Tally marks and notches

Tally marks and notches examines one of the smallest-looking inventions in the Information Transmission Evolution Map: the deliberate production of repeated marks, cuts or notches to retain quantity, recurrence, sequence or obligation outside unaided biological memory.

The topic’s central claim is:

A tally converts repeated events or objects into repeated material marks, allowing quantity to accumulate outside the mind.

This sounds modest. It is not.

A person can look at three animals and recognise a small quantity without counting. Larger or interrupted sets are harder. Objects move. Days pass. Debts outlive conversations. Participants disagree. Memory quietly edits the meeting minutes. A tally changes the task by pairing each relevant object, event or unit with a durable mark.

The simplest version follows one rule:

One counted unit produces one material token.

The token may be:

• a notch cut into bone or wood;
• a line scratched into stone;
• a mark painted on a wall or body;
• a pebble moved into a pile;
• a knot tied in a cord;
• a bead added to a string;
• a stroke written on paper;
• a click stored by a mechanical counter;
• a bit incremented in software.

The material form differs. The underlying operation is accumulation by one-to-one correspondence.

Tallies solve several information problems at once:

• they externalise quantity;
• they preserve an interrupted count;
• they make recurrence visible;
• they support comparison between sets;
• they reduce dependence on number words;
• they permit another person to inspect the record;
• they can record debts, deliveries and obligations;
• they create a trail from event to mark;
• they distribute cognitive effort across time.

A tally is not yet a compact numeral system. Twelve unary marks require twelve marks. A numeral such as 12 represents the same quantity with two conventional signs whose values depend on positional and decimal rules. Tallies are comparatively easy to invent and inspect, but inefficient at scale. Grouping marks, assigning different values to different marks, and replacing bundles of units with higher-order symbols prepare the way for numerical notation, tokens, abaci, ledgers and calculation.

The archaeological record requires unusual caution.

Prehistoric bones, antler pieces and stones with sequential incisions are frequently described as “tallies.” Some may indeed have stored numerical information. Research has argued that sequential markings were used as artificial memory systems from at least the beginning of the European Upper Palaeolithic, around 42,000 years ago [1]. Notched organic artefacts from Border Cave in southern Africa date to roughly 44,000–42,000 years ago, and one incised baboon fibula has been proposed as a counting device [6][7].

Yet a row of regular cuts does not preserve its codebook.

The marks may have been:

• decorative;
• ownership signs;
• records of completed actions;
• calendrical marks;
• ritual participation marks;
• measurements;
• mnemonic prompts;
• numerical tallies;
• stages in manufacture;
• traces of use rather than intentional notation.

Archaeologists therefore cannot move directly from “regular notches” to “calendar,” “arithmetic” or “advanced mathematics.” The stronger the interpretation, the more converging evidence it requires.

The Ishango artefact from present-day Democratic Republic of Congo illustrates the problem. It contains grouped notches arranged in columns and has inspired interpretations involving counting, doubling, prime numbers, base twelve, calendars and mathematical instruction. The grouping is real. The specific mathematical meanings remain debated. It is safer to describe it as a structured notched artefact with probable notational significance than as a prehistoric textbook whose answer key has somehow survived twenty millennia [15].

Historically documented tallies show what the basic technology could become.

Split tally sticks used in Europe recorded debts or payments by cutting value marks into a stick and splitting it lengthwise. Each party retained one half. Because the grain and original notches matched only when reunited, later alteration became detectable. English Exchequer tallies developed into durable administrative and financial instruments. A Bank of England tally from 1694 records one of the institution’s earliest loans to government [18][19]. Some tallies could be transferred, allowing a record of debt to circulate as an asset.

This produces an important transition:

A mark that first helps one person remember a quantity can become a socially recognised claim enforceable between organisations.

Tallying therefore belongs to more than the history of mathematics. It also belongs to the history of:

• memory;
• administration;
• accounting;
• law;
• trust;
• property;
• taxation;
• scientific observation;
• sport;
• voting;
• computing.

The research notes recommends retaining Tally marks and notches as a Core topic and refining its analytical name to:

Tallying and Accumulative Quantity Marks

The public-facing title Tally Marks and Notches remains clear and useful.

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; analytically narrower than the full phenomenon | | Recommended analytical name | Tallying and accumulative quantity marks | | Primary contribution | Externalises quantity, recurrence and sequence through one-to-one material correspondence | | Main successor advantage | Grouped tokens and numerical notation represent larger quantities more compactly and support calculation | | Main caution | Regular notches are not automatically numerical, and numerical intent does not reveal the item or event counted | | Taxonomy pressure | The project needs a clearer distinction between encoding, external memory and elementary information processing |

| Field | Provisional value | |---|---| Tally marks and notches | | Register name | Tally marks and notches | | Recommended analytical name | Tallying and accumulative quantity marks | | Topic type | Quantitative external-memory method and elementary notational system | | Primary category | Encoding and expression | | Secondary categories | Storage; processing; governance; authentication; cultural memory | | Approximate emergence | Sequential artificial markings securely attested by the Upper Palaeolithic; candidate earlier examples remain contested | | Main problem addressed | Reliance on unaided memory for exact quantity, recurrence and sequence | | Key predecessors | Biological memory; one-to-one matching; fingers and body counting; conventional marks | | Key successors | Accounting tokens; numerical notation; knot-record systems; written ledgers; counters; abaci; digital counters | | Current status | Researched and provisionally synthesised |

For this project, tallying is defined as:

The deliberate creation, movement or grouping of material marks or objects so that each relevant unit, event or increment is represented by a corresponding external token.

This definition has seven parts.

3.1 Deliberate

The mark must be intentionally connected to a recording act.

Accidental scratches, butchery marks, tooth marks and manufacturing damage are not tallies merely because they form a row.

3.2 Creation, movement or grouping

Tallies are not limited to incisions.

A tally operation may involve:

• adding marks;
• cutting notches;
• moving stones;
• threading beads;
• tying knots;
• breaking pieces away;
• transferring objects between containers;
• raising fingers;
• clicking a counter.

3.3 Material

The tally exists outside the immediate neural state of the counter.

Even body-based tallies use a perceivable configuration of fingers or body positions as temporary external structure.

3.4 Correspondence

Each counted unit is paired with a tally token, or a mark is assigned a conventional multiple value.

3.5 Relevant unit

A tally does not define the ontology of what is counted.

The unit may be:

• one animal;
• one day;
• one basket;
• one person;
• one completed task;
• one debt instalment;
• one vote;
• one goal;
• one prayer;
• one lunar phase;
• one shipment.

The codebook connecting mark and referent may remain entirely external to the artefact.

3.6 External token

The token provides a persistent stand-in for the counted unit.

3.7 Accumulation

The record grows as relevant units occur.

