchangeable or administratively recognised category.

5.8 Inventory

A record or represented set of goods held at a particular time.

5.9 Stock

A quantity held or available.

5.10 Flow

A quantity moving into, through or out of a system over time.

5.11 Obligation

A socially recognised requirement to deliver goods, labour, payment or another performance.

5.12 Entitlement

A socially recognised claim to receive goods, labour, access or distribution.

5.13 Ration

A standardised allocation of food, drink or another commodity.

5.14 Bulla

A term with several archaeological uses.

It may refer broadly to a clay sealing. It is often used for hollow clay envelopes containing tokens, but some recent scholarship recommends reserving terminology more carefully because not every bulla is an envelope and not every envelope performs the same function [21].

5.15 Clay envelope

A hollow clay container enclosing tokens or, in later contexts, a clay tablet.

5.16 Sealing

Clay or another material impressed with a seal and attached to a container, door, cord, package or document to indicate closure or authority.

5.17 Seal

An engraved object used to create a distinctive impression.

5.18 Seal impression

The mark produced by pressing a seal into clay or another receptive surface.

5.19 Numerical impression

A mark impressed into clay to represent quantity or measure.

5.20 Numerical tablet

A clay tablet bearing numerical impressions without the full sign repertoire of proto-cuneiform writing.

5.21 Proto-cuneiform

The earliest Mesopotamian writing tradition, emerging in the late fourth millennium BCE and used primarily for administration.

5.22 Fungibility

The property by which one unit can be substituted for another equivalent unit.

Tokens can make heterogeneous real goods administratively fungible by representing them through standard units.

5.23 Reconciliation

The process of comparing records, represented quantities or actual goods to identify agreement or discrepancy.

5.24 Audit trail

Evidence linking a recorded state to the actions or transactions that produced it.

Early token systems may support partial auditability, but they rarely preserve the detailed chronological trails expected of modern accounting.

5.25 Tokenisation

The conversion of something into a representative token.

Modern digital tokenisation is conceptually related but technologically distinct from clay counters.

6.1 Boundary with Tally marks and notches: Tallying and Accumulative Quantity Marks

A tally typically records units by adding marks or fixed increments.

A token is independently movable.

This mobility allows sorting, grouping and transfer.

6.2 Boundary with Numerical notation: Numerical Notation

Tokens represent quantity materially through objects.

Numerical notation represents quantity through conventional signs.

A token system may embody numerical values without using written numerals.

6.3 Boundary with Pictograms and ideograms: Conventional Visual Signs

A pictogram is primarily viewed and interpreted.

A token is often physically manipulated.

Some tokens may become signs when impressed into clay.

6.4 Boundary with Clay tablets: Clay Tablets

A token is a discrete object representing information.

A tablet is a surface supporting impressions or writing.

The transition between them is historically important but analytically distinct.

6.5 Boundary with seals and sealings

A token represents a quantity, category or claim.

A seal impression primarily identifies authority, closure or provenance.

Combined systems can represent both what is recorded and who authorised it.

6.6 Boundary with money

A token can represent a commodity or obligation without functioning as generally accepted payment.

Money typically requires broader functions such as:

• medium of exchange;
• unit of account;
• store of value;
• settlement mechanism.

Accounting tokens should not automatically be called money.

6.7 Boundary with game pieces

A game piece represents state within a rule-governed simulation.

An accounting token represents external stocks, flows or obligations.

The same object form could potentially serve either function. Context is decisive.

6.8 Boundary with figurines and ritual miniatures

A miniature object may symbolise a person, animal or offering without recording a quantity or transaction.

6.9 Boundary with weights

A weight acts through its physical mass.

A token acts through assigned symbolic value.

A weight can also bear marks or seals, but its core function differs.

6.10 Boundary with abaci and counting boards

An abacus uses counters within a structured positional or place-value framework.

Loose tokens may lack fixed positions and formal arithmetic rules.

6.11 Boundary with receipts

A receipt is evidence that a transaction occurred.

A token collection may record an obligation, stock or transfer, but whether it functions as a receipt depends on procedure and social recognition.

6.12 Boundary with ownership markers

An ownership marker identifies association with a person or group.

An accounting token records quantity, category or obligation.

An object may perform both functions.

7.1 Participant structure

Common patterns include:

• individual-to-self;
• household-to-household;
• producer-to-administrator;
• administrator-to-storehouse;
• institution-to-worker;
• debtor-to-creditor;
• sender-to-receiver through an authorised carrier.

7.2 Time relationship

Primarily asynchronous.

The represented state survives the original counting or transaction event.

7.3 Spatial relationship

Local in early household contexts, but potentially regional when sealed records or tokens travel with goods.

7.4 Persistence

Moderate to high, especially when clay is dried or fired.

Loose collections remain vulnerable to loss and mixing.

7.5 Direction

Usually one-to-one or institution-to-individual, but records may support many-to-one aggregation.

7.6 Interactivity

High during manipulation and reconciliation.

Low once enclosed and sealed unless the enclosure is opened or external information is provided.

7.7 Addressability

Addressability depends on associated containers, locations, seals, custodians or procedures.

A loose token may represent “one unit” without identifying whose account it belongs to.

7.8 Feedback

Feedback occurs through:

• matching goods to tokens;
• returning or cancelling tokens;
• breaking an envelope;
• comparing impressions with contents;
• resealing or updating the record.

8.1 Source state

A real-world quantity, category, transfer or obligation.

8.2 Unit definition

The socially accepted correspondence between token and referent.

8.3 Encoder

The counter, household manager, clerk, storekeeper or administrator.

8.4 Encoding act

Selecting, shaping, adding, moving or enclosing the relevant tokens.

8.5 Signal

The token collection, arrangement, enclosure, impression or sealed package.

8.6 Channel

Physical handling, storage, transport and presentation.

8.7 Storage

A container, bag, bowl, room, envelope, tablet or institutional archive.

8.8 Authentication layer

A seal, distinctive container, authorised custodian or controlled procedure.

8.9 Receiver

The person or institution expected to interpret or reconcile the record.

8.10 Decoding

Recognising the token classes, quantities, account context and procedural meaning.

8.11 Interpretation

Concluding what is held, owed, delivered, allocated or remaining.

