Translate Archaeological Pottery Records Correctly

To translate archaeological pottery records correctly, preserve what was observed, where each sherd came from, and how certain the interpretation is. A pottery assemblage record may contain excavation context numbers, local fabric codes, rim or base classifications, sherd counts, weights, estimates of vessel equivalents, and a tentative date range. If a translator turns a fabric category into a place of manufacture, moves a sherd into another stratigraphic context or treats a minimum vessel count as a proven total, the translation changes the archaeological evidence.

Readers searching for archaeological pottery report translation, pottery sherd catalogue terminology, fabric code meanings, context numbers, rim sherd identification, sherd weight versus count, vessel equivalent estimates, stratigraphy and pottery dating uncertainty need a clear procedure. The best translation keeps evidence and inference separate. It allows a reader of a different language to understand the assemblage without quietly inventing age, origin or completeness that the excavators did not demonstrate.

This guide follows the recording logic described by the Chartered Institute for Archaeologists’ pottery toolkit: contextual identifiers, ware and fabric, sherd part, vessel form, surface treatment, quantity, condition and dating all serve different purposes. The worked records below are invented teaching examples rather than authentic excavation finds. The central discipline is to protect each original code and its uncertainty while writing readable prose that remains auditable against the source catalogue.

The problem: a readable sentence can still change the underlying task

A translator can produce an elegant paragraph and still fail at archaeological pottery recording sheets and assemblage descriptions. The problem is that the source is a collection of instructions, evidence and relationships, not just a page of sentences. A term might be a code, a measured quantity, a diagram instruction or a descriptive judgement. Replacing every string with an ordinary word is not enough: the translated result needs to leave the same decisions available to the intended reader. If an essential distinction disappears, the text has not transmitted the original knowledge.

Begin by finding the smallest reversible demonstration of excavation context, local fabric typology, count, weight and inference. Mark what may be paraphrased, what must be kept literally and what still needs a specialist’s decision. Read the source twice: once for the main practical goal and once for exceptions, legends, notes and identifiers. A novice reader should be able to answer three questions after the translation: What do I do first? Which values or entities must I not change? How can I tell whether the outcome matches the source?

First principles: the distinctions that control translation accuracy

1. A context is a provenance key

An excavation context ID links a find to the unit from which it was recovered. Its number should stay attached to the sherd throughout the translation, even when neighbouring contexts contain apparently similar pottery. This matters because archaeological pottery recording sheets and assemblage descriptions requires a clear connection between a source element and its practical consequence. When studying a difficult line, write the source term, its role, a permitted explanation and one forbidden confusion in adjacent columns. A learner who can name the forbidden confusion has begun to understand the concept instead of memorising the surface vocabulary.

A useful self-check is to cover the original sentence and explain the relationship in ordinary language, then reveal the source and verify every protected element. If the explanation requires a guess or introduces an actor, time, measurement or outcome that the source did not contain, label the problem for review. The best translation remains faithful even when it must explicitly say that a point has not been established.

2. Fabric codes depend on a type series

A code can be project-specific or part of a local or regional fabric system. It must not be translated as if it were a universally recognised material, maker or country. This matters because archaeological pottery recording sheets and assemblage descriptions requires a clear connection between a source element and its practical consequence. When studying a difficult line, write the source term, its role, a permitted explanation and one forbidden confusion in adjacent columns. A learner who can name the forbidden confusion has begun to understand the concept instead of memorising the surface vocabulary.

3. Ware and fabric are related but not identical

Ware names group traditions or products, while fabric notes describe material characteristics. Translate the roles accurately and preserve the local catalogue’s distinctions. This matters because archaeological pottery recording sheets and assemblage descriptions requires a clear connection between a source element and its practical consequence. When studying a difficult line, write the source term, its role, a permitted explanation and one forbidden confusion in adjacent columns. A learner who can name the forbidden confusion has begun to understand the concept instead of memorising the surface vocabulary.

4. The sherd’s part matters

Rim, body, base, handle and spout are not arbitrary labels. Their occurrence may affect interpretation of vessel form, minimum counts and illustration selection. This matters because archaeological pottery recording sheets and assemblage descriptions requires a clear connection between a source element and its practical consequence. When studying a difficult line, write the source term, its role, a permitted explanation and one forbidden confusion in adjacent columns. A learner who can name the forbidden confusion has begun to understand the concept instead of memorising the surface vocabulary.

5. Count differs from weight

Ten sherds weighing 200 grams is different information from one 200-gram sherd. Preserve both variables and units without reducing them to a single impression of quantity. This matters because archaeological pottery recording sheets and assemblage descriptions requires a clear connection between a source element and its practical consequence. When studying a difficult line, write the source term, its role, a permitted explanation and one forbidden confusion in adjacent columns. A learner who can name the forbidden confusion has begun to understand the concept instead of memorising the surface vocabulary.

6. Vessel counts are often estimates

An EVE or minimum number of vessels is a quantitative approach, not proof that exactly that many whole vessels existed. Explain method and limits where supplied. This matters because archaeological pottery recording sheets and assemblage descriptions requires a clear connection between a source element and its practical consequence. When studying a difficult line, write the source term, its role, a permitted explanation and one forbidden confusion in adjacent columns. A learner who can name the forbidden confusion has begun to understand the concept instead of memorising the surface vocabulary.

7. Date range is not a deposition date

A ware-type date range may reflect production and distribution, not the exact moment a sherd entered the ground. Keep the archaeological inference qualified. This matters because archaeological pottery recording sheets and assemblage descriptions requires a clear connection between a source element and its practical consequence. When studying a difficult line, write the source term, its role, a permitted explanation and one forbidden confusion in adjacent columns. A learner who can name the forbidden confusion has begun to understand the concept instead of memorising the surface vocabulary.