This incremental property distinguishes tallying from simply writing a pre-existing numeral.

4.1 It externalises exact recurrence

Biological memory can retain an impression that something happened often.

A tally can preserve how many times it happened.

4.2 It allows interrupted counting

A count can pause without restarting from the beginning.

4.3 It reduces working-memory demand

The counter does not need to keep the entire running total active internally.

4.4 It creates inspectability

Another person can see, recount or challenge the record.

4.5 It permits comparison

Two rows or piles can be compared visually or by matching their tokens.

4.6 It separates quantity from the objects counted

A herd may move away. A shipment may be consumed. The marks remain.

4.7 It creates administrative memory

Tallies can record:

• delivery;
• debt;
• attendance;
• tax;
• labour;
• inventory;
• payment.

4.8 It supports verification

Grouped, witnessed or split tallies can make alteration visible.

4.9 It prepares the ground for arithmetic

Accumulation, comparison, grouping and substitution create material operations from which addition, subtraction and place-value systems can develop.

4.10 It survives into modern interfaces

The four vertical strokes crossed by a fifth remain recognisable because tallying is cognitively transparent.

5.1 Quantity

How much or how many of something is present.

5.2 Numerosity

The number of elements in a set.

5.3 Number

An abstract concept used to characterise quantity, order or relation.

5.4 Counting

A procedure that maps an ordered sequence of count labels onto items, normally using one-to-one correspondence and interpreting the final label as the set’s cardinality [10][11].

5.5 One-to-one correspondence

The pairing of one count token with one item or event.

5.6 Cardinality

The total number of elements in a set.

5.7 Ordinality

Position within an ordered sequence.

5.8 Tally

A material device or record that accumulates tokens corresponding to units or events.

5.9 Tally mark

An individual visible stroke, notch or other repeated token within a tally.

5.10 Notch

A cut made into the edge or surface of an object.

A notch may be numerical, decorative, functional or accidental.

5.11 Unary notation

A notation in which quantity is represented through repetition of a single unit sign.

For example:

|||| represents four units.

5.12 Grouped tally

A tally in which marks are organised into recurrent bundles, such as groups of five or ten.

5.13 Accumulative device

A material system whose state changes by adding or removing units.

5.14 Numerical notation

A conventional graphic system representing numbers, often more compactly than unary tallies.

5.15 Counter

A manipulable object or mechanism used to track increments.

5.16 Token

A discrete object that stands for a unit, quantity or category.

5.17 Score

A record of points, achievements or obligations. The word may refer either to the total or the recording system.

5.18 Split tally

A notched stick divided lengthwise so that matching halves can authenticate a transaction.

5.19 Stock and foil

Traditional terms for the two halves of a split tally. Terminology varied across periods and organisations.

5.20 Artificial memory system

A material arrangement used to store information beyond unaided biological memory [5].

6.1 Boundary with Neural and cognitive memory: Neural and Cognitive Memory

Memory retains quantity internally.

A tally relocates part of that burden into material form.

6.2 Boundary with Pictograms and ideograms: Conventional Visual Signs

A conventional sign may stand for a category such as “sheep.”

A tally represents repeated units such as twenty-seven sheep.

The two can combine:

sheep symbol + twenty-seven tally units

6.3 Boundary with Tokens and accounting objects: Tokens and Accounting Objects

A tally usually accumulates repeated marks on or in one device.

Accounting tokens use discrete manipulable objects that may also encode categories or different values.

6.4 Boundary with Numerical notation: Numerical Notation

Tallies represent quantity through material repetition or accumulation.

Numerical notation uses conventional signs whose form and position may represent quantities far larger than the number of visible marks.

6.5 Boundary with Khipu and other knot-record systems: Knot-Record Systems

A single row of knots may function as a tally.

Complex knot systems can additionally encode hierarchy, category, place value, colour and administrative structure.

6.6 Boundary with calendars

A tally may count days, lunar phases or recurring seasons.

It becomes a calendar only when the marks are organised within a system that maps them to temporal cycles.

6.7 Boundary with measurement

Tallies count discrete units.

Measurement compares a continuous magnitude against a standard unit.

A notched stick may function as either, depending on whether spacing or number of cuts carries the information.

6.8 Boundary with decoration

Repeated marks can be visually regular without being numerical.

Intent cannot be inferred from neatness alone.

6.9 Boundary with writing

A tally can preserve quantity without representing language.

Writing systematically represents linguistic expressions.

6.10 Boundary with arithmetic

A tally supports elementary operations by material manipulation.

It does not by itself prove that its maker possessed formal arithmetic rules.

7.1 Participant structure

• one person to future self;
• one person to another;
• group to group;
• institution to institution;
• authority to subject;
• distributed public record.

7.2 Time relationship

Primarily asynchronous.

7.3 Spatial relationship

Local at creation but portable when the marked object can move.

7.4 Persistence

Ranges from seconds for finger configurations to centuries for preserved wood, bone, clay, stone or paper.

7.5 Direction

Usually one-way as a record, but often embedded in reciprocal transactions.

7.6 Interactivity

The user can:

• add;
• remove;
• regroup;
• compare;
• reconcile;
• split;
• verify.

7.7 Addressability

A tally may be private, shared by contracting parties, maintained by an institution or displayed publicly.

8.1 Source event

An item arrives, a day passes, a task is completed or an obligation is created.

8.2 Unit definition

The recorder decides what counts as one unit.

8.3 Encoding rule

The system maps the unit to:

• one mark;
• one object;
• one knot;
• one click;
• a weighted mark of conventional value.

8.4 Production

The recorder creates or moves the token.

8.5 Signal

The current arrangement of marks or objects.

8.6 Channel

The material surface or container.

8.7 Storage

The tally remains available after the event.

8.8 Receiver

The recorder, another participant, auditor, official or later researcher.

8.9 Recognition

The receiver identifies intentional marks and their grouping.

8.10 Decoding

The receiver applies the relevant unit and value rules.

8.11 Interpretation

The receiver decides what the total means in context.

8.12 Response

The receiver may:

• deliver goods;
• settle debt;
• challenge the count;
• continue counting;
• transfer the claim;
• archive the device.

8.13 Feedback

Reconciliation occurs by recounting or matching records.

8.14 Noise

Physical noise

• breakage;
• wear;
• decay;
• overlapping cuts;
• lost fragments.

Procedural noise

• skipped units;
• double-counting;
• inconsistent marking;
• failure to define the unit.

Semantic noise

• lost codebook;
• unknown referent;
• uncertain direction;
• ambiguous grouping.

Adversarial noise

• adding marks;
• erasing marks;
• substituting the object;
• falsifying labels;
• controlling who may inspect the record.

9.1 Approximate quantity perception

Humans and many other animals can discriminate quantities approximately without symbolic counting.

Accuracy decreases as quantities become closer in ratio.

