Top Ways to Translate Correctly | Translate Genealogy DNA Matches, Shared cM and Relationship Estimates Without Claiming Kinship

To translate genealogy DNA match reports correctly, preserve the amount of shared DNA, the testing platform’s relationship estimate, every uncertainty label and the distinction between a genetic match and a proven family connection. Genetic genealogy translation is not a word-for-word exercise: centimorgans (cM), shared DNA segments, match lists, predicted cousin relationships and family-tree notes describe different types of evidence. A fluent translation becomes misleading the moment it turns a possible third cousin into an identified relative or mistakes a software estimate for a biological finding.

This guide explains how to translate DNA ancestry match reports, shared centimorgan values, relationship predictions, chromosome-browser notes, maternal or paternal side labels, half relationships, removed cousins, private family trees and match correspondence. It is written for multilingual families, genealogy researchers, students, community archivists and translators who must communicate evidence without exaggerating it. All names, kit labels and figures used in worked examples below are hypothetical teaching examples, not real genetic reports.

The safe answer to ‘what does my DNA match mean?’ begins with three separate statements. First, the testing service reports a similarity measurement. Second, its system may suggest several possible family relationships. Third, documentary research, consent and independent corroboration are needed before anyone responsibly names a specific ancestor or living relationship. Translation must keep those statements separate even when the target language invites a shorter, warmer, more definite family term.

The practical goal is straightforward: after reading the translated report, a person should be able to identify the same measurement, the same limitations, the same unresolved questions and the same next research action as the reader of the source report. If either reader walks away with a different claim about who is related to whom, the translation has changed the evidence.

The visible problem: matching DNA is not the same as naming a family member

A report might say ‘estimated third cousin, 93 cM across six segments.’ The numerical description is one thing. A predicted relationship category is a second. An established relationship supported by records is a third. Translating the line as ‘your third cousin has been found’ collapses all three into a statement that the report never made. This is an error of certainty, not merely terminology, and it can affect sensitive family conversations.

The difficulty increases when a multilingual family uses kinship words that encode relative age, gender, maternal or paternal side, marriage, generation and social obligations. A broad English ‘cousin’ can have several precise equivalents in another language. Selecting one that specifies the family side without evidence inserts a fact. The solution may require a descriptive phrase such as ‘a DNA match whose exact family relationship is not yet confirmed’ rather than a familiar family label.

Genetic genealogy is especially sensitive because records may involve adoption, assisted reproduction, previously unknown relatives or family histories people have not chosen to share. A translation should not convert uncertainty into gossip, infer parentage from a single number or expose a match’s private identifying details. Accuracy includes what we refuse to add.

First principles: twelve distinctions that keep DNA translation honest

Measurement versus identity

A centimorgan is a unit used to describe genetic linkage, not a person’s name, legal relationship or percentage of an estate. Keep the unit and the original numeric value intact. If an explanation is needed, define the term separately rather than silently converting it into a cousin label. For a careful translation, first ask which claim the original sentence actually authorises. Next identify whether a number, relationship category, source label or condition must survive verbatim. An explanation may be added only when visibly distinguished from the record, so readers can separate source evidence from guidance. A 122 cM match can be described as sharing 122 cM; it cannot be described as a confirmed second cousin merely because that relationship appears high on an estimate chart. This distinction is particularly useful when a family member receives a brief screenshot without the report’s explanatory legend.

Reported value versus model interpretation

Testing services use their own algorithms, thresholds and relationship models. A displayed match total and a vendor’s possible relationship are separate fields with separate evidential status. For a careful translation, first ask which claim the original sentence actually authorises. Next identify whether a number, relationship category, source label or condition must survive verbatim. An explanation may be added only when visibly distinguished from the record, so readers can separate source evidence from guidance. Render ‘estimated second to fourth cousin’ as a range; do not select one relationship as the final answer for ease of reading. This distinction is particularly useful when a family member receives a brief screenshot without the report’s explanatory legend.

Total sharing versus longest segment

A total across several segments differs from the largest individual segment. Translate the exact measurement labels rather than treating every cM number as one interchangeable quantity. For a careful translation, first ask which claim the original sentence actually authorises. Next identify whether a number, relationship category, source label or condition must survive verbatim. An explanation may be added only when visibly distinguished from the record, so readers can separate source evidence from guidance. ‘Total 89 cM; longest 28 cM’ is not ‘one segment of 89 cM’. This distinction is particularly useful when a family member receives a brief screenshot without the report’s explanatory legend.

Autosomal versus lineage-specific testing

Autosomal DNA matching, Y-DNA testing and mitochondrial DNA testing can address different genealogical questions. Do not describe a lineage marker as a complete family-tree reconstruction. For a careful translation, first ask which claim the original sentence actually authorises. Next identify whether a number, relationship category, source label or condition must survive verbatim. An explanation may be added only when visibly distinguished from the record, so readers can separate source evidence from guidance. A shared paternal-line marker is not, by itself, evidence that two individuals share a particular recently named grandfather. This distinction is particularly useful when a family member receives a brief screenshot without the report’s explanatory legend.

Possible relationship versus proved relationship

Estimated relationship labels may describe several plausible paths through a family tree. Their ambiguity is evidence to preserve, not noise to remove. For a careful translation, first ask which claim the original sentence actually authorises. Next identify whether a number, relationship category, source label or condition must survive verbatim. An explanation may be added only when visibly distinguished from the record, so readers can separate source evidence from guidance. ‘Half first cousin or first cousin once removed’ is not an editorial invitation to choose the more familiar wording. This distinction is particularly useful when a family member receives a brief screenshot without the report’s explanatory legend.

Degree and removal

A cousin degree and the number of generations removed are distinct. Some languages need a short descriptive phrase to reproduce a term such as 2C1R accurately. For a careful translation, first ask which claim the original sentence actually authorises. Next identify whether a number, relationship category, source label or condition must survive verbatim. An explanation may be added only when visibly distinguished from the record, so readers can separate source evidence from guidance. Explain that second cousin once removed indicates a generation difference; do not translate it as ‘second cousin’ alone. This distinction is particularly useful when a family member receives a brief screenshot without the report’s explanatory legend.

Half relationship versus full relationship

Half sibling, half aunt and half cousin labels reflect different genealogical structures from full relationships. Deleting ‘half’ changes the proposed parentage path. For a careful translation, first ask which claim the original sentence actually authorises. Next identify whether a number, relationship category, source label or condition must survive verbatim. An explanation may be added only when visibly distinguished from the record, so readers can separate source evidence from guidance. A match predicted as ‘half sibling’ must not be translated as ‘sibling confirmed’; retain both the qualifier and uncertainty. This distinction is particularly useful when a family member receives a brief screenshot without the report’s explanatory legend.

Side assignment versus confirmed lineage

A label such as ‘paternal side’ may be based on user tagging, a parental test, or an inference from shared matches. Its evidence source matters. For a careful translation, first ask which claim the original sentence actually authorises. Next identify whether a number, relationship category, source label or condition must survive verbatim. An explanation may be added only when visibly distinguished from the record, so readers can separate source evidence from guidance. If a report says ‘likely paternal’ based on clustering, preserve ‘likely’ and identify the algorithmic source when available. This distinction is particularly useful when a family member receives a brief screenshot without the report’s explanatory legend.

Shared matches versus triangular proof

Two people appearing in the same shared-match list does not on its own establish one common ancestor or demonstrate triangulation on the same chromosome segment. For a careful translation, first ask which claim the original sentence actually authorises. Next identify whether a number, relationship category, source label or condition must survive verbatim. An explanation may be added only when visibly distinguished from the record, so readers can separate source evidence from guidance. Do not translate ‘shared match’ as ‘confirmed shared ancestral branch’. This distinction is particularly useful when a family member receives a brief screenshot without the report’s explanatory legend.