8.12 Response

Distributing goods, updating the collection, settling the obligation, disputing the record or issuing a new record.

8.13 Feedback

Comparison between represented and actual state.

8.14 Noise

Physical noise

• broken tokens;
• missing pieces;
• mixed collections;
• eroded marks;
• damaged envelopes.

Categorical noise

• ambiguous shapes;
• inconsistent classification;
• local variations;
• obsolete token types.

Procedural noise

• incorrect counting;
• unauthorised additions;
• failure to update;
• mistaken account assignment.

Social noise

• disagreement over unit values;
• disputed authority;
• fraud;
• coercive accounting.

Archaeological noise

• incomplete excavation;
• poor early recording;
• objects separated from context;
• functional assumptions based on shape alone.

9.1 One object per unit

The simplest mechanism uses one token for one unit.

9.2 One object per bundle

A different shape or size may represent a higher-order quantity.

This compresses the representation.

9.3 Shape as category

Different forms may identify different commodities or measures.

The degree to which this operated consistently across early periods remains debated.

9.4 Surface modification

Incisions, punctures or applied clay may distinguish subtypes or values.

9.5 Grouping

Tokens placed together form an account set.

9.6 Spatial partitioning

Different bowls, compartments or locations can represent different accounts or states.

9.7 Transfer

Moving a token can represent moving a good, responsibility or claim.

9.8 Enclosure

Sealing tokens inside clay fixes a collection at a particular moment.

9.9 External impression

Pressing tokens into wet clay can expose the represented contents without opening the enclosure.

9.10 Seal impression

A seal can indicate the authority or office responsible for the package.

9.11 Replacement by surface marks

Once the exterior impressions carry the required information, the enclosed tokens may become redundant for some purposes.

9.12 Stylus abstraction

A shaped stylus can reproduce token-like numerical marks without using the token itself.

9.13 Integration with commodity signs

Numerical impressions combined with signs identifying goods create more explicit administrative records.

9.14 Reconciliation

The system closes the loop by comparing:

• represented quantity;
• delivered quantity;
• stored quantity;
• authorised quantity.

10.1 Exact or conventional counting

Users need a repeatable way to map tokens to units or bundles.

10.2 Category formation

The system requires decisions about which goods count as equivalent.

10.3 Material shaping

Users need access to clay or another suitable material and the skill to produce recognisable forms.

10.4 Surplus and storage

Accounting becomes more valuable when goods accumulate and persist beyond immediate consumption.

10.5 Recurring exchange or redistribution

Repeated transactions justify stable recording procedures.

10.6 Delayed settlement

Obligations become record-worthy when action and settlement occur at different times.

10.7 Shared conventions

Participants must understand token meanings.

10.8 Custodianship

Someone must preserve and manage the record.

10.9 Trust or authority

Participants must accept the legitimacy of the system or the institution enforcing it.

10.10 Containers and closure

Bags, bowls, jars, envelopes, strings and seals support grouping and security.

10.11 Institutional continuity

Long-term systems require procedures that outlast individual users.

11.1 Early Neolithic background

Small geometric clay objects appear at early Neolithic sites across parts of West Asia from roughly the tenth or ninth millennium BCE [1][5].

Their presence coincides broadly with major changes in settlement, food production, storage and household organisation.

The old register’s wording, “from about the eighth millennium BCE,” should therefore be broadened.

11.2 Early use remains disputed

The existence of shaped clay objects is secure.

Their universal interpretation as accounting devices is not.

Some assemblages may contain multifunctional utilitarian or ritual objects.

11.3 Long Neolithic and Chalcolithic development

Geometric objects continue across many sites and periods.

Regional variation increases.

Some forms become more elaborate.

11.4 Fourth-millennium administrative intensification

Urbanisation, institutional storage and redistribution in Mesopotamia increase demand for information control.

Tokens, sealings, numerical impressions, tags and tablets appear within increasingly complex administrative assemblages [11][12].

11.5 Token envelopes

Hollow clay envelopes containing tokens appear before or around the emergence of writing.

Some bear seal impressions and external marks.

11.6 Numerical tablets

Flat clay tablets bearing numerical impressions emerge in the late fourth millennium BCE.

They allow quantitative information to be recorded directly on a surface [13].

11.7 Proto-cuneiform

By approximately 3300 BCE, numerical records combine with commodity and administrative signs in southern Mesopotamia [11][14].

11.8 Coexistence after writing

Token-like accounting objects remain in use alongside written administration into later periods [7][8].

This is crucial.

Writing does not instantly assassinate every older technology and bury it behind the archive.

11.9 Later material counters

Counting stones, jettons, abacus beads, gaming counters, ration tokens, tickets and inventory markers continue the broad principle of portable material representation.

11.10 Digital continuation

Modern software still uses “tokens,” counters, objects, records and state transitions, although the material substrate and formal systems are radically different.

12.1 Shape

Regular geometric shapes may indicate deliberate standardisation.

Shape alone does not establish function.

12.2 Repetition

Repeated forms across an assemblage can suggest categories or production conventions.

12.3 Context

Find location is critical.

Relevant contexts include:

• storage areas;
• domestic spaces;
• administrative buildings;
• graves;
• ritual deposits;
• workshops;
• refuse;
• sealed containers.

12.4 Association

Association with seals, sealings, tablets, containers or commodities strengthens administrative interpretations.

12.5 Wear and handling

Surface wear may reveal repeated manipulation.

12.6 Manufacturing traces

Fingerprints, shaping methods and firing can reveal production practices and division of labour [17].

12.7 Enclosed contents

Tokens preserved inside envelopes provide unusually strong evidence that objects were grouped as records.

12.8 Impression correspondence

Matching external impressions to enclosed shapes supports the interpretation that the surface recorded the contents.

12.9 Numerical correspondence

Similarity between some token forms and early numerical impressions supports historical continuity.

12.10 Distribution through time

Continued use can challenge replacement narratives.

12.11 Collection bias

Small plain objects were often under-collected or poorly documented in older excavations.

This can distort typologies and distribution maps.

12.12 Functional ambiguity

A single object class may have multiple uses across sites or even within one community.

