"Genealogy & Privacy: Keeping It In The Family"

When doing research on your family history one of the most important things to consider right off the top, especially in this digital age that we’re all now helplessly immersed in, is the need for personal guidelines on the issue of privacy.
 
Never before in history has it been so easy to gain access to personal information. With the help of the internet, anyone can become a genealogist and post their family history on the web for all to see. However, publicly published family trees can become easy targets for identity thieves. For any living person in your family tree you may have data about them that they may not even be aware could be available to the general public – if any of this information is to be published on a web site, you have some special considerations to make. These may even involve some persons in your tree who are deceased.
 
The first thing I did when setting up my genealogy site was to find a way to keep it in the family. I did this by creating a “splash” page as my home page which requires the entry of a password to get into and be able to view anything on the site. I won’t get into the technical programming know-how of web site password protection here, but there are different ways of doing it – these methods can be Googled by the DIY web site designers out there. Once you get password protection in place, you can control which individuals can see the content on your web site by sending them the password upon request. It’s also a good idea to change the password periodically to be extra safe, just don’t forget to let anyone who has access know when you do.
 
The next thing to do was to figure out what my personal “privacy code” was going to be. This was to be my ethics statement and deal with how I would handle all of the private information I was going to be gathering. Common sense dictated that people in my tree who were still living would need special rules to protect their private information, even from other family members. Some information may be very sensitive in nature. Even data collected about some deceased relatives may be too sensitive to publish. If the web site was to be accessible by the public, it would be appropriate to remove all data associated with any living relatives, as well as to limit data associated with deceased relatives who have close connections to any living relatives. One could replace the name of each living relative with the word “Living” to protect their identities, and only publish the names of their immediate ancestors with no other vital information.
 
Because my site is private, I chose not to restrict information associated with deceased relatives unless the data is extremely sensitive, looking at each case individually. For living relatives, I chose to only publish their names with no other vital data, unless I was expressly granted permission by that person to publish more information about them. That said, I must be prepared to remove them completely from the published version of my tree if they so request it.
 
Once in a while a really juicy story is bound to surface about a long-lost ancestor, and I love to share a juicy story. However, even these may be inappropriate for a public site or blog as they may inadvertently divulge too much private information. It is best to keep these stories within the family as well – I have set up a separate password-protected page on my site for just such a purpose (I call mine “Family Jewels”).
 
It sounds all too obvious, but the genealogist must remember that everyone they collect data on is a real human being. Respect for privacy and the use of common sense must be the order of the day.

"The Spearing Numbering System (Part 3)"

The Canon or Common Law Relationship Chart showing
relationships between 2 members of a family tree

My final post regarding the numbering system is to show how the identification numbers can be modified in such a way as to accommodate moving forward in time from any blood relative so that we understand where every relative is on the tree while at the same time having no two identification numbers exactly alike.

To avoid duplication of identification numbers when the tree branches out and comes back toward the present day, we need to add a sub-level of numbering to the end of our identification numbers. If we use the same S number for a specific relative’s children we can assign them an additional digit to indicate that they are a generation removed. For example, if my father’s father had a younger brother, his identification number would be G02L001F0004S2 (2nd generation prior to my own, 1st line, sharing my grandfather’s F number of 4, sibling #2). If he had a son, that person’s identification number would be G01L001F0004S2.1 (1st generation prior to my own, 1st line, same family & sibling number but with a decimal 1 added to indicate 1st generation moving forward, 1st child). If he had a son, we would add another digit, so his identification number would be G00L001F0004S2.11, with the 2 digits representing the 2nd generation forward, 1st child. Additionally, if that person had a younger sibling, their identification number would be G00L001F0004S2.12 (only the final digit is changed, essentially this is a Sibling number within that particular family).

The system I am using for the S numbers is actually stolen from the Henry and D’Aboville descendant numbering systems already commonly used, with the small modification of using a decimal only once, between the original S number and all subsequent generations.