This system is useful for rough judgements such as:

• more versus fewer;
• larger versus smaller groups;
• abundant versus scarce.

It does not reliably preserve exact large totals.

9.2 Small-number recognition

Humans can rapidly recognise very small quantities without serial counting, a process often called subitising.

9.3 Exact counting

Exact counting requires more than noticing plurality.

Influential accounts identify principles including:

• one-to-one correspondence;
• stable order;
• cardinality;
• abstraction;
• order irrelevance [10][11].

9.4 One-to-one matching before numerals

Two sets can be compared by pairing their elements even without naming the total.

One animal can be matched with one pebble.

At the end:

• leftover animals indicate more animals;
• leftover pebbles indicate more pebbles;
• no leftovers indicate equal sets.

9.5 Material accumulation

A tally extends one-to-one correspondence through time.

The animal and pebble need not remain together. The pebble survives as the animal’s informational proxy.

9.6 Stable sequence

Marks created in a row preserve temporal order as well as quantity, unless the recorder deliberately treats position as irrelevant.

9.7 Cardinal interpretation

The total number of marks can represent the total number of events or items.

However, cardinality is not visible to an archaeologist unless numerical intent can be established.

9.8 Grouping

Grouping reduces recounting cost.

Five-mark bundles, ten-mark bundles and value-coded notches make larger totals easier to scan.

9.9 Material cognition

Tallying distributes cognitive work between brain, body and artefact.

The material arrangement does not merely display a completed mental calculation. It participates in the calculation by holding intermediate state [2][3].

9.10 Error visibility

Material traces can expose disagreement that memory alone would conceal.

The tally does not guarantee truth. It makes some forms of inconsistency inspectable.

10.1 Unit mark

The smallest repeated element.

10.2 Medium

Possible media include:

• bone;
• antler;
• wood;
• stone;
• clay;
• skin;
• paper;
• cord;
• beads;
• metal;
• digital state.

10.3 Surface orientation

Marks may run:

• linearly;
• around an edge;
• in columns;
• in clusters;
• around a circumference.

10.4 Direction

The order of accumulation may depend on a starting point and direction.

10.5 Spacing

Uniform spacing can aid recognition but is not required.

Spacing may distinguish:

• groups;
• sessions;
• categories;
• makers;
• temporal intervals.

10.6 Grouping delimiter

A longer cut, crossed mark, gap or change in orientation may close a bundle.

10.7 Value differentiation

Different widths, depths, lengths or positions may assign different values.

10.8 Category label

A tally often requires an associated label identifying what was counted.

The label may be:

• remembered;
• spoken;
• written;
• pictorial;
• tied to the object’s location;
• embodied in ownership.

10.9 Verification feature

Split tallies, seals, witnesses or matching duplicates can make tampering detectable.

10.10 Portability

Portable tallies detach the record from the place of production.

10.11 Closure

Some systems mark when a count is complete or a debt settled.

10.12 Reusability

Tallies may be permanent, erasable, cancellable or resettable.

11.1 Cardinal quantity

How many units occurred.

11.2 Recurrence

How many times an event happened.

11.3 Sequence

Which event came first, second or later.

11.4 Duration by repeated interval

Days, nights, moons or work periods.

11.5 Progress

Tasks completed towards a target.

11.6 Inventory

Goods received, stored or distributed.

11.7 Debt and obligation

Amounts owed, paid or outstanding.

11.8 Attendance

Persons present or absent.

11.9 Voting

One mark per choice or participant.

11.10 Performance

Goals, points, victories, losses or repetitions.

11.11 Ritual repetition

Prayers, vows, offerings or completed acts.

11.12 Observation

Occurrences of animals, weather events, illnesses or experimental outcomes.

11.13 Calendar information

Possible only when marks are linked to temporal units and cyclical rules.

11.14 Category and quantity together

Multiple rows or labelled devices can distinguish:

• cattle;
• grain;
• labour;
• tax;
• different debtors.

12.1 No single invention event

Tallying is materially simple enough to have been invented repeatedly.

Its archaeological emergence is therefore not expected to resemble the appearance of one patented machine.

12.2 Earlier candidate markings

Humans and earlier hominins produced incised objects long before secure writing.

Some early marks may have been mnemonic or numerical, but many cannot be confidently classified.

12.3 Upper Palaeolithic artificial memory systems

D’Errico and colleagues argue that sequential markings were used to store and retrieve numerical information from the beginning of the European Upper Palaeolithic, around 42,000 years ago, with greater mark numbers and coding complexity appearing later [1].

This is evidence for a broad practice, not proof that every marked object was numerical.

12.4 Border Cave notched artefacts

Border Cave in the Lebombo Mountains preserves notched organic artefacts from the early Later Stone Age, around 44,000–42,000 years ago [6][7].

An incised baboon fibula is frequently called the Lebombo bone and has been interpreted as a number-keeping device.

The safest formulation is:

It is an early deliberately notched artefact plausibly used as an external memory or counting device, but its precise referent and operation are unknown.

Claims that its twenty-nine marks definitely record a menstrual or lunar cycle are speculative.

12.5 European Upper Palaeolithic marked objects

Portable bones, antlers, stones and artworks contain rows or clusters of cuts and dots.

Microscopic analysis may reveal:

• direction of tool movement;
• sequence of production;
• different tools or sessions;
• intentional grouping;
• later additions.

These details can support an interpretation as cumulative notation, but they rarely reveal the subject counted.

12.6 Upper Palaeolithic calendar proposals

Bacon and colleagues proposed that marks associated with animal images encoded lunar months and seasonal animal behaviour [13].

Later critiques questioned the assumptions, dataset construction and ability of the statistical correlations to establish the proposed meanings [14].

The case demonstrates a general rule:

Detecting patterned association is not the same as recovering the code.

12.7 The Ishango artefact

The Ishango artefact was excavated in 1950 near Lake Edward in present-day Democratic Republic of Congo. Its grouped notches occur in multiple columns [15].

Proposed interpretations include:

• simple tallying;
• arithmetic relationships;
• doubling;
• base systems;
• a lunar calendar;
• mathematical instruction;
• structured but non-numerical symbolism.

The object clearly contains deliberate grouping. More elaborate mathematical readings remain hypotheses.

12.8 Neolithic and early administrative contexts

As sedentism, storage, exchange and institutional administration expanded, quantity records became more closely associated with:

• goods;
• labour;
• taxation;
• land;
• ritual obligations;
• redistribution.

Tallies interacted with accounting tokens, seals and early writing rather than vanishing immediately.

12.9 Historical persistence

Tally sticks continued to be used long after numerals and writing existed because they were:

• cheap;
• physically inspectable;
• usable by people with limited literacy;
• difficult to alter when split or witnessed;
• suited to local transactions.