12.13 Analogy risk

Later documented accounting practices can illuminate earlier objects but cannot automatically prove identical functions.

12.14 Imaging technologies

Computed tomography allows researchers to inspect tokens inside sealed envelopes without breaking them [15][16].

Digital models can reveal internal arrangement, shape and correspondence with exterior marks.

Denise Schmandt-Besserat’s work transformed the study of small clay objects by proposing a long developmental relationship between Neolithic tokens and Mesopotamian writing [3][4].

13.1 Core propositions

The broad model proposes that:

1. early farming communities used tokens to count goods;
2. token shapes represented specific quantities or commodities;
3. increasing economic complexity produced more token types;
4. tokens were enclosed in clay envelopes for secure records;
5. token impressions were placed on envelope surfaces;
6. impressed tablets replaced three-dimensional token collections;
7. commodity and numerical signs contributed to proto-cuneiform writing.

13.2 Strengths

The model:

• connects long-term material traditions;
• explains why early writing is administrative;
• relates three-dimensional objects to two-dimensional impressions;
• highlights the role of accounting in writing’s emergence;
• treats writing as a process rather than a miraculous single invention;
• places cognition inside material practice.

13.3 Strongest evidence

The strongest evidence concerns:

• late fourth-millennium accounting devices;
• token envelopes;
• external impressions;
• plain counters and numerical signs;
• the administrative content of the earliest tablets.

13.4 More ambitious claims

More debated claims include:

• a coherent token code operating across the Near East for millennia;
• stable commodity meanings for many early shapes;
• direct derivation of many complex proto-cuneiform signs from complex tokens;
• universal one-token-per-object correspondence;
• complete replacement of tokens by writing.

13.5 Historical importance

Even where revised, the model remains foundational because it forced scholars to take small, easily ignored objects seriously as information technologies.

14.1 “Token” may prejudge function

Calling an object a token can smuggle an accounting interpretation into the catalogue before analysis begins.

A safer initial term may be “small geometric clay object.”

14.2 Multifunctionality

Bennison-Chapman’s large-scale study argues that Neolithic geometric clay objects could have served multiple utilitarian functions rather than one universal accounting system [5][6].

14.3 Ritual materiality

Other research proposes that some objects were ritual offerings or symbolically charged miniatures rather than mundane counters [18].

14.4 Regional variation

Communities may have used similar shapes differently.

14.5 Chronological discontinuity

Visual similarity across thousands of years does not guarantee continuous meaning.

14.6 Sample bias

Older excavations often failed to retain or publish small plain objects systematically.

14.7 Classification problems

Token categories can depend on modern researchers’ choices about which differences matter.

14.8 Continued use after writing

Evidence that tokens continued alongside texts undermines a simple replacement model [7][8].

14.9 Multiple administrative pathways

Writing, seals, tags, counters and containers may have developed through interacting but partly independent practices.

14.10 Not all accounting requires writing

A literate institution may still prefer physical tokens for rapid sorting, allocation or local workflows.

14.11 Not all writing emerges from tokens

The token hypothesis concerns one major pathway in Mesopotamia.

It should not be imposed on independently invented writing systems elsewhere.

14.12 Balanced conclusion

The safest synthesis is:

Some shaped clay objects clearly functioned as counters and accounting devices, especially in later administrative contexts. Their relationship to numerical impressions and early Mesopotamian writing is historically significant. Earlier and broader assemblages were likely more varied in function than a single universal accounting code suggests.

15.1 Simple geometric forms

• spheres;
• cones;
• discs;
• cylinders;
• tetrahedra;
• ovoids;
• rods.

15.2 Size variation

Size may encode value, category or practical handling differences.

15.3 Marked forms

Incisions, perforations, punctures or appliqué can modify meaning.

15.4 Natural counters

Pebbles, seeds, shells or bones can function as counters without being specially manufactured.

15.5 Commodity-shaped miniatures

Objects resembling containers, animals or tools may serve as models, symbols, offerings or counters.

15.6 Containerised sets

Groups held in bags, bowls or clay envelopes form bounded accounts.

15.7 Attached tokens and tags

Objects attached to goods or containers can identify quantities, categories or authority.

15.8 Transfer tokens

A token can authorise receipt or claim rather than merely describe stock.

15.9 Ration tokens

Standard objects can represent entitlement to a fixed allocation.

15.10 Calculation counters

Tokens arranged on a board or surface can support arithmetic operations.

16.1 Why enclose tokens?

Enclosure can:

• preserve a set;
• prevent casual alteration;
• bind the record to a transaction;
• support transport;
• create evidence of closure.

16.2 The visibility problem

Once tokens are enclosed, users cannot inspect them without breaking the envelope.

16.3 External impressions

Impressing tokens on the outer surface makes contents visible while preserving closure.

16.4 Seal impressions

Seals can identify the responsible person, household or office.

16.5 Tamper evidence

Breaking the envelope leaves visible evidence that the closure has been opened.

This does not make fraud impossible. It changes the cost and detectability of alteration.

16.6 Multi-party administration

Multiple seals can represent several offices or stages of authorisation.

16.7 The receipt analogy

A sealed envelope may function somewhat like a receipt, contract or shipment record.

The analogy is useful but should not erase differences in legal and institutional context.

16.8 Digital inspection

CT scanning allows internal objects to be studied non-destructively [15][16].

This creates an elegant modern reversal: contemporary imaging technology decodes an information-security device designed five thousand years ago.

17.1 Three dimensions become two

A token can be pressed into wet clay to leave a corresponding impression.

17.2 The exterior becomes readable

The impression exposes information without opening the container.

17.3 The container becomes optional

Once the surface record is accepted, the tokens and hollow enclosure may be unnecessary for some transactions.

17.4 Flat tablets

A clay surface can hold multiple impressions in a compact format.

17.5 Stylus substitution

Tools can reproduce standard impressions more efficiently than repeatedly manufacturing tokens.

17.6 Numerical systems

Different impression shapes and orientations encode quantities and metrological systems.

Early Mesopotamian numbers were not one simple universal sequence. Different commodities and measures could use different numerical systems [10][13].

17.7 Commodity identification

Adding a sign for the counted commodity solves a key ambiguity.