The only way to make real sense of all this is to illustrate it in a mock family tree. Starting with me, moving backwards a bit then coming forward again on a different branch will give an idea how it works:

Member (description) / Identification Number

Me / G00L001F0001S3
My father / G01L001F0002S6
My grandfather / G02L001F0004S1 (assuming he’s the oldest sibling)
My great-grandfather / G03L001F0008S1 (assuming he’s the oldest sibling)

Moving forward, since this is just an example, I’ll use hypothetical descendants of my great-grandfather for illustration purposes:

Great-grandfather’s sister #1 / G03L001F0008S2
Great-grandfather’s sister #2 / G03L001F0008S3
Sister #2’s husband / G03L001F0008S3M1
   (‘S3M1’ means 1st spouse of sibling #3)
Sister #2’s son #1 / G02L001F0008S3.1
   (takes mother’s S number, plus decimal 1)
Sister #2’s son #2 / G02L001F0008S3.2
Sister #2’s daughter #1 / G02L001F0008S3.3
Sister #2’s son #3 / G02L001F0008S3.4
Son #3’s 1st wife / G02L001F0008S3.4M1
Son #3’s daughter #1 / G01L001F0008S3.41
Son #3’s daughter #2 / G01L001F0008S3.42
Son #3’s 2nd wife / G02L001F0008S3.4M2
Son #3’s son #1 by 2nd wife / G01L001F0008S3.43
Son #1’s wife / G01L001F0008S3.43M1
Son #1’s child #7 / G00L001F0008S3.437
Child #7’s child #3 / G0AL001F0008S3.4373

Using the Canon Law Relationship Chart, I have determined this last relative would be my 3rd cousin once removed. According to the chart, the common progenitor at the top would be my 2nd great-grandfather, and following down the left side, I would be in box 4 (great-great-grandson). Following down the right side of the chart, from my great-grandfather’s sister in box 1 and moving down, the last relative shown above would be in box 5 (great-great-great granddaughter). Where the 2 meet in the middle of the chart is at 3rd cousins one generation removed.

Confirming this data can be done by analyzing this relative’s identification number:

G number = 0A (1 generation ahead of my own)
L number = 001 (1st bloodline – shares a common ancestor from my father’s paternal line)
F number = 0008 (a descendant of my great-grandfather’s family)
S number = 3.4373 (a direct descendant of the 3rd sibling in my great-grandfather’s family; this person is the 3rd child of the 7th child of the 3rd child of the 4th child of that ancestor)

Final Thoughts

It has been said that ancestor and descendant numbering systems should be kept separate, that they mix as well as oil and water. That may be true as the statement comes from much more experienced genealogists than me. However, in my humble opinion, anything is worth trying that will allow me to have one uniform catalogue of everyone I know of who is or has ever been related to me. The identification numbers I’m generating are lengthy by some standards perhaps, but they are meaningful and they work! Now comes the fun part - cataloguing my relatives and growing the family tree.

"The Spearing Numbering System (Part 2)"

The next problem was assigning numbers to individual “families” within family lines. To accomplish this I focused on finding a strict mathematical formula, worrying only about going backward in time and not coming forward again until later.

 

The Sosa-Stradonitz system, showing
numbering for direct ancestors

I gained inspiration from the binary number system, a system every computer geek knows and is something I learned while majoring in electronics at high school. It’s the system computers use to process data – the most basic number system there is, consisting of only ones and zeros (or back in the day, “ons” and “offs”). Considering there are only males and females in a family tree, the use of a base-2 numbering system made sense.