12.10 Digital persistence

Modern systems still tally:

• database rows;
• web visits;
• votes;
• sensor events;
• inventory units;
• sports scores;
• machine cycles.

The material notch has become a state change in memory, but the accumulative logic survives.

13.1 The archaeological problem

A mark survives more readily than the instruction that explains it.

13.2 Regularity is insufficient

Regular marks may be decorative or procedural.

13.3 Numerical intent is graded

Evidence strengthens when marks show:

• serial production;
• deliberate accumulation;
• recurrent grouping;
• corrections;
• repeated conventions across objects;
• relation to countable categories;
• ethnographic or historical parallels;
• wear consistent with repeated handling.

13.4 Taphonomic exclusion

Researchers must distinguish human incisions from:

• carnivore tooth marks;
• trampling;
• root etching;
• butchery;
• manufacturing traces;
• excavation damage.

13.5 Operational sequence

Microscopy can help determine whether marks were made:

• in one sitting;
• over multiple episodes;
• with one tool;
• with several edges;
• before or after other modifications.

13.6 The referent problem

Even a secure tally does not tell us whether it counted:

• days;
• animals;
• people;
• kills;
• exchanges;
• rituals;
• stories;
• tool production.

13.7 The calendar temptation

Rows of twenty-eight, twenty-nine or thirty marks attract lunar interpretations.

Human beings also count many non-lunar things in those quantities.

13.8 The advanced-mathematics temptation

Patterns can be retrospectively fitted to many numerical relations.

A strong interpretation should explain why the proposed structure was useful to the original users and outperform alternative models.

13.9 Ethnographic analogy

Documented tally practices can generate hypotheses, but similarity does not prove direct continuity across tens of thousands of years.

13.10 Confidence ladder

Recommended project labels:

1. deliberately marked object;
2. sequential or grouped notation;
3. probable artificial memory device;
4. probable tally;
5. probable numerical system;
6. specific referent identified;
7. specific mathematical operation identified.

Most famous prehistoric examples do not securely reach the final levels.

14.1 External state

The current total exists materially.

14.2 Incremental update

Each event changes the state by one or another agreed value.

14.3 Persistence

The total survives interruption and distraction.

14.4 Distributed access

Multiple people can inspect the same record.

14.5 Cognitive offloading

The counter no longer rehearses the total continuously.

14.6 Reconstruction

The marks permit later recovery of a total.

14.7 Error localisation

Disagreement can focus on particular marks or recording episodes.

14.8 Accumulation without number words

A person may maintain one token per object without naming the final quantity.

14.9 Material successor function

Adding one mark creates the next tally state.

14.10 Limit

The device stores structure, not explanation.

Without labels or shared context, the record can become an immaculate answer to a vanished question.

15.1 Unary growth

Pure tallying grows linearly with quantity.

One hundred units require one hundred unit marks.

15.2 Perceptual grouping

Humans group marks to reduce visual search.

15.3 Five-bar tally

Four vertical strokes crossed by a fifth creates a compact visual bundle.

15.4 Decimal grouping

Groups of ten align with finger counting and later decimal systems.

15.5 Weighted marks

A longer or differently placed notch can represent a larger value.

15.6 Replacement

Several low-value tokens may be exchanged for one higher-value token.

15.7 Hierarchy

Rows, columns and positions can separate units, bundles and categories.

15.8 Conciseness

Numerals drastically reduce the number of marks required.

15.9 Manipulability trade-off

Loose counters and tokens are easy to rearrange during calculation.

Written numerals are compact but fixed, requiring learned transformation rules.

15.10 Conceptual transition

The path is not simply:

scratches become numbers

It is better understood as interaction among:

• one-to-one matching;
• accumulation;
• grouping;
• language;
• fingers and bodies;
• manipulable tokens;
• written notation;
• institutional need.

16.1 From memory aid to obligation record

A tally can represent not merely what happened, but what another person owes.

16.2 Split construction

Value marks are cut before the stick is split lengthwise.

16.3 Matching halves

The grain, shape and original cuts align when the halves are reunited.

16.4 Tamper evidence

A later notch added to only one half will not match the other [19].

16.5 Distributed trust

Neither party needs to surrender the entire record to the other.

16.6 Administrative use

Exchequers and other organisations used tallies to authorise, record and reconcile payments [17][20].

16.7 Bank of England example

A hazelwood tally records one of the Bank of England’s first loans to the government in 1694 [18].

16.8 Transferability

Some debt tallies could be transferred to another holder, allowing the claim represented by the object to circulate [19][21].

16.9 Literacy independence

The physical match and value notches could support verification even where participants did not share advanced literacy.

16.10 Limits

The system still depended on:

• recognised authorities;
• agreed notch values;
• secure custody;
• legal enforcement;
• identification of parties;
• protection against total replacement or theft.

17.1 Low invention cost

A tally can be created with available materials and minimal specialised equipment.

17.2 Low learning cost

One mark per unit is transparent.

17.3 Exact accumulation

The system can retain exact totals if recording is reliable.

17.4 Incrementality

The record updates one event at a time.

17.5 Auditability

Marks can be recounted.

17.6 Portability

Small tally objects travel easily.

17.7 Durability

Notches can survive longer than spoken agreements.

17.8 Language-light operation

Basic one-to-one tallies require little linguistic complexity.

17.9 Visual comparison

Rows and groups can be compared directly.

17.10 Embeddability

Tallies combine with labels, pictures, contracts and containers.

17.11 Tamper-evident variants

Split tallies support authentication.

17.12 Technological continuity

The same logical operation scales into mechanical and digital counters.

18.1 Poor compression

Large quantities generate many marks.

18.2 Recounting cost

Ungrouped tallies must be scanned serially.

18.3 Weak semantics

The marks do not inherently identify what they count.

18.4 Category confusion

Multiple quantities require labels, separate rows or separate devices.

18.5 Limited calculation

A fixed row supports counting and comparison better than complex arithmetic.

18.6 Recording error

The user may skip, duplicate or misclassify events.

18.7 Tampering

Marks may be added, removed or recut.

18.8 Material loss

The object can decay, burn, break or disappear.

18.9 Authoritative capture

Whoever controls the tally can control the official quantity.

18.10 Context dependence

A tally separated from its owner, place or label can become uninterpretable.

18.11 Ambiguous zero

An empty tally may mean:

• zero;
• not yet started;
• erased;
• missing data;
• wrong device.