17.8 Names and organisations

More developed writing can identify actors, offices, places and conditions beyond the capacity of most token collections.

17.9 Writing’s expanded bandwidth

Written accounting can record:

• quantity;
• commodity;
• agent;
• recipient;
• date;
• purpose;
• balance;
• instruction;
• exception.

17.10 Persistence of older tools

The surface record expands information capacity, but movable counters remain useful for calculation and workflow.

18.1 External representation

Tokens move information from internal memory into the environment.

18.2 Offloading

Users do not need to hold every quantity simultaneously in working memory.

18.3 Epistemic action

Moving tokens can help users discover or simplify a result rather than merely display one already known.

18.4 Object permanence of the account

The represented state remains available after attention shifts.

18.5 Perceptual grouping

Humans can recognise clusters and categories visually or tactually.

18.6 Error localisation

A mismatch can sometimes be located in a specific token class or group.

18.7 Compression

Higher-value tokens reduce the number of objects required.

18.8 Abstraction

A token treats multiple real items as equivalent units.

18.9 Reification

An obligation becomes a thing that can be held, stored or transferred.

18.10 Cognitive risk

Users may confuse the representation with reality.

A token count can be precise while the underlying classification is foolish.

Tokens enable elementary operations without written arithmetic.

19.1 Increment

Add one token.

19.2 Decrement

Remove one token.

19.3 Aggregation

Combine collections.

19.4 Partition

Divide a collection among recipients or categories.

19.5 Comparison

Place sets side by side.

19.6 Matching

Pair tokens against goods, workers or obligations.

19.7 Substitution

Replace multiple unit tokens with a higher-value token.

19.8 Reclassification

Move a token from one category or state to another.

19.9 Reconciliation

Compare represented state with observed state.

19.10 Closure

Seal or archive the final collection.

19.11 Reopening

Break the seal or retrieve the record for verification.

19.12 Exception handling

Token systems are comparatively weak at recording explanations for unusual events.

That weakness favours written annotations.

| Dimension | Provisional assessment | Explanation | |---|---|---| | Reach | Low to moderate | Physical objects must be carried, but can travel separately from goods | | Latency | Low locally; higher over distance | Manipulation is immediate, transport is physical | | Bandwidth | Low to moderate | Efficient for quantity and categories, weak for narrative and conditions | | Fidelity | High within a stable code and controlled set | Vulnerable to loss, substitution and ambiguous meaning | | Persistence | Moderate to high | Clay survives well; loose collections can disperse | | Replication cost | Low | Simple objects are cheap to manufacture | | Distribution cost | Moderate | Physical transport and custodianship required | | Accessibility | Moderate | No literacy required, but code and procedure must be learned | | Portability | High for small sets | Large inventories may require many tokens | | Interactivity | High | Objects can be rearranged and reconciled | | Searchability | Low | Finding one account may require containers, labels or spatial organisation | | Editability | High before sealing | Simple to update, but edits may leave little history | | Scalability | Moderate | Works well locally; complexity grows rapidly with categories and volume | | Authentication | Low alone; moderate to high with seals | Authority depends on procedures and seal control | | Privacy | Moderate | Enclosures conceal contents but external marks may reveal summaries | | Censorship resistance | Low | Authorities can seize, destroy or redefine records | | Infrastructure dependence | Low to moderate | Requires materials, conventions, storage and custodians | | Energy dependence | Low | Human handling and clay production | | Interpretive burden | Moderate | Users must know token classes, values and account context | | Manipulability | Very high | This is the topic’s signature advantage | | Auditability | Moderate | Supports checking, but often lacks a detailed transaction history | | Tamper evidence | Low loose; higher when sealed | Security is procedural and material | | Representation economy | Low to moderate | Bundling improves efficiency, but physical volume remains a constraint |

21.1 Memory constraint

Users no longer need to remember the full inventory or obligation.

21.2 Presence constraint

The record can survive when the original participants are absent.

21.3 Referent constraint

The goods do not need to be physically gathered for every accounting action.

21.4 Heterogeneity constraint

Different token types can represent different categories.

21.5 Interruption constraint

The account can be stored and resumed.

21.6 Transfer constraint

Information about goods can travel separately from the goods.

21.7 Verification constraint

Another party can inspect or reconcile the represented set.

21.8 Trust constraint

Seals and procedures reduce exclusive dependence on personal memory and verbal assurance.

21.9 Calculation constraint

Physical manipulation supports elementary operations.

21.10 Scale constraint

Token systems manage larger stocks than unaided memory, although eventually their own physical scale becomes limiting.

22.1 Agriculture

Tokens can support records of grain, livestock, seed, storage and distribution.

22.2 Household management

They allow households to track goods, labour and obligations.

22.3 Trade

A token record can support delayed delivery and reconciliation.

22.4 Redistribution

Organisations can collect goods and allocate rations using standard units.

22.5 Taxation and tribute

Material accounts can support claims on producers.

22.6 Labour administration

Tokens may represent work owed, completed or rewarded.

22.7 Storage systems

Inventories become inspectable without opening every container.

22.8 Urbanisation

Large settlements require information systems that exceed face-to-face memory.

22.9 Institutional continuity

Records outlive individual managers.

22.10 Standardisation

Accounting encourages agreed categories and measures.

22.11 Writing

Late token and impression systems contribute to the administrative environment from which Mesopotamian writing emerges.

22.12 Mathematics

Manipulable counters support grouping, equivalence, addition, subtraction and higher-order units.

22.13 Law and contract

Sealed or recognised records can support claims and dispute resolution.

22.14 Governance

The ability to count and allocate resources strengthens organisations.

22.15 Information professions

Specialists emerge to manage records, seals, stores and transactions.

23.1 Household authority

Whoever controls the tokens may control the recognised account.

23.2 Storehouse administration

Central stores require authorised record keepers.

23.3 Temples and palaces

Later Mesopotamian organisations use increasingly sophisticated accounting to organise production and distribution.

23.4 Sealing authority

Control of a seal can determine whose record is recognised.

23.5 Specialist knowledge

Token codes may be accessible without literacy but still restricted to trained users.

23.6 Classification power

Administrators decide which differences matter.

Two animals may become identical units for accounting even when herders know they are not equivalent.