8. Condition is separate from manufacture

Abrasion, fresh breaks, burning or water wear may describe post-depositional or use history. Do not automatically convert these into claims about the firing method or date. This matters because archaeological pottery recording sheets and assemblage descriptions requires a clear connection between a source element and its practical consequence. When studying a difficult line, write the source term, its role, a permitted explanation and one forbidden confusion in adjacent columns. A learner who can name the forbidden confusion has begun to understand the concept instead of memorising the surface vocabulary.

9. Surface treatment differs from decoration

Glaze, slip and burnishing refer to treatments, while an incised wavy line or stamped design may be described under decoration. The distinction helps later analysis. This matters because archaeological pottery recording sheets and assemblage descriptions requires a clear connection between a source element and its practical consequence. When studying a difficult line, write the source term, its role, a permitted explanation and one forbidden confusion in adjacent columns. A learner who can name the forbidden confusion has begun to understand the concept instead of memorising the surface vocabulary.

10. Cross-fitting requires traceability

Joined sherds from two contexts may support a relation between pieces without erasing their recorded findspots. Report both contexts and the nature of the join. This matters because archaeological pottery recording sheets and assemblage descriptions requires a clear connection between a source element and its practical consequence. When studying a difficult line, write the source term, its role, a permitted explanation and one forbidden confusion in adjacent columns. A learner who can name the forbidden confusion has begun to understand the concept instead of memorising the surface vocabulary.

11. Residual and intrusive are interpretations

A piece older than a layer may be residual; a later item in older deposits may be intrusive. Such language must be retained with appropriate caution and contextual evidence. This matters because archaeological pottery recording sheets and assemblage descriptions requires a clear connection between a source element and its practical consequence. When studying a difficult line, write the source term, its role, a permitted explanation and one forbidden confusion in adjacent columns. A learner who can name the forbidden confusion has begun to understand the concept instead of memorising the surface vocabulary.

12. A record can be incomplete

Unknown fabric, uncertain form, missing weight and unillustrated sherds should not become confidently classified artefacts. Explicit blanks remain evidence about the limits of recording. This matters because archaeological pottery recording sheets and assemblage descriptions requires a clear connection between a source element and its practical consequence. When studying a difficult line, write the source term, its role, a permitted explanation and one forbidden confusion in adjacent columns. A learner who can name the forbidden confusion has begun to understand the concept instead of memorising the surface vocabulary.

A repeatable twelve-stage method

Use the following workflow in order when accuracy matters. It is deliberately slower than casual instant translation because archaeological pottery recording sheets and assemblage descriptions can contain compact instructions with large consequences. Once the routine becomes familiar, several checks can be combined, but no review should depend on having memorised the entire document.

Stage 1: Identify the audience

Identify whether the reader is a novice, a practitioner, an editor, or a machine that must ingest the translated result. The same source may need explanatory language for a beginner while a database import requires exact original identifiers. In this topic, apply that rule specifically to excavation context, local fabric typology, count, weight and inference. Before leaving the stage, write one line of evidence showing which field, phrase or illustration you used. That brief trace is often enough to diagnose a later disagreement without redoing the entire translation.

Stage 2: Freeze the source version

Record the document, its revision, source language, source date and any accompanying legend before rewriting. A later version may amend directions or rename fields, and silently mixing versions makes a seemingly fluent translation unreliable. In this topic, apply that rule specifically to excavation context, local fabric typology, count, weight and inference. Before leaving the stage, write one line of evidence showing which field, phrase or illustration you used. That brief trace is often enough to diagnose a later disagreement without redoing the entire translation.

Stage 3: Separate data from natural language

Mark codes, measurements, counts, identifiers, diagram relations and timestamps as protected data before drafting sentences. Ordinary prose can be adapted, but a protected token cannot be improved by an elegant synonym or an unapproved number conversion. In this topic, apply that rule specifically to excavation context, local fabric typology, count, weight and inference. Before leaving the stage, write one line of evidence showing which field, phrase or illustration you used. That brief trace is often enough to diagnose a later disagreement without redoing the entire translation.

Stage 4: Read the whole task

Inspect headings, captions, side notes, tables, diagrams, exceptions and cross-references, not merely the first paragraph. A repeated word may have different meanings in different parts of the same instruction, and a footnote may reverse a default. In this topic, apply that rule specifically to excavation context, local fabric typology, count, weight and inference. Before leaving the stage, write one line of evidence showing which field, phrase or illustration you used. That brief trace is often enough to diagnose a later disagreement without redoing the entire translation.

Stage 5: Create a small term map

Write down the five to fifteen terms that determine the procedure and use stable target-language expressions. Do not rely on memory across a long document when one accidental synonym can be interpreted as a new operation. In this topic, apply that rule specifically to excavation context, local fabric typology, count, weight and inference. Before leaving the stage, write one line of evidence showing which field, phrase or illustration you used. That brief trace is often enough to diagnose a later disagreement without redoing the entire translation.

Stage 6: Translate instructions as actions

Convert each instruction into the actor, action, object, order and condition it actually describes. A grammatical translation is only correct if the reader can still perform the same task in the same sequence. In this topic, apply that rule specifically to excavation context, local fabric typology, count, weight and inference. Before leaving the stage, write one line of evidence showing which field, phrase or illustration you used. That brief trace is often enough to diagnose a later disagreement without redoing the entire translation.

Stage 7: Keep uncertainty visible

Retain words such as approximately, possibly, observed, proposed, pending, reported and not yet verified. A source’s limit of knowledge is part of its meaning; translating a hypothesis as a confirmed result manufactures evidence. In this topic, apply that rule specifically to excavation context, local fabric typology, count, weight and inference. Before leaving the stage, write one line of evidence showing which field, phrase or illustration you used. That brief trace is often enough to diagnose a later disagreement without redoing the entire translation.

Stage 8: Compare side by side

Check one source segment against one target segment with its data elements highlighted. Side-by-side comparison catches dropped negations, changed actor roles, swapped directions and lost repeat conditions. In this topic, apply that rule specifically to excavation context, local fabric typology, count, weight and inference. Before leaving the stage, write one line of evidence showing which field, phrase or illustration you used. That brief trace is often enough to diagnose a later disagreement without redoing the entire translation.