18.12 Negative and fractional quantities

Simple tallies do not naturally express debt direction, fractions or signed values without added conventions.

| Dimension | Rating | Explanation | |---|---|---| | Reach | Low to moderate | Depends on portability and institutional recognition | | Latency | Low | Marks can be added immediately | | Bandwidth | Low | Best suited to quantity, recurrence and simple categories | | Fidelity | High for correctly recorded unit counts; low for lost semantics | Marks preserve total better than referent | | Persistence | Moderate to very high | Depends on medium | | Replication cost | Low for simple copies; higher for authenticated split systems | Recopying may introduce error | | Distribution cost | Low | Portable devices are cheap to move | | Accessibility | High for unary use | Advanced grouping conventions require instruction | | Portability | High | Small sticks, bones, cords or sheets travel easily | | Interactivity | Moderate | Users can add, remove, regroup and reconcile | | Searchability | Low | Tallies lack indexing unless embedded in a larger archive | | Editability | High but risky | Ease of update also permits tampering | | Scalability | Low in pure unary form; moderate with grouping | Numerals scale better | | Authentication | Low by default; high in split or witnessed systems | Design determines trust | | Privacy | Moderate | Physical custody can restrict access | | Censorship resistance | Moderate | Simple local devices require no central infrastructure | | Infrastructure dependence | Very low for basic tallying | Institutional tallies require governance | | Energy dependence | Very low | Human labour only | | Interpretive burden | Low when code and context are known; very high when lost | Archaeological problem is severe | | Attention demand | Low during simple increments; moderate during reconciliation | Grouping reduces scanning effort |

20.1 Memory constraint

The total no longer depends entirely on recall.

20.2 Presence constraint

The record can be consulted when the original counter is absent.

20.3 Synchronisation constraint

Events can accumulate asynchronously.

20.4 Verification constraint

Others can recount or match the record.

20.5 Interruption constraint

Counting can resume later.

20.6 Comparison constraint

Separate quantities can be compared materially.

20.7 Trust constraint

Authenticated variants reduce dependence on one party’s testimony.

20.8 Literacy constraint

Simple tallies can operate without full writing.

21.1 Codebook dependence

Users must know the unit and grouping rules.

21.2 Custody dependence

Possession of the tally may determine control over the claim.

21.3 Recording discipline

Every relevant event must be marked consistently.

21.4 Material dependence

The record is vulnerable to physical damage.

21.5 Scale dependence

Large quantities overwhelm unary representation.

21.6 Administrative dependence

Contractual tallies require organisations willing to recognise them.

21.7 Audit dependence

A visible tally still needs trusted procedures for reconciliation.

22.1 Subsistence

Tallies can track:

• animals taken;
• stored food;
• seasonal events;
• tool production;
• shared resources.

22.2 Trade

They preserve deliveries, exchanges and outstanding obligations.

22.3 Governance

They support taxation, tribute, labour and redistribution.

22.4 Law

Authenticated tallies can serve as evidence of agreement or debt.

22.5 Science

Repeated observations become countable data.

22.6 Medicine

Symptoms, doses and recurring events can be monitored.

22.7 Education

Tallies make one-to-one correspondence and grouping visible.

22.8 Timekeeping

Repeated intervals can be accumulated into schedules and calendars.

22.9 Warfare

Tallies can track troops, supplies, casualties and captured goods.

22.10 Religion

Repeated prayers, offerings or ritual acts can be counted.

22.11 Sport and play

Scores convert events into cumulative competitive state.

22.12 Computing

Counters remain fundamental to loops, events, memory addresses, databases, telemetry and algorithms.

Tallies contributed to the work of:

• traders;
• herders;
• storekeepers;
• tax collectors;
• treasurers;
• clerks;
• tally cutters;
• auditors;
• judges;
• scorekeepers;
• enumerators;
• scientists;
• programmers;
• data analysts.

The technology itself is simple. The social machinery around an authoritative count can become enormous.

24.1 Who defines the unit?

Counting requires deciding what counts as one.

That choice can be political.

24.2 Who controls the record?

The tally keeper may control:

• taxation;
• wages;
• debt;
• rationing;
• attendance;
• votes;
• performance metrics.

24.3 Who may inspect it?

Visibility can support accountability or reinforce authority.

24.4 Who can challenge it?

A tally is only as fair as the reconciliation process.

24.5 Who is counted?

States and organisations gain power by making people, land and production legible as quantities.

24.6 What is excluded?

What cannot be easily tallied may be treated as less important.

24.7 Metric substitution

Once a tally becomes a target, participants may optimise the count rather than the underlying purpose.

The ancient notch already contains the seed of the modern dashboard’s favourite sin: mistaking the metric for the thing itself.

25.1 False precision

A clean total can hide an unstable category definition.

25.2 Dehumanisation

People can become units in tax, labour, casualty or productivity records.

25.3 Administrative extraction

Tallies can make resources easier to seize.

25.4 Fraud

Marks may be altered or selectively recorded.

25.5 Surveillance

Repeated behaviour becomes visible to authorities.

25.6 Metric gaming

Participants change behaviour to improve the tally.

25.7 Context loss

Quantity can eclipse quality, cause and circumstance.

25.8 Exclusion

Those unable to access or interpret the official tally may be disadvantaged.

25.9 Heritage distortion

Modern ideological narratives may overstate prehistoric mathematical claims to satisfy present-day identity needs.

Recognising ancient African innovation does not require turning every notch into a spreadsheet with mammoth-skin pivot tables.

26.1 Approximate number sense

Supports rough quantity discrimination.

26.2 Small-number recognition

Supports direct perception of tiny sets.

26.3 One-to-one correspondence

Provides the operational foundation.

26.4 Biological memory

Retains the unit rule and context.

26.5 Gesture and body counting

Fingers and body parts provide ordered count tokens.

26.6 Mark-making

Provides durable visible units.

26.7 Tool use

Supports controlled incision.

26.8 Shared convention

Users must agree on what marks and groups mean.

26.9 Social need

Repeated exchange, storage, observation or obligation makes external counting valuable.

Tallying persists because it remains excellent for:

• live counting;
• low-stakes scorekeeping;
• quick inventory;
• voting checks;
• attendance;
• repetitive tasks;
• informal debt;
• visible progress;
• manual backup;
• teaching number concepts.

The five-bar tally survives because it is:

• immediately incrementable;
• easy to group;
• language-light;
• resistant to mental arithmetic failure;
• readable without specialist equipment.

29.1 A prehistoric recorder adds one more notch

The exact subject is lost, but the act represents a profound transition: an event changes a durable external state.

29.2 Border Cave

Notched organic artefacts demonstrate early sophisticated mark-making and possible artificial memory practices in southern Africa [6][7].

29.3 Ishango

Grouped incisions from Central Africa become a modern symbol of the deep history of numerical thought, while also warning against interpretive overreach [15].

29.4 An Exchequer official splits a tally

A count becomes a bilateral authenticated record.

29.5 The Bank of England’s 1694 loan tally

An ancient recording logic participates in early modern state finance [18].

29.6 A modern election officer records five votes

The same accumulative structure persists beneath organisations of enormous complexity.