Ethnicity estimates versus family ties

An ethnicity estimate summarises a model of genetic similarity to reference populations; it does not identify a specific living relative or prove nationality. For a careful translation, first ask which claim the original sentence actually authorises. Next identify whether a number, relationship category, source label or condition must survive verbatim. An explanation may be added only when visibly distinguished from the record, so readers can separate source evidence from guidance. A predicted population percentage cannot become evidence of a particular named grandparent or civil status. This distinction is particularly useful when a family member receives a brief screenshot without the report’s explanatory legend.

Platform interface versus researcher note

Match-list buttons, user-supplied tree comments and system-generated suggestions have different authors. Keep their attributions visible. For a careful translation, first ask which claim the original sentence actually authorises. Next identify whether a number, relationship category, source label or condition must survive verbatim. An explanation may be added only when visibly distinguished from the record, so readers can separate source evidence from guidance. ‘Potential connection’ entered by a user must not be represented as a laboratory-confirmed finding. This distinction is particularly useful when a family member receives a brief screenshot without the report’s explanatory legend.

Privacy and informed communication

Genetic test and family information can be personally sensitive. A translation preserves access controls, contact preferences and consent restrictions rather than treating relatives as public records. For a careful translation, first ask which claim the original sentence actually authorises. Next identify whether a number, relationship category, source label or condition must survive verbatim. An explanation may be added only when visibly distinguished from the record, so readers can separate source evidence from guidance. ‘Private tree’ means restricted access; it is not an invitation to reconstruct and publish another person’s family history. This distinction is particularly useful when a family member receives a brief screenshot without the report’s explanatory legend.

The twelve-stage method for translating a genetic genealogy report

1. Identify the report and its author

Record whether the item comes from a DNA testing dashboard, third-party relationship chart, uploaded spreadsheet, match’s message or researcher’s annotated tree. Do not let a handwritten hypothesis acquire the authority of a testing laboratory simply because both appear in one screenshot. In practice, write the relevant source expression in a small evidence table beside its target rendering, not in an isolated list of vocabulary. Include a column titled ‘what this does not establish’ and a column for ‘verification needed’. This habit reveals whether the draft has accidentally added a parent, a chromosome, a surname or a probability that nobody supplied. It also gives a second reviewer a concrete reason to approve or question the translation.

2. Locate the reporting date and version

Preserve the date of the displayed estimate and distinguish it from the original sample collection date or the latest message timestamp. A report viewed in October may use an updated matching algorithm despite a much older test. In practice, write the relevant source expression in a small evidence table beside its target rendering, not in an isolated list of vocabulary. Include a column titled ‘what this does not establish’ and a column for ‘verification needed’. This habit reveals whether the draft has accidentally added a parent, a chromosome, a surname or a probability that nobody supplied. It also gives a second reviewer a concrete reason to approve or question the translation.

3. Freeze all identifiers before translating

Treat kit IDs, chromosome numbers, sample labels, segment coordinates, cM values, scientific acronyms and account aliases as protected data. The characters in a kit identifier can carry identity; translating them as ordinary vocabulary destroys the cross-reference. In practice, write the relevant source expression in a small evidence table beside its target rendering, not in an isolated list of vocabulary. Include a column titled ‘what this does not establish’ and a column for ‘verification needed’. This habit reveals whether the draft has accidentally added a parent, a chromosome, a surname or a probability that nobody supplied. It also gives a second reviewer a concrete reason to approve or question the translation.

4. Read the whole relationship label

Identify whether the source gives one suggested relationship, a range, a list of alternatives, or a probability model with a disclaimer. A range such as ‘second–third cousin’ cannot be represented as one point estimate. In practice, write the relevant source expression in a small evidence table beside its target rendering, not in an isolated list of vocabulary. Include a column titled ‘what this does not establish’ and a column for ‘verification needed’. This habit reveals whether the draft has accidentally added a parent, a chromosome, a surname or a probability that nobody supplied. It also gives a second reviewer a concrete reason to approve or question the translation.

5. Check the exact metric name

Determine whether the number is total shared cM, longest segment cM, count of segments, match percentage or an inferred range. Retaining every number while swapping its label is still a factual translation error. In practice, write the relevant source expression in a small evidence table beside its target rendering, not in an isolated list of vocabulary. Include a column titled ‘what this does not establish’ and a column for ‘verification needed’. This habit reveals whether the draft has accidentally added a parent, a chromosome, a surname or a probability that nobody supplied. It also gives a second reviewer a concrete reason to approve or question the translation.

6. Separate relatives from research hypotheses

Mark which family-tree connections are documented, which are based on a user tree and which are speculative links. The verb ‘appears’ in an informal note does not support the categorical verb ‘is’ in the target language. In practice, write the relevant source expression in a small evidence table beside its target rendering, not in an isolated list of vocabulary. Include a column titled ‘what this does not establish’ and a column for ‘verification needed’. This habit reveals whether the draft has accidentally added a parent, a chromosome, a surname or a probability that nobody supplied. It also gives a second reviewer a concrete reason to approve or question the translation.

7. Build a kinship equivalence note

For kinship terms without a one-word match, explain degree, generational removal, half relationship or parental side without adding gender. A descriptive paraphrase is more accurate than a traditional family title that specifies an unverified relationship. In practice, write the relevant source expression in a small evidence table beside its target rendering, not in an isolated list of vocabulary. Include a column titled ‘what this does not establish’ and a column for ‘verification needed’. This habit reveals whether the draft has accidentally added a parent, a chromosome, a surname or a probability that nobody supplied. It also gives a second reviewer a concrete reason to approve or question the translation.

8. Preserve modality and uncertainty

Track words such as estimated, possible, predicted, may, likely, approximately, unconfirmed, disputed, inferred and suggested. In sensitive findings, a single omitted hedge can change a cautious family inquiry into a definitive public claim. In practice, write the relevant source expression in a small evidence table beside its target rendering, not in an isolated list of vocabulary. Include a column titled ‘what this does not establish’ and a column for ‘verification needed’. This habit reveals whether the draft has accidentally added a parent, a chromosome, a surname or a probability that nobody supplied. It also gives a second reviewer a concrete reason to approve or question the translation.

9. Retain attribution and privacy markers

Keep distinct the platform’s estimate, the researcher’s belief and a match’s self-description; preserve private or restricted labels. A match declining contact has not given permission for an intermediary to circulate contact details. In practice, write the relevant source expression in a small evidence table beside its target rendering, not in an isolated list of vocabulary. Include a column titled ‘what this does not establish’ and a column for ‘verification needed’. This habit reveals whether the draft has accidentally added a parent, a chromosome, a surname or a probability that nobody supplied. It also gives a second reviewer a concrete reason to approve or question the translation.

10. Use two passes rather than one

Make a factual first pass protecting fields, then a language pass for readability and natural grammar, followed by a source-to-target comparison. The first pass catches structural errors; the second makes the correct meaning understandable to a real reader. In practice, write the relevant source expression in a small evidence table beside its target rendering, not in an isolated list of vocabulary. Include a column titled ‘what this does not establish’ and a column for ‘verification needed’. This habit reveals whether the draft has accidentally added a parent, a chromosome, a surname or a probability that nobody supplied. It also gives a second reviewer a concrete reason to approve or question the translation.

11. Test the translated report with questions

Ask a reviewer to locate the same total cM, segment count, uncertainty label, possible relationships, identified source and next action in both versions. If a reviewer confidently names a person from the translation but not from the source, the target has overclaimed. In practice, write the relevant source expression in a small evidence table beside its target rendering, not in an isolated list of vocabulary. Include a column titled ‘what this does not establish’ and a column for ‘verification needed’. This habit reveals whether the draft has accidentally added a parent, a chromosome, a surname or a probability that nobody supplied. It also gives a second reviewer a concrete reason to approve or question the translation.