Stage 9: Run an independent data audit

Read only the protected tokens in each version and compare their spelling, units, labels and placement. A natural-language reviewer may miss a one-character discrepancy because the surrounding prose appears smooth. In this topic, apply that rule specifically to excavation context, local fabric typology, count, weight and inference. Before leaving the stage, write one line of evidence showing which field, phrase or illustration you used. That brief trace is often enough to diagnose a later disagreement without redoing the entire translation.

Stage 10: Use a miniature real-world simulation

Try the translated instruction on a small reversible example or a clearly marked mock record. A demonstration reveals whether the target reader can follow the instruction without relying on the translator’s memory. In this topic, apply that rule specifically to excavation context, local fabric typology, count, weight and inference. Before leaving the stage, write one line of evidence showing which field, phrase or illustration you used. That brief trace is often enough to diagnose a later disagreement without redoing the entire translation.

Stage 11: Escalate unresolved ambiguities

Send unknown or conflicting specialist expressions to an appropriate subject reviewer with the original context attached. An editor must not silently invent a rule simply because the available alternatives all sound plausible. In this topic, apply that rule specifically to excavation context, local fabric typology, count, weight and inference. Before leaving the stage, write one line of evidence showing which field, phrase or illustration you used. That brief trace is often enough to diagnose a later disagreement without redoing the entire translation.

Stage 12: Deliver a versioned translation

Keep a change log for corrections, source citations and unresolved questions; state what was and was not validated. Traceability lets the reader distinguish a checked educational explanation from an authoritative operational record. In this topic, apply that rule specifically to excavation context, local fabric typology, count, weight and inference. Before leaving the stage, write one line of evidence showing which field, phrase or illustration you used. That brief trace is often enough to diagnose a later disagreement without redoing the entire translation.

Twenty-four worked translation cases with correction and checking

The following source fragments are invented instructional mini-records, not quotations from live production documents. Their purpose is to make errors visible while keeping each case short enough to test. For each one, compare the source, the tempting but faulty rendering, the faithful reading and the concrete verification action. Do not confuse these teaching cases with an official device manual, competition ruling or archival record.

Worked case 1: Context ID

Imagine that a reader must translate the source fragment CTX 214; rim sherd in a document about archaeological pottery recording sheets and assemblage descriptions. A plausible but faulty target wording would be “sherd number 214 from the whole site.” The problem is not necessarily grammar. It is the underlying interpretation: a context number identifies a recorded depositional unit, not necessarily one individual sherd. A reader relying on that faulty wording could make a different choice while believing the instruction had been preserved.

A safer reader-facing rendering is “rim sherd recovered in context 214.” Keep the original fragment beside the explanation wherever exact notation, codes or numerical labels matter. The translator should not silently add implied details, change an uncertainty into certainty, or use a supposedly simpler word that conceals the distinction. In this case, the decisive checking action is to match the excavation register. If the check cannot be completed, state precisely which element remains unverified rather than marking the entire record as confirmed.

Practice transfer: imagine the same fragment appears as a caption beside a diagram and later as one row in a table. Would your translation preserve the same meaning in both places? State which portion is ordinary descriptive language and which portion must survive unchanged for the next reader. Then write a one-sentence warning to an editor who has not seen the source. The warning should name the particular error in this example, not a generic instruction to ‘be careful’.

Worked case 2: Fabric code

Imagine that a reader must translate the source fragment Fabric Q2, local series in a document about archaeological pottery recording sheets and assemblage descriptions. A plausible but faulty target wording would be “made in Quarry Two.” The problem is not necessarily grammar. It is the underlying interpretation: the code may be arbitrary and cannot establish production origin. A reader relying on that faulty wording could make a different choice while believing the instruction had been preserved.

A safer reader-facing rendering is “fabric code Q2 in the project’s local typology.” Keep the original fragment beside the explanation wherever exact notation, codes or numerical labels matter. The translator should not silently add implied details, change an uncertainty into certainty, or use a supposedly simpler word that conceals the distinction. In this case, the decisive checking action is to consult the type-series key. If the check cannot be completed, state precisely which element remains unverified rather than marking the entire record as confirmed.

Worked case 3: Rim category

Imagine that a reader must translate the source fragment RIM, everted in a document about archaeological pottery recording sheets and assemblage descriptions. A plausible but faulty target wording would be “base of a vessel.” The problem is not necessarily grammar. It is the underlying interpretation: rim and base have distinct morphological meanings. A reader relying on that faulty wording could make a different choice while believing the instruction had been preserved.

A safer reader-facing rendering is “everted rim fragment.” Keep the original fragment beside the explanation wherever exact notation, codes or numerical labels matter. The translator should not silently add implied details, change an uncertainty into certainty, or use a supposedly simpler word that conceals the distinction. In this case, the decisive checking action is to compare catalogue drawing. If the check cannot be completed, state precisely which element remains unverified rather than marking the entire record as confirmed.

Worked case 4: Body sherd

Imagine that a reader must translate the source fragment BODY undiagnostic in a document about archaeological pottery recording sheets and assemblage descriptions. A plausible but faulty target wording would be “complete vessel body.” The problem is not necessarily grammar. It is the underlying interpretation: an undiagnostic fragment cannot support a confident vessel type. A reader relying on that faulty wording could make a different choice while believing the instruction had been preserved.

A safer reader-facing rendering is “body sherd not diagnostic of vessel form.” Keep the original fragment beside the explanation wherever exact notation, codes or numerical labels matter. The translator should not silently add implied details, change an uncertainty into certainty, or use a supposedly simpler word that conceals the distinction. In this case, the decisive checking action is to retain uncertainty. If the check cannot be completed, state precisely which element remains unverified rather than marking the entire record as confirmed.