29.7 A programmer writes counter += 1

The notch has disappeared. The operation has not.

30.1 Biological memory versus tally

| Dimension | Biological memory | Tally | |---|---|---| | Storage location | Neural | Material | | Exact large quantity | Weak | Strong if accurately recorded | | Context | Rich | Often sparse | | Portability | Embodied | Object-dependent | | Auditability | Low | Moderate to high | | Adaptability | High | Low | | Distortion | Reconstructive | Physical and procedural |

30.2 Pictogram versus tally

| Dimension | Pictogram | Tally | |---|---|---| | Primary function | Category or instruction | Quantity or recurrence | | Mark diversity | Multiple sign types | Repeated units | | Meaning source | Iconicity and convention | Correspondence and context | | Scalability | Vocabulary-limited | Quantity-limited | | Successor path | Writing and icons | Numerals and accounting |

30.3 Tally versus token

| Dimension | Tally | Token | |---|---|---| | Form | Repeated marks on one device | Separate manipulable objects | | Update | Add mark | Add, remove or move object | | Calculation | Limited | Stronger physical manipulation | | Portability | Often high | Depends on container | | Fixedness | Usually fixed | Reconfigurable |

30.4 Tally versus numeral

| Dimension | Tally | Numeral | |---|---|---| | Learning cost | Low | Higher | | Conciseness | Low | High | | Incremental transparency | High | Moderate | | Large values | Cumbersome | Efficient | | Arithmetic | Basic | Extensive with rules | | Dependence on convention | Low to moderate | High |

30.5 Tally versus database counter

| Dimension | Manual tally | Database counter | |---|---|---| | Scale | Small | Massive | | Speed | Human | Automated | | Inspectability | Direct | Interface-mediated | | Infrastructure | Minimal | High | | Error source | Human recording | Code, sensors, queries and data definitions | | Governance | Local | Institutional and technical |

| Claim ID | Claim | Confidence | Notes | |---|---|---|---| | ENC006-C01 | Tallying externalises quantity through material correspondence | Established | Operational definition | | ENC006-C02 | Basic tallying can function without compact numeral symbols | Established | One token per unit | | ENC006-C03 | Exact counting involves one-to-one correspondence, stable order and cardinal interpretation | Established | Developmental literature [10][11] | | ENC006-C04 | Material tallies reduce working-memory demand | High | Cognitive offloading framework [2][3] | | ENC006-C05 | Sequential markings were used as artificial numerical memory systems in the Upper Palaeolithic | Broadly accepted with case-specific caution | [1][5] | | ENC006-C06 | Notched artefacts from Border Cave date to roughly 44,000–42,000 years ago | Established | [6][7] | | ENC006-C07 | The Border Cave baboon fibula was definitely a lunar or menstrual calendar | Speculative | Do not state as fact | | ENC006-C08 | Regular notches alone prove numerical intent | False | Multiple possible functions [4] | | ENC006-C09 | The Ishango artefact contains deliberately grouped notches | Established | [15] | | ENC006-C10 | The Ishango artefact definitively encodes prime numbers or base twelve | Debated/speculative | Competing interpretations | | ENC006-C11 | Grouping improves the readability and scalability of tallies | High | Material and perceptual logic [3] | | ENC006-C12 | Pure unary tallies scale poorly to large quantities | Established | Structural property | | ENC006-C13 | Split tally construction can reveal later alteration | Established | [18][19] | | ENC006-C14 | English tally sticks recorded debts and payments | Established | [17][18][20] | | ENC006-C15 | Some tally-based debt claims circulated through transfer | Established for documented systems | [19][21] | | ENC006-C16 | Tallies disappeared when writing emerged | False | Long historical coexistence | | ENC006-C17 | Tallying contributed to the development of numerical notation | High but non-linear | [1][2][9] | | ENC006-C18 | A tally can preserve quantity while losing the identity of what was counted | Established | Semantic limitation | | ENC006-C19 | Counting systems are purely biological and culturally invariant | Unsupported | Cross-cultural evidence stresses cultural construction [8] | | ENC006-C20 | Modern digital counters retain the accumulative logic of tallies | Established as analytical continuity | Not direct descent in every case | | ENC006-C21 | A visible count is necessarily objective | False | Unit definition and recording governance remain social | | ENC006-C22 | Metrics can reshape behaviour once they become targets | High | General institutional consequence | | ENC006-C23 | Prehistoric calendar interpretations require evidence beyond repeated mark counts | High | [14] | | ENC006-C24 | Tallying should be classified only as encoding | Insufficient | It is also storage and elementary processing |

1. Which early notched artefacts meet the strongest criteria for cumulative notation?
2. Can experimental archaeology reliably distinguish decorative incision from episodic tallying?
3. How often did tally systems emerge independently?
4. Did exact number words normally precede, accompany or follow material tallies?
5. How were tally referents labelled in non-literate contexts?
6. What proportion of prehistoric notational objects were made on perishable wood and never survived?
7. How did finger counting influence grouping conventions?
8. Under what conditions do tallies acquire weighted values?
9. How did local tally systems interact with state writing systems?
10. When does a debt record become transferable money-like value?
11. How should African prehistoric numerical artefacts be presented without either colonial dismissal or modern mythmaking?
12. Which tally practices survive in Zimbabwean and southern African communities, trades and games?
13. How do digital event counters reproduce old tally problems through poor event definitions?
14. What is lost when qualitative human activity is converted into countable metrics?
15. Should tallying and one-to-one token matching be separate topics in v0.2?

33.1 The first external counter

Show:

event → notch → accumulated row → recoverable total

33.2 Quantity leaving the brain

Split composition:

• left: person rehearsing a count;
• right: marks holding the total.

33.3 Tally versus numeral

Compare representation of 37:

• thirty-seven unary marks;
• grouped tally marks;
• 37;
• three tens and seven units;
• binary.

33.4 Archaeological confidence ladder

Visual sequence:

scratch → deliberate mark → sequence → probable notation → probable tally → proposed calendar

Each step should require more evidence.

33.5 Border Cave case card

Include:

• location;
• approximate date;
• material;
• notch count;
• secure facts;
• speculative interpretations.

33.6 Ishango interpretation wheel

Place the object at the centre with competing hypotheses around it, clearly colour-coded by confidence.

33.7 Split tally animation

1. notch the stick;
2. split along the grain;
3. distribute halves;
4. attempt alteration;
5. reunite and expose mismatch.

33.8 Evolution chain

fingers → notches → grouped tallies → tokens → numerals → abacus → ledger → spreadsheet → database counter

33.9 Metric trap graphic

Show the object of interest shrinking behind an ever-larger number.

33.10 Interactive tally simulator

Allow users to:

• add units;
• choose grouping size;
• compare unary and positional notation;
• simulate missed and duplicated events;
• test split-tally tampering.