12. Archive the decisions and unresolved items

Save a glossary, a redacted example, source version and notes on uncertain kinship equivalents. Do not store private genetic identifiers in an open teaching repository. A controlled record makes later updates possible without confusing older hypotheses with newly documented relationships. In practice, write the relevant source expression in a small evidence table beside its target rendering, not in an isolated list of vocabulary. Include a column titled ‘what this does not establish’ and a column for ‘verification needed’. This habit reveals whether the draft has accidentally added a parent, a chromosome, a surname or a probability that nobody supplied. It also gives a second reviewer a concrete reason to approve or question the translation.

Twenty-four worked translation cases: practise without inventing ancestors

Worked case 1: Total and longest segment

Imagine the following fictional source record: A fictional dashboard states ‘Shared DNA: 104 cM across 7 segments; longest segment 31 cM.’ A translator is asked to make it accessible to a family member who does not read the reporting language fluently. Before choosing family terms, identify whether the sentence is a measurement, an automated relationship suggestion, a research hypothesis, a privacy setting or an instruction. That classification protects the reader from treating all words in the document as equally established facts.

An unsafe rendering would suggest: ‘One shared segment of 104 cM.’ A more faithful English working interpretation is: ‘104 cM shared across seven reported segments; the longest reported segment is 31 cM.’ This preserves the number of segments and stops one metric from replacing another. This English gloss is an illustration of the decisions to preserve, not a prescribed equivalent in every target language. The actual target wording should be reviewed by someone who understands its kinship system and the source report’s technical conventions.

Ask a reviewer to point to each number and explain its corresponding unit and scope. The source sentence is deliberately short, which makes it tempting to fill its gaps with a plausible story. Instead, separate what the source actually reports from what a reader might want to know. A translation may be clear and readable without answering the unanswered question. The record should leave the reader with the same open question and the same permissible next action. For practice, produce two versions: a concise neutral caption for the screenshot and a longer explanation for a family member. Compare them for unsupported additions. If either version makes a hypothesis sound more certain than the original, revise the expression instead of adding another disclaimer at the bottom.

Worked case 2: Cousin relationship range

Imagine the following fictional source record: A service shows ‘Predicted relationship: second–fourth cousin’ with an information tooltip about overlapping possibilities. A translator is asked to make it accessible to a family member who does not read the reporting language fluently. Before choosing family terms, identify whether the sentence is a measurement, an automated relationship suggestion, a research hypothesis, a privacy setting or an instruction. That classification protects the reader from treating all words in the document as equally established facts.

An unsafe rendering would suggest: ‘This person is your second cousin.’ A more faithful English working interpretation is: ‘The service estimates a possible relationship in the second- to fourth-cousin range.’ A range is not a ranked list of confirmed relatives; the predicted label belongs to the service. This English gloss is an illustration of the decisions to preserve, not a prescribed equivalent in every target language. The actual target wording should be reviewed by someone who understands its kinship system and the source report’s technical conventions.

Identify the word marking prediction and see whether the target still leaves several alternatives open. Check both directions: every number and relationship in the source must survive in the target, and every person or certainty expressed in the target must have support in the source. A reviewer should be able to underline the exact words that warrant each claim. The record should leave the reader with the same open question and the same permissible next action. For practice, produce two versions: a concise neutral caption for the screenshot and a longer explanation for a family member. Compare them for unsupported additions. If either version makes a hypothesis sound more certain than the original, revise the expression instead of adding another disclaimer at the bottom.

Worked case 3: Third cousin example

Imagine the following fictional source record: A family email says ‘Maybe our grandparents’ families crossed in the same village; the 92 cM match is worth checking.’ A translator is asked to make it accessible to a family member who does not read the reporting language fluently. Before choosing family terms, identify whether the sentence is a measurement, an automated relationship suggestion, a research hypothesis, a privacy setting or an instruction. That classification protects the reader from treating all words in the document as equally established facts.

An unsafe rendering would suggest: ‘Our ancestors are from the same family in this village.’ A more faithful English working interpretation is: ‘A relative suggests investigating a possible ancestral connection; the 92 cM match is not yet assigned to a family branch.’ Locality and genetic similarity invite research but do not prove a genealogical line. This English gloss is an illustration of the decisions to preserve, not a prescribed equivalent in every target language. The actual target wording should be reviewed by someone who understands its kinship system and the source report’s technical conventions.

Have the translator distinguish the sender’s hypothesis from the dashboard’s measurement. Look for the smallest word that changes the claim: perhaps likely, possible, estimated, private, shared or unconfirmed. Losing that word can alter how someone approaches a relative or records a family history, even though the rest of the sentence remains fluent. The record should leave the reader with the same open question and the same permissible next action. For practice, produce two versions: a concise neutral caption for the screenshot and a longer explanation for a family member. Compare them for unsupported additions. If either version makes a hypothesis sound more certain than the original, revise the expression instead of adding another disclaimer at the bottom.

Worked case 4: Half-sibling possibility

Imagine the following fictional source record: A report presents ‘Close family: half sibling / aunt-uncle / grandparent’ as alternative relationships. A translator is asked to make it accessible to a family member who does not read the reporting language fluently. Before choosing family terms, identify whether the sentence is a measurement, an automated relationship suggestion, a research hypothesis, a privacy setting or an instruction. That classification protects the reader from treating all words in the document as equally established facts.

An unsafe rendering would suggest: ‘Confirmed half sibling.’ A more faithful English working interpretation is: ‘Close-family match, with several possible relationship categories listed by the service, including half sibling.’ Some close relationships overlap in shared DNA ranges; the translation cannot choose a family role. This English gloss is an illustration of the decisions to preserve, not a prescribed equivalent in every target language. The actual target wording should be reviewed by someone who understands its kinship system and the source report’s technical conventions.

Check every alternative named in the source and preserve their status as possibilities. Teach the reader the meaning of the status before suggesting any family-tree action. If the evidence is incomplete, an accurate next step is to verify, ask permission or compare records, not to announce a discovery. The record should leave the reader with the same open question and the same permissible next action. For practice, produce two versions: a concise neutral caption for the screenshot and a longer explanation for a family member. Compare them for unsupported additions. If either version makes a hypothesis sound more certain than the original, revise the expression instead of adding another disclaimer at the bottom.

Worked case 5: Once removed

Imagine the following fictional source record: A genealogist writes ‘first cousin once removed’ alongside an explanatory diagram of two generations. A translator is asked to make it accessible to a family member who does not read the reporting language fluently. Before choosing family terms, identify whether the sentence is a measurement, an automated relationship suggestion, a research hypothesis, a privacy setting or an instruction. That classification protects the reader from treating all words in the document as equally established facts.

An unsafe rendering would suggest: ‘First cousin.’ A more faithful English working interpretation is: ‘First cousin once removed, indicating a one-generation difference in the cousin relationship.’ The removed qualifier conveys generational structure; it is not an optional flourish. This English gloss is an illustration of the decisions to preserve, not a prescribed equivalent in every target language. The actual target wording should be reviewed by someone who understands its kinship system and the source report’s technical conventions.

Draw a simple generational ladder to see whether the translated expression still fits. The source sentence is deliberately short, which makes it tempting to fill its gaps with a plausible story. Instead, separate what the source actually reports from what a reader might want to know. A translation may be clear and readable without answering the unanswered question. The record should leave the reader with the same open question and the same permissible next action. For practice, produce two versions: a concise neutral caption for the screenshot and a longer explanation for a family member. Compare them for unsupported additions. If either version makes a hypothesis sound more certain than the original, revise the expression instead of adding another disclaimer at the bottom.