23.7 Visibility power

What is counted becomes administratively visible.

What is not counted can disappear from institutional concern.

23.8 Extraction power

Accounting can support taxation, tribute, labour control and rationing.

23.9 Dispute power

The institution controlling the official record may dominate disagreements.

23.10 Historical visibility

Administrative systems preserve the priorities of organisations more readily than the experiences of those being counted.

24.1 Ambiguous referents

A token may retain quantity while losing the identity of what was counted.

24.2 Code dependency

Meaning disappears when conventions are forgotten.

24.3 Physical loss

Loose objects can be lost or mixed.

24.4 Substitution

A similar object may be inserted fraudulently.

24.5 Volume

Large accounts require many objects unless higher-order values are used.

24.6 Limited narrative bandwidth

Tokens struggle to explain why something happened.

24.7 Weak chronology

A final collection may not reveal the sequence of transactions that produced it.

24.8 Account ambiguity

Without labels, containers or seals, users may not know whose account a set belongs to.

24.9 Category rigidity

Standard classes can conceal important qualitative differences.

24.10 Local incompatibility

Different communities may use different conventions.

24.11 Training requirement

Users need procedural knowledge even without literacy.

24.12 Security trade-off

Enclosure protects contents but reduces immediate inspectability.

24.13 Authority dependence

A token is useful only if relevant parties recognise the system.

24.14 Poor exception handling

Unusual conditions are difficult to encode.

24.15 Archaeological opacity

The objects survive more readily than their codebooks.

25.1 Administrative extraction

Improved accounting can make taxation and tribute more efficient.

25.2 Labour control

Ration and labour systems can discipline workers.

25.3 Surveillance

Organisations gain visibility into household production and obligations.

25.4 Reification

People, animals and labour can be reduced to interchangeable units.

25.5 Category violence

Administrative categories may ignore lived differences and local knowledge.

25.6 Record supremacy

An official token set may override testimony even when the record is wrong.

25.7 Fraud

Tokens can be counterfeited, substituted, withheld or manipulated.

25.8 Gatekeeping

Control of code, seals and storage can centralise power.

25.9 Unequal accountability

Organisations may count what subjects owe while failing to record what organisations owe them.

25.10 Metric substitution

The represented quantity can become a target detached from the underlying purpose.

A ration count can be complete while nutrition remains inadequate.

26.1 Predecessors

• Neural and cognitive memory Neural and cognitive memory;
• Tally marks and notches Tallying and accumulative quantity marks;
• one-to-one correspondence;
• fingers and body counting;
• categorisation;
• exchange and obligation;
• containers and storage.

26.2 Direct extensions

• grouped counters;
• higher-value tokens;
• marked tokens;
• attached tags;
• sealed bags or containers;
• clay envelopes;
• sealings.

26.3 Successors

• numerical impressions;
• numerical tablets;
• Numerical notation Numerical notation;
• Writing systems Writing systems;
• Clay tablets Clay tablets;
• written ledgers;
• receipts;
• abaci;
• coinage;
• inventory tags;
• databases;
• digital state records.

26.4 Complement relationships

Tokens coexist with:

• speech;
• memory;
• seals;
• writing;
• containers;
• calculation boards;
• physical goods.

26.5 Convergence

The emergence of written accounting combines:

• numerical representation;
• commodity classification;
• surface inscription;
• authentication;
• institutional procedure;
• archival storage.

28.1 An early household shapes a counter

A small clay object becomes a persistent representation of something not currently present.

28.2 Tokens enter an envelope

A loose account becomes a fixed and protected set.

28.3 An administrator presses the contents onto the exterior

The record gains readability without sacrificing closure.

28.4 A stylus replaces the counter

The sign no longer needs the original three-dimensional object.

28.5 Numerical tablets appear

Quantitative state moves onto a portable surface.

28.6 Proto-cuneiform adds categories and organisations

Accounting begins acquiring the expressive capacity of writing.

28.7 Tokens continue beside tablets

The supposedly obsolete method refuses to read the evolutionary press release.

28.8 A modern worker scans an inventory tag

The material marker now points into a digital database.

29.1 Tally versus token

| Dimension | Tally | Token | |---|---|---| | Form | Fixed or accumulated mark | Movable object | | Main strength | Simple quantity retention | Sorting, grouping, transfer and state manipulation | | Category handling | Limited unless marks differ | Different shapes or sets can represent categories | | Update | Add or remove marks | Move, add, remove or substitute objects | | Portability | Depends on marked surface | Usually high | | Main risk | Ambiguous referent | Loss, mixing and code ambiguity |

29.2 Token versus numeral

| Dimension | Token | Numeral | |---|---|---| | Representation | Material object | Conventional sign | | Manipulability | Very high | Primarily visual or written | | Compression | Limited unless values differ | High with positional systems | | Transport | Physical volume grows | Compact surface record | | Literacy | Token code required | Numerical literacy required | | Narrative integration | Weak | Can combine with written language |

29.3 Token versus seal

| Dimension | Token | Seal | |---|---|---| | Primary function | Quantity, category or claim | Identity, authority or closure | | Information question | What or how much? | Who authorised or secured it? | | Combination | Tokens plus seals create richer authenticated records | Same |

29.4 Token versus money

| Dimension | Accounting token | Money | |---|---|---| | Domain | Specific account or institution | Broader exchange system | | Acceptance | Procedurally limited | More generalised | | Function | Record or claim | Exchange, account and value storage | | Transferability | Variable | Usually central |

29.5 Token versus written ledger

| Dimension | Token system | Written ledger | |---|---|---| | Interaction | Tactile and spatial | Symbolic and textual | | Bandwidth | Quantities and categories | Quantities, names, dates, conditions and explanations | | Edit history | Often weak | Can preserve sequential entries | | Scale | Moderate | High | | Search | Physical sorting | Indexing and structured records possible | | Exception handling | Weak | Stronger |

29.6 Token versus database object

Both represent entities and states through formal types and permitted operations.

The database adds:

• massive scale;
• rapid search;
• remote access;
• automated validation;
• timestamped histories;
• replication;
• permissions.

