Foundational Myths

Branch 214 · 22 birth
Foundational Myths
Tracing Greek mythological familial connections and roots

Tracing Greek mythological familial connections and roots

When tracing a figure like Heracles back through his purported familial lines—for example, connecting him to figures from earlier generations or distant collateral branches—you must approach…

11 Sep
View parish
Understanding the mythic origins of Adam and Eve

Understanding the mythic origins of Adam and Eve

“If Adam and Eve were the original pair, how can we even begin tracing their descendants using modern genealogical methods?”…

8 Sep
View parish

The single most consequential thing to understand about family trees is that "removed" counts generations, not people, and that it is a measure of the gap between two cousins, not a rank. Two people who are first cousins once removed are one generation apart; first cousins twice removed are two generations apart. That is the whole rule, and almost every foundational myth in genealogy is a variation on misreading it — or on confusing the two diagrams everyone draws, the pedigree chart and the family tree, which answer different questions and are not interchangeable.

Cousin "removed" is a generation gap, not a distance up the tree

The number before "removed" is the number of generations between the two people you are comparing. Once removed means the two people are one generation apart. Twice removed means two generations apart. Nothing more is being said. A first cousin once removed is not a more distant relative than a first cousin; they are the same cousin line, just offset by a generation.

The confusion starts because people learn "first cousin" as a single fixed thing — your parent's sibling's child — and then try to extend the label upward and downward by feel. The reliable way to do it is to name the common ancestor and count generations down from them to each of you separately. The smaller number of generations is the cousin degree; the difference between the two counts is the removal. If your shared ancestor is a great-grandparent to you and a grandparent to the other person, you are offset by one generation, and the relationship is a cousin relationship once removed. If the offset is two generations, it is twice removed. The counting is mechanical, and it works for every branch, which is why it is worth learning properly rather than memorising a handful of cases.

The trade-off in doing it the mechanical way is that it sounds stiff in conversation — nobody says "my first cousin once removed" at a wedding. Most people say "my cousin's daughter" or "my mother's cousin," which is perfectly clear and carries more information for a listener who knows the family. The formal term earns its keep in writing, in databases, and in any situation where the reader does not know your family. Use the plain description socially and the formal term in records.

The word "cousin" doesn't mean what most people think it means

In ordinary speech, "cousin" covers an enormous range — second cousins, cousins once removed, sometimes anyone sharing a grandparent or a great-grandparent. In genealogy, "cousin" is a specific claim: you share a set of grandparents, and the degree tells you which set. First cousins share grandparents. Second cousins share great-grandparents. The degree is determined by the nearest shared ancestor, and it is the same for both people, because you are both the same distance from that ancestor.

Where this goes wrong in practice is that people use "cousin" as a catch-all for anyone in the same generation of a collateral line and then get surprised when a DNA match labelled "second cousin" turns out to share a great-great-grandparent instead. The label was doing social work, not genealogical work. When a match or a record uses the word, the useful question is not "how close are we?" but "which shared ancestor produces this degree?" — because the degree is derived from that ancestor, not assigned independently.

The limit of the formal system is that it only describes blood relationships through a shared ancestor. It says nothing useful about half-cousins, step-relations, or adoptive lines, all of which are common and none of which the standard degree-and-removal vocabulary handles cleanly. You will end up writing "half first cousin" and explaining it, which is honest and better than forcing the term.

A pedigree chart and a family tree are not the same diagram

A pedigree chart shows direct ancestors generation by generation, starting with the person at the centre. A family tree is a diagram of family relationships that can include ancestors, descendants, and other relatives. The difference is what is on the page: a pedigree chart is a single line of descent — you, your parents, your grandparents, and back — and it deliberately excludes siblings, cousins, and anyone who is not your direct ancestor. A family tree can hold all of them.

This matters because beginners often download a "family tree" template, fill it with direct ancestors only, and then cannot work out where a cousin belongs. The template was a pedigree chart and the question was a family tree question. If you are trying to place a cousin, a sibling's descendants, or a second wife, you need the wider diagram. If you are trying to prove a line of descent for a heritage society or a citizenship claim, the pedigree chart is the correct and expected form, and adding collateral relatives to it will only make it harder to read.

The cost of running both is duplication and drift: the same person appears on two diagrams, and if you correct a date on one you have to remember to correct it on the other. Most people end up with a pedigree chart as the working summary and a database that holds the full tree, treating the pedigree as a view rather than a separate document. That is the right instinct — one source of truth, two ways of looking at it.

Everyone is the centre of their own chart, and that is a feature

A pedigree chart starts with the person at the centre, which means your chart and your cousin's chart are different documents covering overlapping ground. Neither is wrong. This trips people up when they compare charts and find the same ancestor in a different position, or when they inherit a chart drawn from a great-aunt's perspective and try to read it as if it were theirs.

The practical consequence is that a pedigree chart is always partly a statement about the person who drew it, and you should check who that is before you trust the layout. A chart with no named starting person is close to useless, because the whole structure is relative to that centre. When you draw your own, label the centre explicitly, and when you file someone else's, note whose it is on the back or in the file name.