Worked case 6: Maternal side inferred

Imagine the following fictional source record: A matching tool says ‘Likely maternal cluster’ based on connections to a previously tagged group. A translator is asked to make it accessible to a family member who does not read the reporting language fluently. Before choosing family terms, identify whether the sentence is a measurement, an automated relationship suggestion, a research hypothesis, a privacy setting or an instruction. That classification protects the reader from treating all words in the document as equally established facts.

An unsafe rendering would suggest: ‘Mother’s verified relatives.’ A more faithful English working interpretation is: ‘A cluster the tool tentatively associates with the maternal side.’ The system uses an inference that may need independent family-tree confirmation. This English gloss is an illustration of the decisions to preserve, not a prescribed equivalent in every target language. The actual target wording should be reviewed by someone who understands its kinship system and the source report’s technical conventions.

Trace the word ‘likely’ to an explicit target-language hedge rather than deleting it. Check both directions: every number and relationship in the source must survive in the target, and every person or certainty expressed in the target must have support in the source. A reviewer should be able to underline the exact words that warrant each claim. The record should leave the reader with the same open question and the same permissible next action. For practice, produce two versions: a concise neutral caption for the screenshot and a longer explanation for a family member. Compare them for unsupported additions. If either version makes a hypothesis sound more certain than the original, revise the expression instead of adding another disclaimer at the bottom.

Worked case 7: Paternal side documented

Imagine the following fictional source record: A person reports that their biological father has tested and an identified match is shared with that parent. A translator is asked to make it accessible to a family member who does not read the reporting language fluently. Before choosing family terms, identify whether the sentence is a measurement, an automated relationship suggestion, a research hypothesis, a privacy setting or an instruction. That classification protects the reader from treating all words in the document as equally established facts.

An unsafe rendering would suggest: ‘Every match in this cluster is definitely paternal.’ A more faithful English working interpretation is: ‘This particular match is also observed in the tested father’s match list; other cluster members still require checking.’ A direct parental test can support a narrower claim than a complete lineage assignment. This English gloss is an illustration of the decisions to preserve, not a prescribed equivalent in every target language. The actual target wording should be reviewed by someone who understands its kinship system and the source report’s technical conventions.

Mark which match has been directly observed and which ones remain inferred. Look for the smallest word that changes the claim: perhaps likely, possible, estimated, private, shared or unconfirmed. Losing that word can alter how someone approaches a relative or records a family history, even though the rest of the sentence remains fluent. The record should leave the reader with the same open question and the same permissible next action. For practice, produce two versions: a concise neutral caption for the screenshot and a longer explanation for a family member. Compare them for unsupported additions. If either version makes a hypothesis sound more certain than the original, revise the expression instead of adding another disclaimer at the bottom.

Worked case 8: Unassigned side

Imagine the following fictional source record: A dashboard displays ‘Parent 1’ and ‘Parent 2’ because parent identities have not been entered. A translator is asked to make it accessible to a family member who does not read the reporting language fluently. Before choosing family terms, identify whether the sentence is a measurement, an automated relationship suggestion, a research hypothesis, a privacy setting or an instruction. That classification protects the reader from treating all words in the document as equally established facts.

An unsafe rendering would suggest: ‘Mother’ and ‘father.’ A more faithful English working interpretation is: ‘Parent 1’ and ‘Parent 2’, as labelled by the platform, without assigning sex or identity. Some interfaces use neutral arbitrary labels pending user research. This English gloss is an illustration of the decisions to preserve, not a prescribed equivalent in every target language. The actual target wording should be reviewed by someone who understands its kinship system and the source report’s technical conventions.

Explain whether the labels are numbering devices rather than established maternal and paternal categories. Teach the reader the meaning of the status before suggesting any family-tree action. If the evidence is incomplete, an accurate next step is to verify, ask permission or compare records, not to announce a discovery. The record should leave the reader with the same open question and the same permissible next action. For practice, produce two versions: a concise neutral caption for the screenshot and a longer explanation for a family member. Compare them for unsupported additions. If either version makes a hypothesis sound more certain than the original, revise the expression instead of adding another disclaimer at the bottom.

Worked case 9: Shared-match list

Imagine the following fictional source record: Two kit aliases appear in a shared-match list under a third match. A translator is asked to make it accessible to a family member who does not read the reporting language fluently. Before choosing family terms, identify whether the sentence is a measurement, an automated relationship suggestion, a research hypothesis, a privacy setting or an instruction. That classification protects the reader from treating all words in the document as equally established facts.

An unsafe rendering would suggest: ‘They all descend from the same great-grandparents.’ A more faithful English working interpretation is: ‘The platform lists the individuals as shared matches; the common ancestry is not established here.’ List membership is not the same as proof of a particular common ancestor. This English gloss is an illustration of the decisions to preserve, not a prescribed equivalent in every target language. The actual target wording should be reviewed by someone who understands its kinship system and the source report’s technical conventions.

Look for segment-level and documentary evidence before converting a grouping into genealogy. The source sentence is deliberately short, which makes it tempting to fill its gaps with a plausible story. Instead, separate what the source actually reports from what a reader might want to know. A translation may be clear and readable without answering the unanswered question. The record should leave the reader with the same open question and the same permissible next action. For practice, produce two versions: a concise neutral caption for the screenshot and a longer explanation for a family member. Compare them for unsupported additions. If either version makes a hypothesis sound more certain than the original, revise the expression instead of adding another disclaimer at the bottom.

Worked case 10: Chromosome browser

Imagine the following fictional source record: A user note says ‘A segment on chromosome 8 overlaps the region displayed for another match.’ A translator is asked to make it accessible to a family member who does not read the reporting language fluently. Before choosing family terms, identify whether the sentence is a measurement, an automated relationship suggestion, a research hypothesis, a privacy setting or an instruction. That classification protects the reader from treating all words in the document as equally established facts.

An unsafe rendering would suggest: ‘These people have the same grandfather.’ A more faithful English working interpretation is: ‘The displayed chromosome 8 regions overlap; the ancestral explanation has not been established.’ An apparent overlap requires careful matching, phasing and context before genealogical inference. This English gloss is an illustration of the decisions to preserve, not a prescribed equivalent in every target language. The actual target wording should be reviewed by someone who understands its kinship system and the source report’s technical conventions.

Keep the evidence statement separate from the family-tree theory. Check both directions: every number and relationship in the source must survive in the target, and every person or certainty expressed in the target must have support in the source. A reviewer should be able to underline the exact words that warrant each claim. The record should leave the reader with the same open question and the same permissible next action. For practice, produce two versions: a concise neutral caption for the screenshot and a longer explanation for a family member. Compare them for unsupported additions. If either version makes a hypothesis sound more certain than the original, revise the expression instead of adding another disclaimer at the bottom.

Worked case 11: X-DNA label

Imagine the following fictional source record: A dashboard refers to an ‘X-DNA match’ without explaining inheritance in the message. A translator is asked to make it accessible to a family member who does not read the reporting language fluently. Before choosing family terms, identify whether the sentence is a measurement, an automated relationship suggestion, a research hypothesis, a privacy setting or an instruction. That classification protects the reader from treating all words in the document as equally established facts.

An unsafe rendering would suggest: ‘A paternal-line match.’ A more faithful English working interpretation is: ‘An X-chromosome DNA match as labelled by the testing service.’ The label should not be reduced to a simplistic paternal or maternal route. This English gloss is an illustration of the decisions to preserve, not a prescribed equivalent in every target language. The actual target wording should be reviewed by someone who understands its kinship system and the source report’s technical conventions.