The clay token adds:

• direct tangibility;
• low infrastructure dependence;
• visible physical possession;
• intuitive manipulation.

|---|---|---|---| | STO002-C01 | Small geometric clay objects occur in West Asia from the early Neolithic | High | Exact earliest dates and functions vary by site | | STO002-C02 | Some clay objects functioned as counters or accounting tokens | High | Especially secure in later administrative contexts | | STO002-C03 | Every early geometric clay object was an accounting token | Low | Rejected as overgeneralisation | | STO002-C04 | Token systems externalised quantity and category | High | Functional interpretation supported by accounting contexts | | STO002-C05 | Portable tokens enabled sorting and physical manipulation of represented state | High | Follows from material properties and documented counter use | | STO002-C06 | Token shapes had stable commodity meanings across all Neolithic West Asia | Low | Regional and functional variation remain substantial | | STO002-C07 | Late fourth-millennium token envelopes contained grouped counters | High | Archaeologically demonstrated | | STO002-C08 | Exterior impressions sometimes correspond to enclosed tokens | High | Supports content-display interpretation | | STO002-C09 | Seals added an authentication or closure layer | High | Exact institutional meanings vary | | STO002-C10 | Plain tokens contributed to early numerical impressions | Moderate to high | Broad scholarly support, details debated | | STO002-C11 | Complex tokens directly generated the full proto-cuneiform sign inventory | Low to moderate | Influential but contested | | STO002-C12 | Mesopotamian writing emerged within administrative and accounting practices | High | Earliest corpus is overwhelmingly administrative | | STO002-C13 | Writing replaced tokens immediately | Low | Archaeological evidence shows coexistence | | STO002-C14 | Tokens continued in administrative use into the first millennium BCE | Moderate to high | Supported by recent synthesis, context-specific | | STO002-C15 | Clay tokens were the origin of all writing systems | Very low | Mesopotamian pathway cannot be universalised | | STO002-C16 | Token accounting required shared conventions and procedures | High | Meaning is socially assigned | | STO002-C17 | Tokens reduced working-memory demand | High | Strong cognitive inference from external representation | | STO002-C18 | Tokens supported elementary information processing | High | Sorting, grouping and reconciliation are operations on representation | | STO002-C19 | Token envelopes were tamper-proof | Low | Better described as tamper-evident or tamper-resistant | | STO002-C20 | CT imaging can reveal enclosed contents non-destructively | High | Demonstrated research method | | STO002-C21 | Some token-like objects had ritual functions | Moderate | Supported for particular contexts, not universal | | STO002-C22 | Accounting technologies contributed to institutional power | High | Strong historical and systems inference | | STO002-C23 | Tokens were money | Low as a general claim | Some may represent claims or values without general monetary functions | | STO002-C24 | Tokenisation is a continuing information pattern | High conceptually | Avoid direct technological genealogy claims | | STO002-C25 | The topic should be treated as Core | High analytical judgment | It marks manipulable external representation |

1. Which earliest geometric clay objects can be securely identified as accounting devices?
2. How much regional variation existed in token meaning?
3. Were token conventions household-specific, community-wide or interregional?
4. Which objects were multifunctional?
5. How often were tokens used for play, teaching or ritual?
6. Did users operate with exact one-to-one correspondence or bundled values?
7. How were accounts assigned to particular people or transactions before seals and labels?
8. Were perishable containers more important than the surviving clay record suggests?
9. How were tokens transported with goods?
10. Who manufactured tokens?
11. Did token production itself become specialised?
12. How were errors corrected?
13. How were settled obligations cancelled?
14. How did seal authority interact with token quantity?
15. Which token forms correspond most securely to numerical impressions?
16. Which proposed correspondences with proto-cuneiform signs survive current scrutiny?
17. Why did physical tokens persist after writing?
18. Which tasks favoured tokens over tablets?
19. How should “bulla,” “envelope” and “sealing” be standardised in the project?
20. What comparable accounting objects existed outside West Asia without imposing a Mesopotamian template?
21. How can experimental archaeology test handling, storage and reconciliation procedures?
22. Can residue, micro-wear or fingerprint analysis reveal use and production?
23. How did token systems shape concepts of equivalence and value?
24. When does a token become a transferable claim rather than a record?
25. How should this topic connect to the histories of money, contracts and databases without flattening their differences?

32.1 Quantity becomes movable

Show five sheep beside five clay counters.

The sheep remain in the field while the counters move across an accounting surface.

32.2 Tally versus token

Split screen:

• left: five fixed notches;
• right: five movable counters being sorted.

32.3 The first data objects

Display token shapes as typed data objects:

TOKEN TYPE: cone
VALUE: disputed / context-dependent
STATE: loose → grouped → sealed → impressed

32.4 The envelope sequence

Animate:

1. loose tokens;
2. clay wrapping;
3. seal impression;
4. external token impressions;
5. break-open verification.

32.5 Three dimensions become two

Show a cone pressed into wet clay and lifted away, leaving a numerical impression.

32.6 Classic model versus revised model

Classic line

tokens → envelopes → tablets → writing

Revised network

counters ↔ containers ↔ seals ↔ tags ↔ tablets ↔ writing

Include continued token use after writing.

32.7 Confidence ladder

geometric clay object ↓
portable tool or symbol ↓
probable counter ↓
accounting set ↓
authenticated administrative record ↓
precursor to numerical inscription

32.8 Archaeological context map

Compare:

• storehouse;
• domestic floor;
• grave;
• ritual deposit;
• sealed envelope.

The same shape receives different confidence depending on context.

32.9 State machine graphic

AVAILABLE → ALLOCATED → TRANSFERRED → RECEIVED → SETTLED

Move a physical token through each state.

32.10 Token to database

A visual lineage of representation:

clay counter → marked tablet → ledger entry → punch card → database row

Label this as a conceptual comparison, not a direct uninterrupted genealogy.

32.11 Security layers

Show:

• tokens represent quantity;
• envelope preserves set;
• seal identifies authority;
• impression reveals summary.

32.12 Manipulation simulator

An interactive graphic allows users to:

• assign token types;
• add stock;
• allocate rations;
• transfer goods;
• seal an account;
• reconcile the final state.