Worked case 5: Weight

Imagine that a reader must translate the source fragment 7 sherds, 184 g in a document about archaeological pottery recording sheets and assemblage descriptions. A plausible but faulty target wording would be “184 separate fragments.” The problem is not necessarily grammar. It is the underlying interpretation: count and mass are separate recorded quantities. A reader relying on that faulty wording could make a different choice while believing the instruction had been preserved.

A safer reader-facing rendering is “seven fragments with combined weight 184 grams.” Keep the original fragment beside the explanation wherever exact notation, codes or numerical labels matter. The translator should not silently add implied details, change an uncertainty into certainty, or use a supposedly simpler word that conceals the distinction. In this case, the decisive checking action is to count and weigh source columns. If the check cannot be completed, state precisely which element remains unverified rather than marking the entire record as confirmed.

Worked case 6: Surface glaze

Imagine that a reader must translate the source fragment Green glaze on interior in a document about archaeological pottery recording sheets and assemblage descriptions. A plausible but faulty target wording would be “the clay itself is green.” The problem is not necessarily grammar. It is the underlying interpretation: surface finish is not automatically the underlying ceramic fabric. A reader relying on that faulty wording could make a different choice while believing the instruction had been preserved.

A safer reader-facing rendering is “green glaze present on the inner surface.” Keep the original fragment beside the explanation wherever exact notation, codes or numerical labels matter. The translator should not silently add implied details, change an uncertainty into certainty, or use a supposedly simpler word that conceals the distinction. In this case, the decisive checking action is to identify interior versus exterior. If the check cannot be completed, state precisely which element remains unverified rather than marking the entire record as confirmed.

Worked case 7: Burnishing

Imagine that a reader must translate the source fragment Burnished outer surface in a document about archaeological pottery recording sheets and assemblage descriptions. A plausible but faulty target wording would be “painted outer surface.” The problem is not necessarily grammar. It is the underlying interpretation: mechanically smoothed surfaces differ from applied paint. A reader relying on that faulty wording could make a different choice while believing the instruction had been preserved.

A safer reader-facing rendering is “outer surface burnished.” Keep the original fragment beside the explanation wherever exact notation, codes or numerical labels matter. The translator should not silently add implied details, change an uncertainty into certainty, or use a supposedly simpler word that conceals the distinction. In this case, the decisive checking action is to inspect method description. If the check cannot be completed, state precisely which element remains unverified rather than marking the entire record as confirmed.

Worked case 8: Decoration

Imagine that a reader must translate the source fragment Incised wavy line in a document about archaeological pottery recording sheets and assemblage descriptions. A plausible but faulty target wording would be “wavy line stamped into clay.” The problem is not necessarily grammar. It is the underlying interpretation: incision and stamping are different manufacturing actions. A reader relying on that faulty wording could make a different choice while believing the instruction had been preserved.

A safer reader-facing rendering is “wavy line made by incision.” Keep the original fragment beside the explanation wherever exact notation, codes or numerical labels matter. The translator should not silently add implied details, change an uncertainty into certainty, or use a supposedly simpler word that conceals the distinction. In this case, the decisive checking action is to keep the technique. If the check cannot be completed, state precisely which element remains unverified rather than marking the entire record as confirmed.

Worked case 9: Form estimate

Imagine that a reader must translate the source fragment Possible jug rim in a document about archaeological pottery recording sheets and assemblage descriptions. A plausible but faulty target wording would be “certainly a complete jug.” The problem is not necessarily grammar. It is the underlying interpretation: the source’s classification is provisional. A reader relying on that faulty wording could make a different choice while believing the instruction had been preserved.

A safer reader-facing rendering is “rim possibly from a jug.” Keep the original fragment beside the explanation wherever exact notation, codes or numerical labels matter. The translator should not silently add implied details, change an uncertainty into certainty, or use a supposedly simpler word that conceals the distinction. In this case, the decisive checking action is to check presence of ‘possible’. If the check cannot be completed, state precisely which element remains unverified rather than marking the entire record as confirmed.

Worked case 10: Rim diameter

Imagine that a reader must translate the source fragment Estimated rim dia. 18 cm in a document about archaeological pottery recording sheets and assemblage descriptions. A plausible but faulty target wording would be “sherd length exactly 18 cm.” The problem is not necessarily grammar. It is the underlying interpretation: a reconstructed vessel measurement is not the surviving fragment length. A reader relying on that faulty wording could make a different choice while believing the instruction had been preserved.

A safer reader-facing rendering is “estimated original rim diameter eighteen centimetres.” Keep the original fragment beside the explanation wherever exact notation, codes or numerical labels matter. The translator should not silently add implied details, change an uncertainty into certainty, or use a supposedly simpler word that conceals the distinction. In this case, the decisive checking action is to retain estimate label. If the check cannot be completed, state precisely which element remains unverified rather than marking the entire record as confirmed.

Worked case 11: Fabric date

Imagine that a reader must translate the source fragment Ware dated c. AD 900–1100 in a document about archaeological pottery recording sheets and assemblage descriptions. A plausible but faulty target wording would be “context deposited in AD 1000.” The problem is not necessarily grammar. It is the underlying interpretation: typological range does not directly fix the context’s depositional date. A reader relying on that faulty wording could make a different choice while believing the instruction had been preserved.

A safer reader-facing rendering is “ware type broadly dated around AD 900 to 1100.” Keep the original fragment beside the explanation wherever exact notation, codes or numerical labels matter. The translator should not silently add implied details, change an uncertainty into certainty, or use a supposedly simpler word that conceals the distinction. In this case, the decisive checking action is to separate ware and context dates. If the check cannot be completed, state precisely which element remains unverified rather than marking the entire record as confirmed.

Worked case 12: EVE value

Imagine that a reader must translate the source fragment Rim EVE 0.25 in a document about archaeological pottery recording sheets and assemblage descriptions. A plausible but faulty target wording would be “a quarter of a complete pot existed.” The problem is not necessarily grammar. It is the underlying interpretation: a quantitative reconstruction is not literal fraction of one observed complete pot. A reader relying on that faulty wording could make a different choice while believing the instruction had been preserved.