The trade-off is that a centring convention makes charts hard to merge. If you and a cousin each draw a pedigree chart, combining them into one document is not a matter of pasting them side by side — the centres have to be reconciled, and one of them stops being the centre. That is one reason a shared database beats a shared chart for collaborative work, and why charts are best treated as outputs from the data rather than as the data itself.

Records lie in predictable ways, and the myths are usually about which record wins

When two records disagree, the older one is not automatically right and the more official-looking one is not automatically right either. What matters is who supplied the information and how close they were to the event. A death certificate gives a date of birth that the informant — often a grieving relative — was recalling from memory, sometimes decades after the fact. A birth registration made shortly after the birth, by someone present, is generally stronger evidence for that date. The reverse can be true for other facts on the same certificate.

This is where a lot of beginner error concentrates: treating one document as the answer rather than as a witness. Each record is a person's account of an event, filtered through what they knew, what they were asked, and what they had reason to say. A census asks for age and place of birth and gets an approximation. A marriage record asks for parents' names and gets what the couple believed or chose to state. A family bible records what the family wanted remembered.

The honest limit is that you will rarely resolve a conflict to certainty. The workable standard is to record the competing values, note which source you are treating as primary and why, and move on. Chasing a single "correct" date for a person born before civil registration is often not achievable, and the effort is better spent on a different person in the same generation.

Spelling was not fixed, and name changes were ordinary

Surname spelling in older records varies constantly, and treating each variant as a different family is one of the most common ways beginners lose a line. The same family appears under several spellings across parish registers, census returns, and civil records, often within a single document set. The clerk wrote what he heard; the family may not have been able to read it back.

Name changes were also ordinary rather than scandalous. Families anglicised names, dropped prefixes, split or merged compounds, and adopted a stepfather's surname. A person who appears under one surname as a child and another as an adult is not necessarily hiding anything. The trap is assuming that a name change implies adoption, illegitimacy, or a deliberate break, when the likeliest explanation is often a clerk, a migration, or a family decision to fit in.

The practical response is to search variants systematically rather than one at a time, and to record every spelling you find against the person so the next search starts from the full set. The cost is time and a certain amount of noise — variant searches pull in unrelated families, and you have to check each hit. That is cheaper than missing a whole branch because you searched only the spelling on the birth certificate.

What you actually have to decide, and in what order

The decisions come in a sequence, and getting the order right saves most of the rework. First, decide what question you are answering: a single line of descent, or the full set of relationships around a person. That choice determines whether you are drawing a pedigree chart or building a family tree, and it determines what counts as finished. Second, decide how you will record evidence — one place per fact, with the source attached — before you have enough facts for it to matter. Retrofitting source citations onto a tree you built without them is the single most tedious job in this work, and it is the job most people abandon halfway.

Third, decide your standard for a conclusion. Some people will not enter a relationship without a record that states it directly; others will enter a probable relationship flagged as unproven. Both are defensible, but you have to pick one and apply it consistently, or your tree becomes a mixture of proven and guessed facts with no way to tell them apart. Fourth, decide how you will handle the cousin question in writing — formal degree-and-removal terms, or plain description — and stick to it, because switching mid-document is what makes a family history confusing to read.

The cost of all this front-loading is that you spend your first sessions setting up rather than finding anyone. The return is that you never have to stop and fix a structural decision later, which is the point at which most family history projects quietly stall. The myths about cousins, charts, records, and names are not hard to see through once you know the underlying rule. What is hard is applying the rule consistently for long enough that the tree stays trustworthy — and that is a matter of deciding, early, what you will accept as proof.

Half-siblings break the counting, not the vocabulary

The degree-and-removal system assumes both people descend from the same pair of ancestors. When they share only one, the count still runs but every label needs a qualifier. A half first cousin shares one grandparent rather than a grandparent couple; the arithmetic of generations is unchanged, the shared ancestry is halved. Say so explicitly in writing rather than letting "first cousin" imply a full set of grandparents that the records do not support.

Adoptive and step lines fail the system differently: there is no shared ancestor to count down from at all, so no degree can be derived. The relationship is real and often central to the family, but it is a social and legal fact, not a genetic one, and forcing it into cousin terminology produces claims the documents cannot back. Describe it in words — "raised alongside," "adopted by" — and keep the blood lines separate in the database. The cost is that the written account reads more clumsily; the return is that a reader can tell which relationships rest on descent and which on household.

Draw the chart from the data, never the other way round

Charts are outputs, not archives. If the only place a relationship exists is a line drawn on a pedigree chart, you have stored a conclusion without its evidence, and the next person to inherit the document cannot check it. Keep the facts and sources in one place — a database, a spreadsheet, a set of notes with citations — and generate the chart from that when you need to show someone.

The practical test is whether you could redraw the chart from scratch after losing the file. If you cannot, the chart was the source of truth and the underlying work was never done. Regenerating also catches drift: a date corrected in the database but not on the wall chart, a cousin placed on the wrong branch and never moved. Charts go stale silently; data with citations attached does not.