Preserve the technical term and invite specialist interpretation rather than guessing inheritance. Look for the smallest word that changes the claim: perhaps likely, possible, estimated, private, shared or unconfirmed. Losing that word can alter how someone approaches a relative or records a family history, even though the rest of the sentence remains fluent. The record should leave the reader with the same open question and the same permissible next action. For practice, produce two versions: a concise neutral caption for the screenshot and a longer explanation for a family member. Compare them for unsupported additions. If either version makes a hypothesis sound more certain than the original, revise the expression instead of adding another disclaimer at the bottom.

Worked case 12: Y-DNA project

Imagine the following fictional source record: A project page reports a shared Y-DNA haplogroup code for two participants. A translator is asked to make it accessible to a family member who does not read the reporting language fluently. Before choosing family terms, identify whether the sentence is a measurement, an automated relationship suggestion, a research hypothesis, a privacy setting or an instruction. That classification protects the reader from treating all words in the document as equally established facts.

An unsafe rendering would suggest: ‘They are brothers.’ A more faithful English working interpretation is: ‘The two results share the reported Y-DNA haplogroup label; the precise relationship is not established by that label alone.’ Lineage markers can refer to broader ancestry rather than a specific recent relationship. This English gloss is an illustration of the decisions to preserve, not a prescribed equivalent in every target language. The actual target wording should be reviewed by someone who understands its kinship system and the source report’s technical conventions.

Retain the code exactly and identify the type of test. Teach the reader the meaning of the status before suggesting any family-tree action. If the evidence is incomplete, an accurate next step is to verify, ask permission or compare records, not to announce a discovery. The record should leave the reader with the same open question and the same permissible next action. For practice, produce two versions: a concise neutral caption for the screenshot and a longer explanation for a family member. Compare them for unsupported additions. If either version makes a hypothesis sound more certain than the original, revise the expression instead of adding another disclaimer at the bottom.

Worked case 13: Mitochondrial match

Imagine the following fictional source record: A researcher records a matching mitochondrial haplogroup and the words ‘maternal lineage’. A translator is asked to make it accessible to a family member who does not read the reporting language fluently. Before choosing family terms, identify whether the sentence is a measurement, an automated relationship suggestion, a research hypothesis, a privacy setting or an instruction. That classification protects the reader from treating all words in the document as equally established facts.

An unsafe rendering would suggest: ‘They have the same mother.’ A more faithful English working interpretation is: ‘The records share a reported mitochondrial-lineage classification, without establishing the identity of a recent maternal ancestor.’ Maternal-line inheritance is not the same as sharing one immediate mother. This English gloss is an illustration of the decisions to preserve, not a prescribed equivalent in every target language. The actual target wording should be reviewed by someone who understands its kinship system and the source report’s technical conventions.

Check the implied generation and refuse to add a person absent from the source. The source sentence is deliberately short, which makes it tempting to fill its gaps with a plausible story. Instead, separate what the source actually reports from what a reader might want to know. A translation may be clear and readable without answering the unanswered question. The record should leave the reader with the same open question and the same permissible next action. For practice, produce two versions: a concise neutral caption for the screenshot and a longer explanation for a family member. Compare them for unsupported additions. If either version makes a hypothesis sound more certain than the original, revise the expression instead of adding another disclaimer at the bottom.

Worked case 14: Ethnicity estimate

Imagine the following fictional source record: An ancestry panel shows ‘Estimated 14% in reference region R; model may change as data improve.’ A translator is asked to make it accessible to a family member who does not read the reporting language fluently. Before choosing family terms, identify whether the sentence is a measurement, an automated relationship suggestion, a research hypothesis, a privacy setting or an instruction. That classification protects the reader from treating all words in the document as equally established facts.

An unsafe rendering would suggest: ‘One grandparent was from R.’ A more faithful English working interpretation is: ‘The platform estimates 14% similarity to its reference category R; this does not identify a particular grandparent.’ Population models and family relationships answer different questions. This English gloss is an illustration of the decisions to preserve, not a prescribed equivalent in every target language. The actual target wording should be reviewed by someone who understands its kinship system and the source report’s technical conventions.

Preserve the model disclaimer with the number rather than burying it in a footnote. Check both directions: every number and relationship in the source must survive in the target, and every person or certainty expressed in the target must have support in the source. A reviewer should be able to underline the exact words that warrant each claim. The record should leave the reader with the same open question and the same permissible next action. For practice, produce two versions: a concise neutral caption for the screenshot and a longer explanation for a family member. Compare them for unsupported additions. If either version makes a hypothesis sound more certain than the original, revise the expression instead of adding another disclaimer at the bottom.

Worked case 15: Pedigree collapse

Imagine the following fictional source record: A research note says ‘Some ancestors appear more than once in the tree; relationship paths may be complex.’ A translator is asked to make it accessible to a family member who does not read the reporting language fluently. Before choosing family terms, identify whether the sentence is a measurement, an automated relationship suggestion, a research hypothesis, a privacy setting or an instruction. That classification protects the reader from treating all words in the document as equally established facts.

An unsafe rendering would suggest: ‘The DNA result is invalid.’ A more faithful English working interpretation is: ‘Repeated ancestral paths may complicate a simple relationship estimate; the report does not declare the result invalid.’ Pedigree collapse changes interpretation, not necessarily whether a match exists. This English gloss is an illustration of the decisions to preserve, not a prescribed equivalent in every target language. The actual target wording should be reviewed by someone who understands its kinship system and the source report’s technical conventions.

Check that the target preserves complexity without diagnosing the underlying family structure. Look for the smallest word that changes the claim: perhaps likely, possible, estimated, private, shared or unconfirmed. Losing that word can alter how someone approaches a relative or records a family history, even though the rest of the sentence remains fluent. The record should leave the reader with the same open question and the same permissible next action. For practice, produce two versions: a concise neutral caption for the screenshot and a longer explanation for a family member. Compare them for unsupported additions. If either version makes a hypothesis sound more certain than the original, revise the expression instead of adding another disclaimer at the bottom.

Worked case 16: Endogamous community

Imagine the following fictional source record: A researcher writes ‘Higher shared DNA in this community may reflect multiple ancestral links.’ A translator is asked to make it accessible to a family member who does not read the reporting language fluently. Before choosing family terms, identify whether the sentence is a measurement, an automated relationship suggestion, a research hypothesis, a privacy setting or an instruction. That classification protects the reader from treating all words in the document as equally established facts.

An unsafe rendering would suggest: ‘Every shared match is a close cousin.’ A more faithful English working interpretation is: ‘Multiple ancestral links in an endogamous population may affect how a simple cousin prediction is interpreted.’ One population history can make simplistic relationship labels less useful. This English gloss is an illustration of the decisions to preserve, not a prescribed equivalent in every target language. The actual target wording should be reviewed by someone who understands its kinship system and the source report’s technical conventions.

Verify that the translation does not pathologise a community or make a universal claim. Teach the reader the meaning of the status before suggesting any family-tree action. If the evidence is incomplete, an accurate next step is to verify, ask permission or compare records, not to announce a discovery. The record should leave the reader with the same open question and the same permissible next action. For practice, produce two versions: a concise neutral caption for the screenshot and a longer explanation for a family member. Compare them for unsupported additions. If either version makes a hypothesis sound more certain than the original, revise the expression instead of adding another disclaimer at the bottom.

Worked case 17: Private tree

Imagine the following fictional source record: The match page states ‘Family tree private. Access available only by owner invitation.’ A translator is asked to make it accessible to a family member who does not read the reporting language fluently. Before choosing family terms, identify whether the sentence is a measurement, an automated relationship suggestion, a research hypothesis, a privacy setting or an instruction. That classification protects the reader from treating all words in the document as equally established facts.

An unsafe rendering would suggest: ‘Open the relative’s public family tree.’ A more faithful English working interpretation is: ‘The family tree is private; the owner must choose whether to grant access.’ Privacy status is an operational rule and a consent boundary. This English gloss is an illustration of the decisions to preserve, not a prescribed equivalent in every target language. The actual target wording should be reviewed by someone who understands its kinship system and the source report’s technical conventions.