Family Numbers

Using myself as Family (or F) number 1, and alternating backwards between the males and females it works like this:

F Number / Represents / Binary Equivalent

0001 / Me / 000001
0002 / Dad / 000010
0003 / Mom / 000011
0004 / Dad’s Dad / 000100
0005 / Dad’s Mom / 000101
0006 / Mom’s Dad / 000110
0007 / Mom’s Mom / 000111
0008 / Dad’s Dad’s Dad / 001000
0009 / Dad’s Dad’s Mom / 001001
0010 / Dad’s Mom’s Dad / 001010
0011 / Dad’s Mom’s Mom / 001011
0012 / Mom’s Dad’s Dad / 001100
0013 / Mom’s Dad’s Mom / 001101
0014 / Mom’s Mom’s Dad / 001110
0015 / Mom’s Mom’s Mom / 001111
0016 / Dad’s Dad’s Dad’s Dad / 010000
0017 / Dad’s Dad’s Dad’s Mom / 010001
0018 / Dad’s Dad’s Mom’s Dad / 010010
0019 / Dad’s Dad’s Mom’s Mom / 010011
and so on…

There are obvious patterns with this system. First, each male is represented by an even number (“0”) while the females are odd (“1”) – nothing to do with temperament, it just worked out that way. A glance at any of the binary numbers can tell you who that person is – if the leading “1” is interpreted as the generation level minus 1, the remaining digits show the Dads and Moms. For example, F number 19 shows a 4th generation prior to my own (the leading “1” is in the 5th position from the right, minus 1), then “0011” which can be interpreted as “Dad Dad Mom Mom” or my Dad’s Dad’s Mom’s Mom.

Second, the following mathematical rules apply:

1.   Each generation going back paternally – multiply the F # by 2
2.   Each generation going back maternally – multiply the F # by 2 then add one

For example, my F number is 1, my father’s is 2 (mine x 2), and my father’s father’s is 4 (my father’s x 2) and so on. My mother’s F number is 3 (mine x 2 + 1), while her father’s is 6 (her’s x 2) and her mother’s is 7 (her’s x 2 + 1). Having such a pattern comes in quite handy for tracking purposes, and in fact this is the Sosa-Stradonitz (or Ahnentafel) system itself exactly as it is used in genealogical systems which do not include peripheral relatives. However, instead of using it to designate individuals, I am using it to designate the families to which each individual belongs. My system requires an additional (fourth) level of numbering to accommodate each individual on the tree, i.e. to include all siblings and spouses.

The purpose of the F number is to distinguish relatives who share the same G and L numbers but belong to different families. For the purpose of this numbering system, the definition of family will be “group of siblings”. And of course, the numbering is relative to my own family, therefore my brother, sister and I all have an F number of 1. My father and all six of his siblings share the same F number of 2. My mother and her siblings will have an F number of 3, and so on. Outside of my direct bloodline, spouses who have married into the family tree share the same F number as that of their partner, e.g. my mother’s sister’s husband (uncle) shares her F number of 3.

I have set aside up to 4 digits for F numbers as that many digits accommodates over 10 prior generations (going beyond the 1600’s). Unless I exceed my expectations, that should be plenty. If an additional digit is required in future, then I can just magically add one later.

Sibling Numbers

The final level of numbering in the system is the Sibling (or S) number. This is the simple way we distinguish between siblings within the same family. Known siblings are numbered according to the order of their birth, beginning at 1. Therefore in my own family, my brother has an S number of 1, my sister is 2, and I’m 3 since I am the 3rd and youngest child. Every family (or group of siblings) works this way. S numbers have only 1 digit – for families with more than 9 siblings, we use the letter ‘A’ for 10, ‘B’ for 11, ‘C’ for 12 and so on. In my grandmother’s family, one of my great-aunts has an S number of A, since she was the youngest of ten children.

The S number is at the end of the complete identification number (e.g. my own final identification number is G00L001F0001S3). Using four levels of numbering provides sufficient data to give every one of my direct blood ancestors their own unique and meaningful identification number, however we need to make slight modifications to handle my collateral relatives along with their spouses and offspring. My next blog post will explain this final piece of the puzzle…

"The Spearing Numbering System (Part 1)"

Line Number Chart, showing 'L' numbers
 with colour-coding by generation

When I began compiling genealogy data it occurred to me that it might be a prudent idea to incorporate a way of cataloguing my ancestors database-style as they come into re-existence. One quick internet search later and I had found that there are several “accepted” numbering systems out there, and it was a relief to know that at least my cataloguing method had already been figured out for me. Count that as my first mistake - all of the systems (save for one) are descendant-based, meaning that the numbering starts with the oldest ancestor and moves forward in time. This implies that your starting point has to be an ancestor that you have already determined to be the earliest entry in your historical database, which implies that my research will have to be somewhat complete before I can assign anyone a number at all. I was going backwards in time and how far back was still unknown, leaving only one option.