A safer reader-facing rendering is “rim-based estimated vessel equivalent value 0.25.” Keep the original fragment beside the explanation wherever exact notation, codes or numerical labels matter. The translator should not silently add implied details, change an uncertainty into certainty, or use a supposedly simpler word that conceals the distinction. In this case, the decisive checking action is to explain measure. If the check cannot be completed, state precisely which element remains unverified rather than marking the entire record as confirmed.

Worked case 13: Minimum count

Imagine that a reader must translate the source fragment MNV at least 3 in a document about archaeological pottery recording sheets and assemblage descriptions. A plausible but faulty target wording would be “exactly three original vessels.” The problem is not necessarily grammar. It is the underlying interpretation: a lower bound cannot be changed into a precise total. A reader relying on that faulty wording could make a different choice while believing the instruction had been preserved.

A safer reader-facing rendering is “minimum number of vessels estimated as three.” Keep the original fragment beside the explanation wherever exact notation, codes or numerical labels matter. The translator should not silently add implied details, change an uncertainty into certainty, or use a supposedly simpler word that conceals the distinction. In this case, the decisive checking action is to preserve at least. If the check cannot be completed, state precisely which element remains unverified rather than marking the entire record as confirmed.

Worked case 14: Residual note

Imagine that a reader must translate the source fragment Residual Roman sherd? in a document about archaeological pottery recording sheets and assemblage descriptions. A plausible but faulty target wording would be “definitely in situ Roman activity.” The problem is not necessarily grammar. It is the underlying interpretation: the question mark qualifies depositional interpretation. A reader relying on that faulty wording could make a different choice while believing the instruction had been preserved.

A safer reader-facing rendering is “possibly residual Roman-period sherd.” Keep the original fragment beside the explanation wherever exact notation, codes or numerical labels matter. The translator should not silently add implied details, change an uncertainty into certainty, or use a supposedly simpler word that conceals the distinction. In this case, the decisive checking action is to record source doubt. If the check cannot be completed, state precisely which element remains unverified rather than marking the entire record as confirmed.

Worked case 15: Cross-context join

Imagine that a reader must translate the source fragment Join: CTX 214 / 219 in a document about archaeological pottery recording sheets and assemblage descriptions. A plausible but faulty target wording would be “same findspot 214.” The problem is not necessarily grammar. It is the underlying interpretation: a physical join does not erase separate recovery contexts. A reader relying on that faulty wording could make a different choice while believing the instruction had been preserved.

A safer reader-facing rendering is “joining fragments recorded from contexts 214 and 219.” Keep the original fragment beside the explanation wherever exact notation, codes or numerical labels matter. The translator should not silently add implied details, change an uncertainty into certainty, or use a supposedly simpler word that conceals the distinction. In this case, the decisive checking action is to retain both numbers. If the check cannot be completed, state precisely which element remains unverified rather than marking the entire record as confirmed.

Worked case 16: Abraded edges

Imagine that a reader must translate the source fragment Very abraded in a document about archaeological pottery recording sheets and assemblage descriptions. A plausible but faulty target wording would be “freshly broken.” The problem is not necessarily grammar. It is the underlying interpretation: condition can inform depositional history but is not automatically a date. A reader relying on that faulty wording could make a different choice while believing the instruction had been preserved.

A safer reader-facing rendering is “surfaces or edges heavily abraded.” Keep the original fragment beside the explanation wherever exact notation, codes or numerical labels matter. The translator should not silently add implied details, change an uncertainty into certainty, or use a supposedly simpler word that conceals the distinction. In this case, the decisive checking action is to compare condition category. If the check cannot be completed, state precisely which element remains unverified rather than marking the entire record as confirmed.

Worked case 17: Sooting

Imagine that a reader must translate the source fragment External sooting in a document about archaeological pottery recording sheets and assemblage descriptions. A plausible but faulty target wording would be “black glaze applied outside.” The problem is not necessarily grammar. It is the underlying interpretation: use-related deposits and intentional treatment are different possibilities. A reader relying on that faulty wording could make a different choice while believing the instruction had been preserved.

A safer reader-facing rendering is “sooting observed on outer surface.” Keep the original fragment beside the explanation wherever exact notation, codes or numerical labels matter. The translator should not silently add implied details, change an uncertainty into certainty, or use a supposedly simpler word that conceals the distinction. In this case, the decisive checking action is to distinguish observation from cause. If the check cannot be completed, state precisely which element remains unverified rather than marking the entire record as confirmed.

Worked case 18: Wall thickness

Imagine that a reader must translate the source fragment Wall 6–7 mm in a document about archaeological pottery recording sheets and assemblage descriptions. A plausible but faulty target wording would be “precisely seven centimetres.” The problem is not necessarily grammar. It is the underlying interpretation: both units and ranges are integral to the measurement. A reader relying on that faulty wording could make a different choice while believing the instruction had been preserved.

A safer reader-facing rendering is “wall thickness measured in the six-to-seven-millimetre range.” Keep the original fragment beside the explanation wherever exact notation, codes or numerical labels matter. The translator should not silently add implied details, change an uncertainty into certainty, or use a supposedly simpler word that conceals the distinction. In this case, the decisive checking action is to check mm. If the check cannot be completed, state precisely which element remains unverified rather than marking the entire record as confirmed.

Worked case 19: Sampling

Imagine that a reader must translate the source fragment Selected for petrography in a document about archaeological pottery recording sheets and assemblage descriptions. A plausible but faulty target wording would be “petrography proved origin.” The problem is not necessarily grammar. It is the underlying interpretation: selection is a plan, not an analytical result. A reader relying on that faulty wording could make a different choice while believing the instruction had been preserved.