34.1 Flagship article

The First Data Structure Was a Scratch

Angle:

A tally is an append-only log built from physical marks.

34.2 Historical article

Before Spreadsheets, Humans Counted With Bone and Wood

34.3 Myth-correction article

The Lebombo Bone Is Fascinating. It Is Not a Proven Menstrual Calendar

34.4 African history article

Africa’s Deep History of Numerical Record-Keeping

The tone should protect both significance and uncertainty.

34.5 Computer-history video

From Tally Marks to counter += 1

34.6 Information-theory video segment

Why Counting Requires a Codebook

34.7 Finance video

The Wooden Stick That Worked Like a Financial Instrument

34.8 Systems-thinking article

When the Tally Becomes the Target

Connect ancient counting to modern KPIs.

34.9 Demonstration video

Create three systems for the same transaction:

• remembered count;
• unary tally;
• split tally.

Then introduce disagreement and tampering.

34.10 Short-form hooks

• “A tally is a database with one field and a knife.”
• “Humans invented append-only logs before they invented writing.”
• “The first spreadsheet may have been a bone, but its column headings are missing.”
• “A number can be accurate while the thing it measures is nonsense.”

Option A: Bone to Spreadsheet

Visual: notched bone on the left, glowing spreadsheet grid on the right.
Text: FIRST DATABASE?

Option B: One More Mark

Visual: extreme close-up of a hand cutting the final notch.
Text: COUNTING BEGINS

Option C: Ishango Mystery

Visual: the notched artefact surrounded by competing mathematical overlays, most crossed out.
Text: WHAT DID IT COUNT?

Option D: Split Tally

Visual: two matching wooden halves snapping together.
Text: UNHACKABLE DEBT?

Option E: Metric Trap

Visual: human figure buried beneath enormous KPI numbers.
Text: THE COUNT TOOK OVER

For Clueless Pundit, the strongest treatment is an archival object hero with yellow annotation marks and one restrained modern data contrast. The treatment should preserve one dominant idea, phone-size readability and Clueless Pundit recognition through archival texture and yellow electricity.

| Field | Value | |---|---| Tally marks and notches | | Method name | Tally marks and notches | | Recommended analytical name | Tallying and accumulative quantity marks | | Alternative names | Tallying; score marks; notched records; unary quantity marks | | Topic type | Quantitative external-memory method and elementary notational system | | Primary category | Encoding and expression | | Secondary categories | Storage; processing; governance; authentication | | Earliest known evidence | Probable Upper Palaeolithic accumulative notation; specific earlier claims disputed | | Practical introduction | Prehistoric and independently recurrent | | Mass-adoption period | Not applicable as one global event | | Dominance period | Persistent alongside later systems | | Current status | Surviving specialised method and foundational computational operation | | Geographic origin | Multiple independent origins probable | | Communication pattern | Asynchronous external record; one-to-self, one-to-one and institutional | | Persistence type | Temporary to durable, depending on medium | | Primary prerequisites | One-to-one correspondence; memory; mark-making; shared unit definition | | Primary predecessors | Neural and cognitive memory; Pictograms and ideograms; body counting | | Primary successors | Tokens and accounting objects; Numerical notation; Khipu and other knot-record systems; ledgers; counters | | Main problem addressed | Exact quantity and recurrence beyond unaided memory | | New trade-offs and dependencies | Scaling, codebook dependence, custody and semantic ambiguity | | Reach | Low to moderate | | Latency | Low | | Bandwidth | Low | | Fidelity | High for total when accurately recorded; low for lost context | | Persistence | Medium to high | | Replication cost | Low | | Distribution cost | Low | | Accessibility | High for unary tallies | | Portability | High | | Interactivity | Moderate | | Searchability | Low | | Editability | High | | Authentication | Variable; high in split systems | | Privacy | Moderate | | Censorship resistance | Moderate | | Infrastructure dependence | Very low to moderate | | Energy dependence | Very low | | Interpretive burden | Low with codebook; extreme without it | | Main benefits | Cognitive offloading; exact accumulation; auditability; low cost | | Main harms | Tampering; false precision; administrative extraction; metric fixation | | Organisations created | Tally keepers, clerks, treasuries, auditors, scorekeepers | | Representative event | Split Exchequer tally and Bank of England 1694 loan tally | | Source confidence | High for documented systems; variable for prehistory |

Replace the existing row with:

|---|---|---|---|---|---|---|---|---|---|---| | Tally marks and notches | Tallying and accumulative quantity marks | Quantitative external-memory method | Encoding & expression | Storage; processing; governance; authentication | Probable Upper Palaeolithic use; exact earliest examples disputed | Externalises quantity, recurrence and sequence beyond unaided memory | Memory; one-to-one matching; body counting; conventional marks | Tokens; numerical notation; knot records; ledgers; counters | Core | Researched |

38.1 Add accumulative versus descriptive encoding

Distinguish systems that describe a state from systems whose material state changes incrementally as events occur.

38.2 Add unit-definition field

Every quantitative topic should identify what constitutes one unit.

38.3 Add representation economy

Record the relationship between quantity represented and number of physical signs required.

38.4 Add manipulability

Distinguish fixed marks from moveable counters.

38.5 Add auditability

Can another person reconstruct how the total was produced?

38.6 Add tamper evidence

Can alteration be detected?

38.7 Add referent persistence

Does the record preserve what was counted, or only how many?

38.8 Add metric-governance risk

Quantitative systems can reshape behaviour when counts become institutional targets.

38.9 Clarify processing

Tallying is elementary information processing because the system updates state, supports comparison and can perform addition or subtraction materially.

Archaeology, numerical cognition and materiality

[1] d’Errico, Francesco, Luc Doyon, Ivan Colagé, Alain Queffelec, Emma Le Vraux, Giacomo Giacobini, Bernard Vandermeersch and Roberto Macchiarelli. “From Number Sense to Number Symbols: An Archaeological Perspective.” Philosophical Transactions of the Royal Society B 373 (2018): 20160518.
https://doi.org/10.1098/rstb.2016.0518

[2] Overmann, Karenleigh A. The Materiality of Numbers: Emergence and Elaboration from Prehistory to Present. Cambridge University Press, 2023.
https://doi.org/10.1017/9781009361262

[3] Overmann, Karenleigh A. “Tallies and Other Devices That Accumulate.” In The Materiality of Numbers, 2023.
https://www.cambridge.org/core/services/aop-cambridge-core/content/view/A8083D534D9E478FFF7A313742FD2848/9781009361248c10_220-243.pdf/tallies-and-other-devices-that-accumulate.pdf

[4] Overmann, Karenleigh A. “Interpreting Prehistoric Artifacts.” In The Materiality of Numbers, 2023.
https://doi.org/10.1017/9781009361262.013