The “Sosa-Stradonitz” system (also known as the Ahnentafel system) is ancestor-based and it works by simply assigning each member a unique number, beginning with you as number 1. Then your father is assigned #2, your mother #3, your paternal grandfather #4, your paternal grandmother #5, etc. In this system, someone’s father’s number is always twice that person’s number, and that person’s mother’s number is twice plus one. It sounds simple and effective, and it is if you are only tracking your direct bloodline. My problem was that I wanted to include collateral relatives and for this, the widely popular Ahnentafel system alone wouldn’t cut it. It was suggested that a second system was needed to combine with it – a descendant-based one, to catch all of those stray uncles, aunts and cousins, but I just wasn’t comfortable with doing it that way. It sounded untidy, or to describe it with my own made-up word, “un-uniform”. So despite the fact that genealogical societies everywhere frown on the layman who creates his own numbering system, I decided to give it a shot.

About 13 seconds after I began laying out the framework for my numbering system I discovered that this was going to require a steady supply of eye drops and coffee. It looked like it was going to involve diagrams, flow-charts and something I had not familiarized myself with in a long time – mathematics. I envisioned the “family tree” to be like an erupting geyser – shooting out from a single point, spreading outward as it rises up as far as it can go, reaching its apex, then falling back down to the ground, spattering randomly all over the place while covering a wide radius. And I wanted a system that would give a meaningful number to every individual – one that would uniquely mark them as if it was part of their DNA. However, I did not expect that this was going to be neat and tidy… or any easier.

The idea was to build each person’s unique number from a collection of data according to where they placed themselves in the tree in relation to my position in it. My first two identifiers would be the ‘Generation’ and the ‘Family Line’ (or ‘Line’) numbers.

"Are You Your Own Cousin?"

We are one big happy family

Here’s an interesting problem – for every generation back you go, you should be able to double your number of ancestors; 2 parents, 4 grandparents, 8 great-grandparents, and so on. At 12 generations back, you should have 4,096 10th great-grandparents; at 20 generations back, over 1 million 18th great-grandparents. At 30 generations back, or roughly around the time of the “Black Death” in Europe in the middle of the 14th century, you should have 1,073,741,824 28th great-grandparents – that’s over 1 billion. The problem is that the world population just before the plague is estimated to be around only 450 million people. So what happened to a half-billion of my ancestors?

The answer is a phenomenon called pedigree collapse. It occurs when people from the same family tree procreate, and the result is a shrinking of the tree instead of the standard expansion. It has occurred in everyone’s family history – if you go back far enough (for some not so far back as others) you should find it common that 3rd, 2nd or even 1st cousins have married, thus pulling the branches of the family tree back inward, reducing the overall number of ancestors you have. British social theorist Robin Fox recently estimated that 80% of historical marriages have taken place between 1st or 2nd cousins, causing many people to take up more than one spot in their pedigree charts.

It actually had to be that way, if not more so. British genealogist Brian Pears figured out that "if every single marriage was between 2nd cousins, then 30 generations ago [residents of Britain] would all have needed exactly 4,356,616 ancestors - still more than the English population at the time" and concludes that each of Britain’s citizens around the year 1300 was an ancestor to almost every Briton currently living. Mr. Pears’ fascinating essays on “The Ancestor Paradox” can be read here.

Also, check out this article by Joelle Steele that attempts to uncover the truth about cousin marriages.

So the next time you meet up with someone of questionable character, and feel the need to fling the insult “that’s what happens when cousins marry”, just remember that somewhere in your past, it happened to you too.