A safer reader-facing rendering is “selected for future petrographic study.” Keep the original fragment beside the explanation wherever exact notation, codes or numerical labels matter. The translator should not silently add implied details, change an uncertainty into certainty, or use a supposedly simpler word that conceals the distinction. In this case, the decisive checking action is to keep status pending. If the check cannot be completed, state precisely which element remains unverified rather than marking the entire record as confirmed.

Worked case 20: Diagnostic count

Imagine that a reader must translate the source fragment 5 diagnostic of 32 sherds in a document about archaeological pottery recording sheets and assemblage descriptions. A plausible but faulty target wording would be “32 diagnostic fragments.” The problem is not necessarily grammar. It is the underlying interpretation: the numerator and denominator describe different subsets. A reader relying on that faulty wording could make a different choice while believing the instruction had been preserved.

A safer reader-facing rendering is “five of thirty-two sherds carry diagnostic features.” Keep the original fragment beside the explanation wherever exact notation, codes or numerical labels matter. The translator should not silently add implied details, change an uncertainty into certainty, or use a supposedly simpler word that conceals the distinction. In this case, the decisive checking action is to compare record totals. If the check cannot be completed, state precisely which element remains unverified rather than marking the entire record as confirmed.

Worked case 21: Illustration number

Imagine that a reader must translate the source fragment Drawing P-17 in a document about archaeological pottery recording sheets and assemblage descriptions. A plausible but faulty target wording would be “object inventory P-17.” The problem is not necessarily grammar. It is the underlying interpretation: an illustration ID is not necessarily the artefact’s accession code. A reader relying on that faulty wording could make a different choice while believing the instruction had been preserved.

A safer reader-facing rendering is “associated illustration reference P-17.” Keep the original fragment beside the explanation wherever exact notation, codes or numerical labels matter. The translator should not silently add implied details, change an uncertainty into certainty, or use a supposedly simpler word that conceals the distinction. In this case, the decisive checking action is to inspect table heading. If the check cannot be completed, state precisely which element remains unverified rather than marking the entire record as confirmed.

Worked case 22: Provisional fabric

Imagine that a reader must translate the source fragment Fabric uncertain: Q2/Q3 in a document about archaeological pottery recording sheets and assemblage descriptions. A plausible but faulty target wording would be “fabric confirmed as Q2.” The problem is not necessarily grammar. It is the underlying interpretation: a choice must not be made without reviewer evidence. A reader relying on that faulty wording could make a different choice while believing the instruction had been preserved.

A safer reader-facing rendering is “fabric classification remains between Q2 and Q3.” Keep the original fragment beside the explanation wherever exact notation, codes or numerical labels matter. The translator should not silently add implied details, change an uncertainty into certainty, or use a supposedly simpler word that conceals the distinction. In this case, the decisive checking action is to retain both candidates. If the check cannot be completed, state precisely which element remains unverified rather than marking the entire record as confirmed.

Worked case 23: Fresh break

Imagine that a reader must translate the source fragment Two sherds joined along fresh break in a document about archaeological pottery recording sheets and assemblage descriptions. A plausible but faulty target wording would be “two pots manufactured together.” The problem is not necessarily grammar. It is the underlying interpretation: the join observation is not a statement about how many original vessels existed. A reader relying on that faulty wording could make a different choice while believing the instruction had been preserved.

A safer reader-facing rendering is “two fragments meet along a fresh fracture.” Keep the original fragment beside the explanation wherever exact notation, codes or numerical labels matter. The translator should not silently add implied details, change an uncertainty into certainty, or use a supposedly simpler word that conceals the distinction. In this case, the decisive checking action is to report break status. If the check cannot be completed, state precisely which element remains unverified rather than marking the entire record as confirmed.

Worked case 24: Not excavated

Imagine that a reader must translate the source fragment Deposit sampled, not fully excavated in a document about archaeological pottery recording sheets and assemblage descriptions. A plausible but faulty target wording would be “the entire layer was examined.” The problem is not necessarily grammar. It is the underlying interpretation: coverage limits what assemblage counts can prove. A reader relying on that faulty wording could make a different choice while believing the instruction had been preserved.

A safer reader-facing rendering is “only a sample of the deposit was excavated.” Keep the original fragment beside the explanation wherever exact notation, codes or numerical labels matter. The translator should not silently add implied details, change an uncertainty into certainty, or use a supposedly simpler word that conceals the distinction. In this case, the decisive checking action is to preserve sampling condition. If the check cannot be completed, state precisely which element remains unverified rather than marking the entire record as confirmed.

Failure modes and how to repair them

Failure mode 1: Literal word substitution

A normal dictionary replacement changes the operation rather than explaining it. Read the relevant field definition or illustrated procedure and translate its function; never choose the most familiar everyday meaning by reflex. In archaeological pottery recording sheets and assemblage descriptions, the most effective repair is to identify the affected token or action, compare it with the original and write a concise correction note. Correct the particular mistake rather than rewriting an entire paragraph that was otherwise accurate. Finally repeat the miniature source-to-target test so the amended version demonstrably restores the intended distinction.

Failure mode 2: Inconsistent names

One field or operation acquires several target-language names. Create a term map and run a consistency pass, keeping unique identifiers alongside any reader-friendly description. In archaeological pottery recording sheets and assemblage descriptions, the most effective repair is to identify the affected token or action, compare it with the original and write a concise correction note. Correct the particular mistake rather than rewriting an entire paragraph that was otherwise accurate. Finally repeat the miniature source-to-target test so the amended version demonstrably restores the intended distinction.

Failure mode 3: Unannounced conversion

Units, notation, dates or numeric forms shift without a declared basis. Keep the original alongside any clearly requested conversion and ask what precision, standard and locale the recipient requires. In archaeological pottery recording sheets and assemblage descriptions, the most effective repair is to identify the affected token or action, compare it with the original and write a concise correction note. Correct the particular mistake rather than rewriting an entire paragraph that was otherwise accurate. Finally repeat the miniature source-to-target test so the amended version demonstrably restores the intended distinction.