33.1 Flagship article

Before Spreadsheets, Humans Used Clay Objects to Model the Economy

Argument:

Tokens transformed quantities and obligations into manipulable objects.

33.2 Myth-correction article

Did Clay Tokens Really Invent Writing?

Structure:

• classic hypothesis;
• strongest evidence;
• modern critiques;
• revised conclusion.

33.3 Computer-history bridge

The First Data Objects Were Made of Clay

Compare types, values, states and operations with software objects.

33.4 Accounting history

When a Pebble Became an Account

Trace the movement from one-to-one counters to institutional records.

33.5 Information-security video

Ancient Mesopotamia’s Sealed Data Packets

Use token envelopes carefully as an analogy for bounded, authenticated records.

33.6 Archaeology explainer

Why Archaeologists Cannot Call Every Clay Shape a Token

Focus on context, multifunctionality and confirmation bias.

33.7 Systems-thinking article

The Record Is Not the Reality

Use tokens to explain classification, metrics and administrative abstraction.

33.8 Governance article

The Technology That Made Taxation Scalable

Balance coordination benefits with extraction and control.

33.9 Experimental video

Create a small token system to track:

• grain;
• goats;
• labour;
• rations.

Then test:

• memory;
• error rates;
• update speed;
• security;
• exception handling.

33.10 Short-form hooks

• “The first spreadsheet did not have cells. It had clay balls.”
• “This cone may have been data.”
• “Before writing, people could physically hold an account.”
• “A seal did not protect the goods. It protected the story about the goods.”
• “Writing did not replace tokens. Bureaucracy kept both.”
• “Not every ancient clay pebble was somebody’s invoice.”

Option A: The First Data Object

• Giant clay token in the foreground;
• ghosted sheep and grain behind it;
• text: DATA BEFORE WRITING.

Option B: Clay Spreadsheet

• Rows of tokens arranged like spreadsheet cells;
• a hand moving one token;
• text: THE FIRST SPREADSHEET?

Option C: Sealed Account

• Cracked clay envelope with tokens visible inside;
• seal impression glowing;
• text: ANCIENT DATA SECURITY.

Option D: Object to Sign

• clay cone on the left;
• its impression on the right;
• text: HOW WRITING BEGAN.

Use carefully because the relationship is important but not the whole story.

Option E: Not Just Counters

• several token shapes;
• one highlighted with a question mark;
• text: WHAT WERE THESE?

Replace the current entry with:

|---|---|---|---|---|---|---|---|---|---|---| | Tokens and accounting objects | Tokens and accounting objects | Material accounting system and manipulable representation | Storage and persistence | Processing; encoding; governance; authentication | Small geometric clay objects from the tenth or ninth millennium BCE in parts of West Asia; secure complex accounting use especially visible by the fourth millennium BCE | Tracks quantities, categories, stocks, transfers and obligations outside memory and apart from the goods themselves | Counting; tallying; categorisation; exchange; storage | Numerical impressions; tablets; writing; ledgers; abaci | Core | Researched |

Changes:

1. move the earliest range earlier than the current eighth-millennium wording;
2. upgrade priority from Supporting to Core;
3. change topic type from Medium to Material accounting system and manipulable representation;
4. add encoding and authentication as secondary categories;
5. avoid implying that all early geometric clay objects were confirmed accounting tokens;
6. record coexistence with writing rather than simple replacement.

37.1 Add manipulability

The evaluation framework should ask:

Can users operate on the representation itself by moving, sorting, grouping or transforming it?

37.2 Add state representation

Some media describe events. Others model current state.

The project should distinguish:

• event record;
• stock record;
• flow record;
• obligation record;
• entitlement record.

37.3 Add procedural dependence

A token’s meaning depends heavily on workflow.

Add a field for:

Required operating procedure and institutional context.

37.4 Add object-surface transition

The map needs an explicit relationship type for information moving from:

• three-dimensional object;
• to impression;
• to drawn or written sign.

37.5 Add coexistence after succession

Successor topics should record whether older methods continue as workflow tools.

37.6 Add multifunctionality confidence

Archaeological objects may have different functions across contexts.

Add:

• single-function confidence;
• multifunctionality evidence;
• context dependence.

37.7 Add auditability

Distinguish:

• inspectable state;
• tamper evidence;
• chronological audit trail;
• independent reconciliation.

37.8 Add authentication stack

Records often combine several layers:

• content representation;
• container;
• closure;
• seal;
• custodian;
• procedure.

37.9 Refine “storage”

This topic shows that a storage medium can also be a processor.

The taxonomy should allow one topic to store information through an arrangement that is actively manipulated.

37.10 Add representational substitution risk

The project should track the danger that administrators optimise the representation while neglecting reality.

Foundational token hypothesis

[1] Schmandt-Besserat, Denise. “Tokens: Their Significance for the Origin of Counting and Writing.” University of Texas at Austin.
https://sites.utexas.edu/dsb/tokens/tokens/

[2] Schmandt-Besserat, Denise. “From Accounting to Writing.” University of Texas at Austin.
https://sites.utexas.edu/dsb/tokens/from-accounting-to-writing/

[3] Schmandt-Besserat, Denise. Before Writing, Volume I: From Counting to Cuneiform. University of Texas Press, 1992.

[4] Schmandt-Besserat, Denise. “The Envelopes That Bear the First Writing.” Technology and Culture 21, no. 3, 1980.