Check that the reader would not be told to bypass a restricted account. The source sentence is deliberately short, which makes it tempting to fill its gaps with a plausible story. Instead, separate what the source actually reports from what a reader might want to know. A translation may be clear and readable without answering the unanswered question. The record should leave the reader with the same open question and the same permissible next action. For practice, produce two versions: a concise neutral caption for the screenshot and a longer explanation for a family member. Compare them for unsupported additions. If either version makes a hypothesis sound more certain than the original, revise the expression instead of adding another disclaimer at the bottom.

Worked case 18: No response to a message

Imagine the following fictional source record: An email log reads ‘Message sent 4 May; no reply as of 20 May.’ A translator is asked to make it accessible to a family member who does not read the reporting language fluently. Before choosing family terms, identify whether the sentence is a measurement, an automated relationship suggestion, a research hypothesis, a privacy setting or an instruction. That classification protects the reader from treating all words in the document as equally established facts.

An unsafe rendering would suggest: ‘The match refuses to cooperate.’ A more faithful English working interpretation is: ‘A message was sent on 4 May; no response had been received by 20 May.’ Silence is not proof of refusal, disagreement or even receipt. This English gloss is an illustration of the decisions to preserve, not a prescribed equivalent in every target language. The actual target wording should be reviewed by someone who understands its kinship system and the source report’s technical conventions.

Keep the temporal cut-off and avoid assigning motive. Check both directions: every number and relationship in the source must survive in the target, and every person or certainty expressed in the target must have support in the source. A reviewer should be able to underline the exact words that warrant each claim. The record should leave the reader with the same open question and the same permissible next action. For practice, produce two versions: a concise neutral caption for the screenshot and a longer explanation for a family member. Compare them for unsupported additions. If either version makes a hypothesis sound more certain than the original, revise the expression instead of adding another disclaimer at the bottom.

Worked case 19: Managed kit

Imagine the following fictional source record: The interface shows ‘Kit managed by another account holder’. A translator is asked to make it accessible to a family member who does not read the reporting language fluently. Before choosing family terms, identify whether the sentence is a measurement, an automated relationship suggestion, a research hypothesis, a privacy setting or an instruction. That classification protects the reader from treating all words in the document as equally established facts.

An unsafe rendering would suggest: ‘The manager is the person tested.’ A more faithful English working interpretation is: ‘A separate account holder manages access to the test results; they may not be the person who provided the sample.’ Account management and biological identity are distinct roles. This English gloss is an illustration of the decisions to preserve, not a prescribed equivalent in every target language. The actual target wording should be reviewed by someone who understands its kinship system and the source report’s technical conventions.

Preserve any permission requirements for contacting or sharing data. Look for the smallest word that changes the claim: perhaps likely, possible, estimated, private, shared or unconfirmed. Losing that word can alter how someone approaches a relative or records a family history, even though the rest of the sentence remains fluent. The record should leave the reader with the same open question and the same permissible next action. For practice, produce two versions: a concise neutral caption for the screenshot and a longer explanation for a family member. Compare them for unsupported additions. If either version makes a hypothesis sound more certain than the original, revise the expression instead of adding another disclaimer at the bottom.

Worked case 20: Adoption hypothesis

Imagine the following fictional source record: A family-tree annotation says ‘Possible birth-family lead, do not contact without consent.’ A translator is asked to make it accessible to a family member who does not read the reporting language fluently. Before choosing family terms, identify whether the sentence is a measurement, an automated relationship suggestion, a research hypothesis, a privacy setting or an instruction. That classification protects the reader from treating all words in the document as equally established facts.

An unsafe rendering would suggest: ‘DNA confirms the birth parent; send a message now.’ A more faithful English working interpretation is: ‘A possible lead has been noted, with an explicit instruction not to make contact without consent.’ This is both an evidential hedge and a privacy instruction. This English gloss is an illustration of the decisions to preserve, not a prescribed equivalent in every target language. The actual target wording should be reviewed by someone who understands its kinship system and the source report’s technical conventions.

Audit whether the prohibition survives translation in an unambiguous form. Teach the reader the meaning of the status before suggesting any family-tree action. If the evidence is incomplete, an accurate next step is to verify, ask permission or compare records, not to announce a discovery. The record should leave the reader with the same open question and the same permissible next action. For practice, produce two versions: a concise neutral caption for the screenshot and a longer explanation for a family member. Compare them for unsupported additions. If either version makes a hypothesis sound more certain than the original, revise the expression instead of adding another disclaimer at the bottom.

Worked case 21: Historical name spelling

Imagine the following fictional source record: An archival note contains two spellings for one surname, with a question mark after the earlier reading. A translator is asked to make it accessible to a family member who does not read the reporting language fluently. Before choosing family terms, identify whether the sentence is a measurement, an automated relationship suggestion, a research hypothesis, a privacy setting or an instruction. That classification protects the reader from treating all words in the document as equally established facts.

An unsafe rendering would suggest: ‘The surname was definitely changed.’ A more faithful English working interpretation is: ‘The record shows an uncertain reading of an earlier surname spelling; the discrepancy is unresolved.’ Historical orthography cannot be settled by smoothing names into one modern spelling. This English gloss is an illustration of the decisions to preserve, not a prescribed equivalent in every target language. The actual target wording should be reviewed by someone who understands its kinship system and the source report’s technical conventions.

Keep the original spellings in the source script and note transliteration choices separately. The source sentence is deliberately short, which makes it tempting to fill its gaps with a plausible story. Instead, separate what the source actually reports from what a reader might want to know. A translation may be clear and readable without answering the unanswered question. The record should leave the reader with the same open question and the same permissible next action. For practice, produce two versions: a concise neutral caption for the screenshot and a longer explanation for a family member. Compare them for unsupported additions. If either version makes a hypothesis sound more certain than the original, revise the expression instead of adding another disclaimer at the bottom.

Worked case 22: Match threshold

Imagine the following fictional source record: A vendor note says ‘Small segments below the current display threshold are not shown.’ A translator is asked to make it accessible to a family member who does not read the reporting language fluently. Before choosing family terms, identify whether the sentence is a measurement, an automated relationship suggestion, a research hypothesis, a privacy setting or an instruction. That classification protects the reader from treating all words in the document as equally established facts.

An unsafe rendering would suggest: ‘All genetic similarities have been analysed and excluded.’ A more faithful English working interpretation is: ‘The interface does not display segments below its stated threshold; absence on screen is not proof of no matching DNA.’ A display setting is not a universal scientific negative. This English gloss is an illustration of the decisions to preserve, not a prescribed equivalent in every target language. The actual target wording should be reviewed by someone who understands its kinship system and the source report’s technical conventions.

Capture the threshold if the source provides it and avoid inventing a number. Check both directions: every number and relationship in the source must survive in the target, and every person or certainty expressed in the target must have support in the source. A reviewer should be able to underline the exact words that warrant each claim. The record should leave the reader with the same open question and the same permissible next action. For practice, produce two versions: a concise neutral caption for the screenshot and a longer explanation for a family member. Compare them for unsupported additions. If either version makes a hypothesis sound more certain than the original, revise the expression instead of adding another disclaimer at the bottom.

Worked case 23: Updated relationship model

Imagine the following fictional source record: A dashboard change note says ‘Estimated relationship categories updated in September; cM values unchanged.’ A translator is asked to make it accessible to a family member who does not read the reporting language fluently. Before choosing family terms, identify whether the sentence is a measurement, an automated relationship suggestion, a research hypothesis, a privacy setting or an instruction. That classification protects the reader from treating all words in the document as equally established facts.