Failure mode 4: Reversed sequence

Two otherwise correct steps are delivered in the wrong order. Number the source steps, map every transition and verify each prerequisite before allowing the next instruction to run. In archaeological pottery recording sheets and assemblage descriptions, the most effective repair is to identify the affected token or action, compare it with the original and write a concise correction note. Correct the particular mistake rather than rewriting an entire paragraph that was otherwise accurate. Finally repeat the miniature source-to-target test so the amended version demonstrably restores the intended distinction.

Failure mode 5: Missing condition

Words such as only, unless, if, optional or after disappear. Highlight every condition and separately test the branch in which the condition is false; a conditional instruction is not a universal command. In archaeological pottery recording sheets and assemblage descriptions, the most effective repair is to identify the affected token or action, compare it with the original and write a concise correction note. Correct the particular mistake rather than rewriting an entire paragraph that was otherwise accurate. Finally repeat the miniature source-to-target test so the amended version demonstrably restores the intended distinction.

Failure mode 6: An attractive inference

The translator fills a gap with a plausible but unsupported detail. Use a visible ‘not specified’ note rather than presenting the inferred detail as a fact from the source. In archaeological pottery recording sheets and assemblage descriptions, the most effective repair is to identify the affected token or action, compare it with the original and write a concise correction note. Correct the particular mistake rather than rewriting an entire paragraph that was otherwise accurate. Finally repeat the miniature source-to-target test so the amended version demonstrably restores the intended distinction.

Failure mode 7: Overnormalised language

A project-specific code is rewritten to resemble a more familiar national convention. Keep the source code verbatim until a verified project mapping explicitly authorises a change. In archaeological pottery recording sheets and assemblage descriptions, the most effective repair is to identify the affected token or action, compare it with the original and write a concise correction note. Correct the particular mistake rather than rewriting an entire paragraph that was otherwise accurate. Finally repeat the miniature source-to-target test so the amended version demonstrably restores the intended distinction.

Failure mode 8: Diagram–text disagreement

The prose says one thing and the numbered image or diagram shows another. Flag the conflict with the source owner; do not secretly edit the diagram or choose the version you like. In archaeological pottery recording sheets and assemblage descriptions, the most effective repair is to identify the affected token or action, compare it with the original and write a concise correction note. Correct the particular mistake rather than rewriting an entire paragraph that was otherwise accurate. Finally repeat the miniature source-to-target test so the amended version demonstrably restores the intended distinction.

Failure mode 9: No reader test

The translated passage is approved because it sounds natural to an editor. Ask a fresh reader to perform a miniature sample and document which phrase led them to an unintended result. In archaeological pottery recording sheets and assemblage descriptions, the most effective repair is to identify the affected token or action, compare it with the original and write a concise correction note. Correct the particular mistake rather than rewriting an entire paragraph that was otherwise accurate. Finally repeat the miniature source-to-target test so the amended version demonstrably restores the intended distinction.

Failure mode 10: No revision record

A correction enters the final document without showing what changed. Record source version, translation version, revised terms and reviewer decisions so later users can reconstruct the change. In archaeological pottery recording sheets and assemblage descriptions, the most effective repair is to identify the affected token or action, compare it with the original and write a concise correction note. Correct the particular mistake rather than rewriting an entire paragraph that was otherwise accurate. Finally repeat the miniature source-to-target test so the amended version demonstrably restores the intended distinction.

Further demonstrations: a better way to think and teach

Build an evidence table before prose

For each row, separate a source identifier, observed material properties, quantities, comparative classification and inferred date. Every target-language sentence should be traceable to one or more of these fields. Never let a stylistic summary overwrite an unknown value in the original form.

Turn this idea into a classroom task: give two learners the same short source and ask them to annotate protected elements independently before discussing wording. Compare their chosen renderings only after they have agreed which source relationships must survive. Where they disagree, ask for the actual page, key, illustration or reference that decides the matter. This encourages evidence-based discussion rather than an argument over which sentence merely sounds more natural.

A two-context exercise

Invent two contextual units with different pottery counts and a reported cross-fit. Ask learners to translate the source twice: once as a table and once as explanatory prose. Compare whether both versions retain the separate provenances, the exact counts and the provisional nature of the join interpretation.

What this article does not own

This guide deals with translating archaeological pottery sherd data and its epistemic limits. Museum accession labels, fossil specimen geologic ages, and general excavation context or stratigraphy records are related but different canonical jobs.

A four-week practice and transfer programme

Week one: read and label

Collect six short examples of archaeological pottery recording sheets and assemblage descriptions. Identify identifiers, ordinary words, conditional instructions and unknowns. Build a glossary containing both the target wording and the role of each concept. Use a pencil or digital annotation rather than altering the source record. Your weekly test is to explain five distinctions to a new learner without looking at a dictionary. Record which explanations required source consultation.

For additional transfer, explain one of this week’s corrections to a reader with no specialist background. Avoid oversimplifying the protected meaning. Ask the reader to identify the action, the numerical or coded constraint and the remaining uncertainty in their own words. If they cannot do so, improve the explanation and test again. A lesson is complete when clarity and fidelity coexist.

Week two: translate and reverse-check

Translate eight small examples with all original protected tokens visible. Then set the source aside, ask another person to reconstruct the intended action from the translation and compare that reconstruction with the original. Mark every divergence as an omission, reversal, overstatement, unit error, actor confusion or term inconsistency. Revise only after the evidence is clear.

Week three: simulate and audit

Choose four medium-length examples and test them through a reversible miniature simulation, an appropriate viewer or a controlled mock record. Create an audit sheet with columns for element, source value, target value, test and reviewer comment. Explain which parts require subject expertise. A failed test is useful evidence because it identifies exactly where a fluent translation misled the reader.