Critiques and revised interpretations

[5] Bennison-Chapman, Lucy E. “Reconsidering ‘Tokens’: The Neolithic Origins of Accounting or Multifunctional, Utilitarian Tools?” Cambridge Archaeological Journal 29, no. 2, 2019.
https://www.cambridge.org/core/journals/cambridge-archaeological-journal/article/reconsidering-tokens-the-neolithic-origins-of-accounting-or-multifunctional-utilitarian-tools/7E6C04CB040AD8AA0EA84B94D4D275C4

[6] Bennison-Chapman, Lucy E. “The Role and Function of ‘Tokens’ and Sealing Practices in the Neolithic of the Near East.” Doctoral thesis, University of Liverpool, 2014.
https://livrepository.liverpool.ac.uk/2008477/

[7] Bennison-Chapman, Lucy E. “Tools of the Trade: Accounting Tokens as an Alternative to Text in the Cuneiform World.” Journal of Near Eastern Studies 82, no. 2, 2023.
https://www.journals.uchicago.edu/doi/10.1086/727776

[8] University of Cambridge. “Prehistoric ‘Book-Keeping’ Continued Long After Invention of Writing.” 2014.
https://www.cam.ac.uk/research/news/prehistoric-book-keeping-continued-long-after-invention-of-writing

[9] Zimansky, Paul. Review of Schmandt-Besserat, Before Writing. Journal of Field Archaeology 20, no. 4, 1993.
https://urkesh.org/attach/Zimansky1993.pdf

[10] Englund, Robert K. “Account-Keeping and the Emergence of Writing.” In The Oxford Handbook of Cuneiform Culture, 2011.
https://cdli.mpiwg-berlin.mpg.de/files-up/publications/englund2011a.pdf

Early writing and administrative systems

[11] Nissen, Hans J., Peter Damerow and Robert K. Englund. Archaic Bookkeeping: Early Writing and Techniques of Economic Administration in the Ancient Near East. University of Chicago Press, 1993.
https://cdli.earth/publications/1657148

[12] Woods, Christopher. “The Origins of Writing in Mesopotamia.” Institute for the Study of Ancient Cultures.
https://isac.uchicago.edu/sites/default/files/uploads/shared/docs/nn207.pdf

[13] Englund, Robert K. “The State of Decipherment of Proto-Elamite.”
https://cdli.earth/files-up/publications/englund2004c.pdf

[14] Metropolitan Museum of Art. “The Origins of Writing.”
https://www.metmuseum.org/essays/the-origins-of-writing

Envelopes, seals and imaging

[15] Woods, Christopher. “Writing in Early Mesopotamia: The Token Envelope Project.” Institute for the Study of Ancient Cultures.
https://isac.uchicago.edu/sites/default/files/uploads/shared/docs/nn215.pdf

[16] Woods, Christopher. Research report on scanning proto-literate accounting devices. Institute for the Study of Ancient Cultures.
https://isac.uchicago.edu/sites/default/files/uploads/shared/docs/ar/11-20/11-12/11_12_Woods.pdf

[17] Bennison-Chapman, Lucy E., and Lori D. Hager. “Tracking the Division of Labour Through Handprints: Applying Reflectance Transformation Imaging to Clay ‘Tokens’ in Neolithic West Asia.” Journal of Archaeological Science 99, 2018.

Alternative material interpretations

[18] Rowan, Yorke M. “Not Just Counters: Clay Tokens and Ritual Materiality in the Ancient Near East.” Journal of Archaeological Method and Theory 28, 2021.
https://link.springer.com/article/10.1007/s10816-020-09457-8

Regional and recent synthesis

[19] Matthews, Roger. “Counting, Sealing, Writing in Iran, 10000–300 BC.” Journal of Ancient Near Eastern History, 2025.
https://www.degruyter.com/document/doi/10.1515/janeh-2024-0027/html

[20] Met Museum. “Mesopotamia, 8000–2000 B.C.” Heilbrunn Timeline of Art History.
https://www.metmuseum.org/toah/ht/02/wam.html

[21] Bennison-Chapman, Lucy E. Terminological discussion of tokens, sealings and envelopes in recent administrative archaeology. See [7] and Matthews [19].

Comparative and conceptual context

[22] Wilding, David. “A Conversation with Denise Schmandt-Besserat and Bill Maurer.” Exchanges, 2017.
https://exchanges.warwick.ac.uk/index.php/exchanges/article/view/196

[23] Mattessich, Richard. “Archaeology of Accounting and Schmandt-Besserat’s Contribution.” Accounting, Business & Financial History 4, no. 1, 1994.

[24] Metropolitan Museum of Art. Proto-cuneiform administrative objects and collection entries.
https://www.metmuseum.org/art/collection/search/321992

[25] Cuneiform Digital Library Initiative. Publications and corpora on proto-cuneiform accounting.
https://cdli.earth/

39.1 The breakthrough is manipulable representation

Tallies preserve increments. Tokens make increments movable.

That mobility changes the information system.

A user can act on the model rather than repeatedly acting on the world.

39.2 The object is a data structure

A token has:

• a type;
• a value;
• a referent;
• a state;
• an owner or account context;
• permitted operations.

The clay is the substrate. The social procedure is the software.

39.3 Accounting begins before writing

Humans could record, classify and manipulate economic information without encoding full spoken language.

This is a major reminder:

Writing is not the beginning of information processing.

39.4 Administration drives abstraction

When organisations manage grain, livestock, labour and rations, they need standard units and categories.

The system turns messy reality into administratively manageable equivalences.

39.5 Security requires layers

A token set alone can be altered.

Containers, closures, seals, custodians and procedures add trust.

The result resembles a security stack rather than one magical anti-fraud object.

39.6 The classic origins story is partly right and too tidy

Tokens, impressions, numerical tablets and proto-cuneiform are meaningfully connected.

Yet not every early clay object was an account, not every sign came from a token, and writing did not sweep older tools away.

39.7 The representational gain creates political power

Counting makes coordination possible.

It also makes extraction possible.

An institution that can represent stocks and obligations can distribute resources, collect tribute and enforce claims at larger scale.

39.8 The record can dominate reality

Once a token system defines the official state, a discrepancy becomes somebody’s problem.

That may improve accountability. It may also allow the record keeper to define truth.

39.9 Successors solve the bandwidth problem

Tokens are excellent for quantities and categories but poor at names, explanations, exceptions and chronology.

Numerical notation and writing compress richer information onto surfaces.

39.10 The logic remains alive

Modern systems still convert real-world entities into manipulable representations.

A database row, inventory tag or authentication token follows the same broad intellectual move:

define a unit, create a representation, assign rules, update state and trust the result enough to act on it.

39.11 Final topic statement

Tokens and accounting objects transformed information from a mark that could be inspected into a model that could be manipulated. By turning goods, quantities and obligations into portable objects, they enabled sorting, transfer, reconciliation and institutional memory. Their later interaction with seals, impressions and tablets helped produce written accounting, but their deeper contribution was older and broader: they taught humans to operate on representations of reality rather than reality alone.