[5] Courtenay, Lloyd Austin, Francesco d’Errico, Rafael Núñez and Damián E. Blasi. “Identifying Potential Palaeolithic Artificial Memory Systems via Spatial Statistics: Implications for the Origin of Quantification.” Archaeological and Anthropological Sciences 17 (2025): 171.
https://doi.org/10.1007/s12520-025-02286-4

[6] d’Errico, Francesco, Lucinda Backwell, Paola Villa, Ingrid Degano, Jeannette J. Lucejko, Marion K. Bamford, Thomas F. G. Higham, Maria Perla Colombini and Peter B. Beaumont. “Early Evidence of San Material Culture Represented by Organic Artifacts from Border Cave, South Africa.” Proceedings of the National Academy of Sciences 109 (2012): 13214–13219.
https://doi.org/10.1073/pnas.1204213109

[7] Villa, Paola, et al. “Border Cave and the Beginning of the Later Stone Age in South Africa.” Proceedings of the National Academy of Sciences 109 (2012): 13208–13213.
https://doi.org/10.1073/pnas.1202629109

[8] O’Shaughnessy, David M., et al. “The Cultural Origins of Symbolic Number.” Psychological Review 129 (2022): 1442–1456.
https://pmc.ncbi.nlm.nih.gov/articles/PMC8678391/

[9] Schlaudt, Oliver. “Type and Token in the Prehistoric Origins of Numbers.” Cambridge Archaeological Journal 30 (2020): 629–646.
https://doi.org/10.1017/S0959774320000165

Counting principles and exact number

[10] Gelman, Rochel and C. R. Gallistel. The Child’s Understanding of Number. Harvard University Press, 1978.

[11] Slaughter, Virginia, et al. “Learning to Count Begins in Infancy: Evidence from 18-Month-Olds’ Visual Preferences.” Proceedings of the Royal Society B 278 (2011): 2979–2984.
https://pmc.ncbi.nlm.nih.gov/articles/PMC3151703/

[12] Koopman, Sarah E., Alyssa M. Arre, Steven T. Piantadosi and Jessica F. Cantlon. “One-to-One Correspondence Without Language.” Royal Society Open Science 6 (2019): 190495.
https://doi.org/10.1098/rsos.190495

Upper Palaeolithic notation debate

[13] Bacon, Bennett, Azadeh Khatiri, James Palmer, Tony Freeth, Paul Pettitt and Robert Kentridge. “An Upper Palaeolithic Proto-writing System and Phenological Calendar.” Cambridge Archaeological Journal 33 (2023): 371–389.
https://doi.org/10.1017/S0959774322000415

[14] Nowell, April, Paul Bahn and Jean-Loïc Le Quellec. “Evaluating the Evidence for Lunar Calendars in Upper Palaeolithic Parietal Art.” Cambridge Archaeological Journal 35 (2025).
https://doi.org/10.1017/S0959774324000155

Ishango

[15] Institute of Natural Sciences, Brussels. “The Ishango Bone.”
https://www.naturalsciences.be/en/museum/exhibitions-activities/exhibitions/250-years-of-natural-sciences/the-ishango-bone

[16] Pletser, Vladimir. “Does the Ishango Bone Indicate Knowledge of the Base 12?” 2012.
https://arxiv.org/abs/1204.1019
Use as an example of a proposed interpretation, not as settled consensus.

Historical tally systems

[17] The National Archives, United Kingdom. “Revealing the (Corrupt) Record in 14th-Century Ireland.”
https://www.nationalarchives.gov.uk/explore-the-collection/stories/revealing-the-corrupt-record-in-14th-century-ireland/

[18] Bank of England Museum. “Payments Through Time: Wooden Tally Stick.”
https://www.bankofengland.co.uk/museum/whats-on/2019/325-years-exhibition/payments-through-time

[19] Bank of England Museum. The Future of Money: Large Print Guide. 2024.
https://www.bankofengland.co.uk/-/media/boe/files/museum/the-future-of-money-large-print-guide.pdf

[20] British Museum. Money Gallery: Large Print Guide, Room 68.
https://www.britishmuseum.org/sites/default/files/2021-05/Money_Gallery_LPG_2020_Room_68.pdf

[21] Cunliffe, Jon. “It’s Time to Talk About Money.” Bank of England, 2020.
https://www.bankofengland.co.uk/-/media/boe/files/speech/2020/its-time-to-talk-about-money-speech-by-jon-cunliffe.pdf

[22] Levavi, Yuval. “Tallying in the Eanna.” Iraq 84 (2022): 49–62.
https://doi.org/10.1017/irq.2022.4

Time, notational systems and successor technologies

[23] Cooperrider, Kensy. “Time Tools.” Behavioral and Brain Sciences / open-access review, 2025.
https://pmc.ncbi.nlm.nih.gov/articles/PMC12831611/

[24] Overmann, Karenleigh A. “Handwritten Notations.” In The Materiality of Numbers, 2023.
https://www.cambridge.org/core/books/materiality-of-numbers/handwritten-notations/8AC3F658B5CD3C52FBD21D530AC3FA15

[25] Overmann, Karenleigh A. “Devices That Accumulate and Group.” In The Materiality of Numbers, 2023.
https://www.cambridge.org/core/services/aop-cambridge-core/content/view/055DF6D51F15DDF36B355B34A1228ACB/9781009361248c12_277-308.pdf/devices-that-accumulate-and-group.pdf

40.1 The breakthrough is external accumulation

A mark can stand for an event that no longer exists.

A row of marks can stand for a history of repeated events.

40.2 The tally is a process, not merely a picture

Its meaning arises from an update rule:

when one unit occurs, change the material state

40.3 It makes quantity portable

The counted objects may move, perish or be consumed. Their numerical trace remains.

40.4 It turns memory into evidence

A personal reminder can become a shared record, an audit trail or a legal claim.

40.5 It exposes the politics of counting

Every tally depends on decisions about:

• what counts;
• who counts it;
• when it is recorded;
• who may inspect it;
• what consequences follow.

40.6 It is powerful because it is simple

One mark per unit is easy to invent, teach and verify.

40.7 It is limited for the same reason

A large total produces a large forest of marks, while the subject and context may remain invisible.

40.8 Numerical notation solves the compression problem

Grouped and positional symbols represent far more with far fewer signs.

40.9 The logic never disappears

Mechanical counters, database increments, website analytics and sports scoreboards all preserve the old operation.

40.10 Final topic statement

Tallying transformed quantity from a fragile mental state into an accumulative material record. By pairing events or objects with durable marks, humans could preserve exact recurrence, resume interrupted counts, compare sets and turn remembered obligations into inspectable evidence. Its simplicity made it nearly universal; its poor compression and weak semantics drove the development of tokens, numerals, ledgers and computational counters.