Week four: publish a miniature guide

Create a short bilingual teaching sheet about archaeological pottery recording sheets and assemblage descriptions, with three examples, three common errors, a tiny glossary and an explicit source section. Ask a beginner to use it unaided. Revise the hardest instruction and document the change. Your goal is not the fastest translation; it is a method that another person can reproduce and verify without relying on personal trust.

Frequently asked questions

Is a context number an artefact number?

Not necessarily. It identifies an excavation unit, whereas artefacts may also have bag, find or accession identifiers. The practical rule is to keep excavation context, local fabric typology, count, weight and inference traceable to the specific source passage and to distinguish a demonstrated fact from a translator’s explanatory note. If the answer remains dependent on the source system, show that dependency instead of guessing.

Can I translate a fabric code into a place name?

Only if an authoritative project key supports that mapping; otherwise preserve the code. The practical rule is to keep excavation context, local fabric typology, count, weight and inference traceable to the specific source passage and to distinguish a demonstrated fact from a translator’s explanatory note. If the answer remains dependent on the source system, show that dependency instead of guessing.

Does sherd count equal vessel count?

No. Several sherds may derive from one vessel, and separate vessels may be represented by similar fragments. The practical rule is to keep excavation context, local fabric typology, count, weight and inference traceable to the specific source passage and to distinguish a demonstrated fact from a translator’s explanatory note. If the answer remains dependent on the source system, show that dependency instead of guessing.

What does EVE mean?

Estimated Vessel Equivalent is one way of quantifying pottery, often using a measurable surviving portion such as rim circumference. The practical rule is to keep excavation context, local fabric typology, count, weight and inference traceable to the specific source passage and to distinguish a demonstrated fact from a translator’s explanatory note. If the answer remains dependent on the source system, show that dependency instead of guessing.

Can a sherd date a context precisely?

Usually not on its own; fabric dates and depositional histories require contextual interpretation. The practical rule is to keep excavation context, local fabric typology, count, weight and inference traceable to the specific source passage and to distinguish a demonstrated fact from a translator’s explanatory note. If the answer remains dependent on the source system, show that dependency instead of guessing.

Is abrasion proof of age?

No. It is an observed condition that may have multiple causes. The practical rule is to keep excavation context, local fabric typology, count, weight and inference traceable to the specific source passage and to distinguish a demonstrated fact from a translator’s explanatory note. If the answer remains dependent on the source system, show that dependency instead of guessing.

Should I remove question marks?

No. They often record classification uncertainty and should remain legible. The practical rule is to keep excavation context, local fabric typology, count, weight and inference traceable to the specific source passage and to distinguish a demonstrated fact from a translator’s explanatory note. If the answer remains dependent on the source system, show that dependency instead of guessing.

What if two contexts share joining sherds?

Retain both recorded contexts and explain the observed join without inventing a single original findspot. The practical rule is to keep excavation context, local fabric typology, count, weight and inference traceable to the specific source passage and to distinguish a demonstrated fact from a translator’s explanatory note. If the answer remains dependent on the source system, show that dependency instead of guessing.

How do I handle a missing mass?

Mark it not recorded or unknown, rather than inserting a plausible estimate. The practical rule is to keep excavation context, local fabric typology, count, weight and inference traceable to the specific source passage and to distinguish a demonstrated fact from a translator’s explanatory note. If the answer remains dependent on the source system, show that dependency instead of guessing.

Are Roman and medieval terms worldwide categories?

They are context-dependent historical labels; explain the chronology rather than transplanting them uncritically to unrelated regions. The practical rule is to keep excavation context, local fabric typology, count, weight and inference traceable to the specific source passage and to distinguish a demonstrated fact from a translator’s explanatory note. If the answer remains dependent on the source system, show that dependency instead of guessing.

Do I need a pottery specialist?

Yes for unresolved fabric, form or dating interpretation; language editors should not manufacture archaeological certainty. The practical rule is to keep excavation context, local fabric typology, count, weight and inference traceable to the specific source passage and to distinguish a demonstrated fact from a translator’s explanatory note. If the answer remains dependent on the source system, show that dependency instead of guessing.

What is the final record test?

Verify context IDs, fabric codes, sherd type, count, mass, date qualifier and any cross-context relationship against the source table. The practical rule is to keep excavation context, local fabric typology, count, weight and inference traceable to the specific source passage and to distinguish a demonstrated fact from a translator’s explanatory note. If the answer remains dependent on the source system, show that dependency instead of guessing.

Quick review checklist

Before using the translation, verify the exact original source and version; its intended audience; all identifiers and units; the distinction between a measurement, instruction, opinion and uncertain inference; ordered actions; directional or visual relationships; every conditional phrase; the complete glossary; specialist questions; and the final miniature test. A useful final question is: Would a careful reader following only the translated version make the same choice as a reader following the source? If not, the work is not finished.

References and careful further reading

For domain-specific terms, consult the primary reference and the supplementary reference. For general search-facing writing, Google Search Central’s people-first content guidance encourages genuinely useful explanations, source transparency and original value rather than mechanical keyword repetition. Sources support the named technical concepts; all teaching mini-records above are illustrative and should not be mistaken for authentic historical or production data.

Your next useful route

Continue with the original eduKateSG translation foundation: Translate meaning, not just words; Context, tone and register; Idioms, culture and non-literal language; Terminology, glossaries and quality checks. For stronger vocabulary distinctions, visit the Vocabulary Learning Hub. For meaning, sentence structure and communication, use How English Works. The Top Ways to Translate Correctly series index leads to adjacent specialist topics without treating them as replacements for the current lesson.

The central lesson is simple: translation is correct when the next reader can recover the same information and act on it without an invented assumption. The task for this page is narrow and concrete—archaeological pottery recording sheets and assemblage descriptions—and its examples demonstrate the boundaries of that job. Keep the source, preserve its relationships, explain what the symbols do and check the target against reality wherever a safe controlled test is possible.

Top Ways to Translate Correctly | Translate Archaeological Pottery Sherd Records, Fabric Codes and Context Numbers Without Inventing Provenance