An unsafe rendering would suggest: ‘Your DNA changed in September.’ A more faithful English working interpretation is: ‘The service changed its relationship estimates in September; the reported shared-cM values stayed the same.’ An algorithm update is not a biological change in the underlying sample. This English gloss is an illustration of the decisions to preserve, not a prescribed equivalent in every target language. The actual target wording should be reviewed by someone who understands its kinship system and the source report’s technical conventions.

Ask the reviewer to identify what changed and what remained stable. Look for the smallest word that changes the claim: perhaps likely, possible, estimated, private, shared or unconfirmed. Losing that word can alter how someone approaches a relative or records a family history, even though the rest of the sentence remains fluent. The record should leave the reader with the same open question and the same permissible next action. For practice, produce two versions: a concise neutral caption for the screenshot and a longer explanation for a family member. Compare them for unsupported additions. If either version makes a hypothesis sound more certain than the original, revise the expression instead of adding another disclaimer at the bottom.

Worked case 24: Translation request to family

Imagine the following fictional source record: A relative asks ‘Could you translate this result for Grandma without naming the unknown match?’ A translator is asked to make it accessible to a family member who does not read the reporting language fluently. Before choosing family terms, identify whether the sentence is a measurement, an automated relationship suggestion, a research hypothesis, a privacy setting or an instruction. That classification protects the reader from treating all words in the document as equally established facts.

An unsafe rendering would suggest: ‘Here is the stranger’s identity and address.’ A more faithful English working interpretation is: ‘Explain the reported measurements and uncertainties while withholding the match’s identifying details as requested.’ Good translation includes audience, purpose and data minimisation. This English gloss is an illustration of the decisions to preserve, not a prescribed equivalent in every target language. The actual target wording should be reviewed by someone who understands its kinship system and the source report’s technical conventions.

Prepare a privacy-safe explanation that preserves the finding but omits identifiers. Teach the reader the meaning of the status before suggesting any family-tree action. If the evidence is incomplete, an accurate next step is to verify, ask permission or compare records, not to announce a discovery. The record should leave the reader with the same open question and the same permissible next action. For practice, produce two versions: a concise neutral caption for the screenshot and a longer explanation for a family member. Compare them for unsupported additions. If either version makes a hypothesis sound more certain than the original, revise the expression instead of adding another disclaimer at the bottom.

Common failures, the damage they cause and how to repair them

1. Changing a unit

Replacing cM with a percentage or confusing centimorgans with centimetres makes the source measurement unrecoverable. Copy the number and cM unit exactly, define the unit once and check the metric name on both sides. A reliable repair involves checking the original screenshot or source document again, not merely replacing one target word with a synonym. Ask what decision the reader may make after seeing the sentence. If that decision could differ from the one supported by the original report, make the missing distinction explicit. Record the correction in a small glossary so the same error does not recur across pages or family correspondence.

2. Selecting one cousin title

Choosing a specific kinship word from a range can falsely assign parentage, generation and family side. Use a descriptive relationship range, retain the service attribution and let documentary evidence decide. A reliable repair involves checking the original screenshot or source document again, not merely replacing one target word with a synonym. Ask what decision the reader may make after seeing the sentence. If that decision could differ from the one supported by the original report, make the missing distinction explicit. Record the correction in a small glossary so the same error does not recur across pages or family correspondence.

3. Dropping ‘half’ or ‘removed’

These words are structural facts in kinship descriptions, not optional politeness markers. Use an explanatory phrase if the target language lacks a compact equivalent. A reliable repair involves checking the original screenshot or source document again, not merely replacing one target word with a synonym. Ask what decision the reader may make after seeing the sentence. If that decision could differ from the one supported by the original report, make the missing distinction explicit. Record the correction in a small glossary so the same error does not recur across pages or family correspondence.

4. Confusing alias and identity

A screen name, kit manager and DNA participant may be different individuals. Preserve the field label and avoid expanding an account alias into a person’s legal name. A reliable repair involves checking the original screenshot or source document again, not merely replacing one target word with a synonym. Ask what decision the reader may make after seeing the sentence. If that decision could differ from the one supported by the original report, make the missing distinction explicit. Record the correction in a small glossary so the same error does not recur across pages or family correspondence.

5. Treating a tree as laboratory evidence

Public user trees may contain researched facts, guesses, copied errors or deliberate privacy exclusions. Attribute family-tree claims to their authors; do not convert them into system conclusions. A reliable repair involves checking the original screenshot or source document again, not merely replacing one target word with a synonym. Ask what decision the reader may make after seeing the sentence. If that decision could differ from the one supported by the original report, make the missing distinction explicit. Record the correction in a small glossary so the same error does not recur across pages or family correspondence.

6. Misreading an overlap

Overlapping chromosome display regions, shared-match networks and predicted ancestral branches carry different implications. Name exactly which kind of overlap the interface shows and what it does not establish. A reliable repair involves checking the original screenshot or source document again, not merely replacing one target word with a synonym. Ask what decision the reader may make after seeing the sentence. If that decision could differ from the one supported by the original report, make the missing distinction explicit. Record the correction in a small glossary so the same error does not recur across pages or family correspondence.

7. Overstating ethnic ancestry

Population similarity estimates do not identify nationality, ancestry certificates or immediate relatives. Keep the term estimate and its reference-population framing. A reliable repair involves checking the original screenshot or source document again, not merely replacing one target word with a synonym. Ask what decision the reader may make after seeing the sentence. If that decision could differ from the one supported by the original report, make the missing distinction explicit. Record the correction in a small glossary so the same error does not recur across pages or family correspondence.

8. Turning silence into rejection

A missing reply does not prove that a relative refuses contact or that the account is inactive. Translate dates, message status and absence of response without assigning motives. A reliable repair involves checking the original screenshot or source document again, not merely replacing one target word with a synonym. Ask what decision the reader may make after seeing the sentence. If that decision could differ from the one supported by the original report, make the missing distinction explicit. Record the correction in a small glossary so the same error does not recur across pages or family correspondence.

9. Publishing sensitive findings

A family translation may be forwarded outside the intended group or reveal adoption and non-parentage questions. Work with minimum necessary identifiers, retain consent conditions and discuss distribution before sharing. A reliable repair involves checking the original screenshot or source document again, not merely replacing one target word with a synonym. Ask what decision the reader may make after seeing the sentence. If that decision could differ from the one supported by the original report, make the missing distinction explicit. Record the correction in a small glossary so the same error does not recur across pages or family correspondence.

10. Ignoring a model update

A relationship prediction can be revised without any change to an underlying cM figure. Retain the version date and state clearly whether measurements or interpretations changed. A reliable repair involves checking the original screenshot or source document again, not merely replacing one target word with a synonym. Ask what decision the reader may make after seeing the sentence. If that decision could differ from the one supported by the original report, make the missing distinction explicit. Record the correction in a small glossary so the same error does not recur across pages or family correspondence.

11. Inventing the platform’s certainty

A numerical probability or confidence level must not be supplied because another service uses one. Preserve only labels actually displayed; request a source screenshot or technical legend when unsure. A reliable repair involves checking the original screenshot or source document again, not merely replacing one target word with a synonym. Ask what decision the reader may make after seeing the sentence. If that decision could differ from the one supported by the original report, make the missing distinction explicit. Record the correction in a small glossary so the same error does not recur across pages or family correspondence.

12. Over-annotating as if confirmed

A translator’s helpful notes sometimes blend into the translated report and become mistaken for evidence. Separate translator notes typographically and label them as explanation rather than original reporting. A reliable repair involves checking the original screenshot or source document again, not merely replacing one target word with a synonym. Ask what decision the reader may make after seeing the sentence. If that decision could differ from the one supported by the original report, make the missing distinction explicit. Record the correction in a small glossary so the same error does not recur across pages or family correspondence.

Practice programme: from reading the report to explaining it responsibly

Session one — annotate the evidence

Use three invented match cards with total cM, longest segment and candidate relationship ranges. Highlight measurements in one colour, estimates in another and messages in a third. Translate only after you can explain what each category contains. Review both source and target without relying on your familiarity with the family’s narrative. The standard of success is not speed of translation or number of labels memorised. It is whether a new reader can correctly state the measurement, recognise that multiple relationships remain possible, and refrain from presenting the translation as genetic or legal proof. Keep exercises fictional and do not upload identifiable third-party DNA results into unapproved tools.

Session two — map difficult kinship terms

Write a small glossary for first cousin, second cousin, once removed, half sibling, maternal side, paternal side, unknown side, match and ancestor. For each term list its meaning, possible equivalents and things the equivalent must not imply. Include a natural descriptive fallback when one target-language family title is too specific. The standard of success is not speed of translation or number of labels memorised. It is whether a new reader can correctly state the measurement, recognise that multiple relationships remain possible, and refrain from presenting the translation as genetic or legal proof. Keep exercises fictional and do not upload identifiable third-party DNA results into unapproved tools.

Session three — practise sensitive communication

Compose a short private explanation of a fictional uncertain match. Preserve the uncertainty and avoid identifying someone without consent. Ask a peer to mark which sentences report facts, which interpret a possibility and which request next actions. Revise until those three functions are plainly separable. The standard of success is not speed of translation or number of labels memorised. It is whether a new reader can correctly state the measurement, recognise that multiple relationships remain possible, and refrain from presenting the translation as genetic or legal proof. Keep exercises fictional and do not upload identifiable third-party DNA results into unapproved tools.

Session four — audit transfer and updates

Translate a mock report, then receive a second version showing a changed prediction but identical measurements. Produce a two-column change log. Explain in ordinary language what was updated, what was not updated and which family-tree questions remain unresolved. Repeat this task several days later without looking at the model answer. The standard of success is not speed of translation or number of labels memorised. It is whether a new reader can correctly state the measurement, recognise that multiple relationships remain possible, and refrain from presenting the translation as genetic or legal proof. Keep exercises fictional and do not upload identifiable third-party DNA results into unapproved tools.

Frequently asked questions about DNA match translation

Does 100 cM prove a particular cousin relationship?

No. A shared-cM figure can support consideration of several relationships and depends on the wider record. Translate the measurement accurately and preserve the report’s uncertainty rather than naming a definitive cousin. The practical test is whether the reader remains free to ask the right next question, rather than being pushed toward a conclusion that the original report did not reach. When a word is ambiguous, keep the ambiguity visible and seek the relevant testing-platform explanation.

Can I translate cM into a percentage?

Do not silently convert. If the platform itself provides both values, preserve both with their labels. Otherwise a conversion requires a specified model and technical context; it is not a substitute for the source measurement. The practical test is whether the reader remains free to ask the right next question, rather than being pushed toward a conclusion that the original report did not reach. When a word is ambiguous, keep the ambiguity visible and seek the relevant testing-platform explanation.

How should I translate a kinship term with no exact equivalent?

Describe the structure: cousin degree, generation removed and half relationship where applicable. An explanatory phrase is safer than a familiar target term that assigns unverified gender or parental side. The practical test is whether the reader remains free to ask the right next question, rather than being pushed toward a conclusion that the original report did not reach. When a word is ambiguous, keep the ambiguity visible and seek the relevant testing-platform explanation.

Is ‘predicted relationship’ the same as ‘confirmed relationship’?

No. Predicted labels are outputs of a method or model. A confirmed genealogical relationship needs additional independent evidence. The target should make the report’s attribution and uncertainty visible. The practical test is whether the reader remains free to ask the right next question, rather than being pushed toward a conclusion that the original report did not reach. When a word is ambiguous, keep the ambiguity visible and seek the relevant testing-platform explanation.

What do shared matches prove?

They show a platform-defined overlap in match lists. They do not automatically prove a shared named ancestor or a particular chromosome-based triangulation. Preserve the interface label precisely. The practical test is whether the reader remains free to ask the right next question, rather than being pushed toward a conclusion that the original report did not reach. When a word is ambiguous, keep the ambiguity visible and seek the relevant testing-platform explanation.

Should a translator interpret chromosome data?

A translator can accurately convey the source labels and cite neutral explanations. Specialist genetic interpretation should be identified as such, not inserted into the translated report as an undisputed fact. The practical test is whether the reader remains free to ask the right next question, rather than being pushed toward a conclusion that the original report did not reach. When a word is ambiguous, keep the ambiguity visible and seek the relevant testing-platform explanation.

Can ethnicity percentages identify my grandfather?

No. They are model-based population estimates and do not identify a specific family member. Translating them as a certain ancestral person would add information not in the source. The practical test is whether the reader remains free to ask the right next question, rather than being pushed toward a conclusion that the original report did not reach. When a word is ambiguous, keep the ambiguity visible and seek the relevant testing-platform explanation.

What if the match profile is private?

Keep the privacy status, consent requirements and contact boundaries explicit. Access restrictions remain restrictions in every language, and a family member’s curiosity is not permission to disclose more. The practical test is whether the reader remains free to ask the right next question, rather than being pushed toward a conclusion that the original report did not reach. When a word is ambiguous, keep the ambiguity visible and seek the relevant testing-platform explanation.

How do I translate uncertain historical surnames?

Keep the source spelling, transliteration choices and uncertainty markers visible. Do not force every archival spelling into a modern form that falsely suggests certainty about identity. The practical test is whether the reader remains free to ask the right next question, rather than being pushed toward a conclusion that the original report did not reach. When a word is ambiguous, keep the ambiguity visible and seek the relevant testing-platform explanation.

Should family-tree hypotheses appear in the translation?

Only if the source includes them, with their original attribution and confidence. Separate researcher notes from testing-service data so readers know which statements have documentary support. The practical test is whether the reader remains free to ask the right next question, rather than being pushed toward a conclusion that the original report did not reach. When a word is ambiguous, keep the ambiguity visible and seek the relevant testing-platform explanation.

Can AI translate DNA reports safely?

AI may help draft phrasing or terminology notes, but numbers, names, uncertainty and privacy must be checked against the original. Sensitive third-party information should not be uploaded to tools without appropriate authorisation. The practical test is whether the reader remains free to ask the right next question, rather than being pushed toward a conclusion that the original report did not reach. When a word is ambiguous, keep the ambiguity visible and seek the relevant testing-platform explanation.

What should the reader do after translation?

Compare the translated report with the original, verify important genealogical claims against documentary sources, respect consent and distinguish a promising lead from a settled family relationship. The practical test is whether the reader remains free to ask the right next question, rather than being pushed toward a conclusion that the original report did not reach. When a word is ambiguous, keep the ambiguity visible and seek the relevant testing-platform explanation.

Sources, scope and further learning

For transparent interpretation of shared-centimorgan ranges, consult the DNA Painter Shared cM Project tool. Its relationship groupings and assumptions illustrate why one shared-DNA amount can correspond to multiple genealogical paths. These examples are educational and not a substitute for professional genetic interpretation, identity verification or counselling.

This guide has one canonical job: protect the difference between a DNA match report and a claimed family relationship. The broader Translate Meaning, Not Just Words owner explains semantic equivalence. Context, Tone and Register teaches audience-sensitive choices; Idioms and Culture covers non-literal language; and Terminology, Glossaries and Quality Checks explains consistent specialist vocabulary.

For connected learning, explore the Vocabulary Learning Hub for precise words and transfer, and How English Works: Translation and Meaning Across Languages for the larger language framework. Accurate family-history translation preserves what a report actually knows, marks what it merely suggests and leaves uncertainty intact until further evidence can resolve it.