* Irish / Scottish / Presbyterian Naming Practices

[davidf]

I am trying to research naming Conventions - particularly those used for determining First Names for children born into Irish, Scottish or Presbyterian families. I have seen reference to "naming conventions" applied to all three of those family types.

The convention that I have seen most often is:

Basic Naming Convention

Screenshot from 2022-07-21 15-37-43.png (75.55 KiB) Viewed 1680 times

First, is this generally recognised and when we refer to Irish and Scottish Naming Conventions do we actually mean Presbyterian Naming Conventions? Or do these conventions also apply to Catholic Families? From sites like Ireland XO, I get the impression that the practice is agnostic as to particular Christian Church.

On Find My Past, I have found reference to John B Robb, and his 2012 paper; “The Scottish Onomastic Child-naming Pattern”, which indicates an alternative convention for 3rd sons and daughters onwards:

Basic Naming + Robb's Convention

Screenshot from 2022-07-21 15-50-37.png (123.66 KiB) Viewed 1680 times

I can't see a ready logic to the order in which 3rd to 6th sons or daughters are named. I wonder whether there is a typo in the original.

Looking at the practice for 1st and 2nd sons/daughters (common to both alternatives) you can see your way through it if you imagine an ancestor's tree (with the child at the left, paternal line along the top). For sons, you read through the grand-parents top to bottom (preferencing the paternal), and for daughters, you read through the grand-parents bottom to top (preferencing the maternal).

However, for Robb's Ancestral Convention that logic does not apply. I would expect to go through the great-grand-parents in a similar manner (top to bottom for sons, bottom to top for daughters) which would give the following alternative - which makes more sense to me.

Basic Naming + Robb's Convention + Alternative

Screenshot from 2022-07-21 16-00-29.png (174.91 KiB) Viewed 1680 times

I can't from a quick search find alternative sources for the "Ancestral Convention", which make me wonder:

1. Is this "ancestral convention" (as opposed to the "parental convention" on the left of the above tables) so rare as to be not worth bothering with?
2. Has anyone seen alternative references supporting the convention documented by Robb?
3. Is there a logic behind Robb's convention that I have missed?

(For the time being I am ignoring naming children after deceased siblings or a deceased wife.)

Thanks

[AdrianBruce]

Let me weigh in with my size 6 boots (yes, really, size 6).

With a surname like Bruce, it's no surprise that I have Scots ancestry - to whit, 1/8 (despite what Ancestry's ethnicity might claim...) centred upon Dundee. In all my Scottish families I have only found two families that came close to following the so-called Scottish pattern. One was a family that a cousin of my bloodline married into. The other was actually in San Francisco where the children's maternal GF was a Dundonian and the parents (a Californian mother and Tennessean father) followed the pattern for their children (3 daughters) - except that the first and second names were reversed. And that's all I have. Just the two families.

The counter is that loads of people will say, "Well my lot follow it and they really follow it...". And I don't doubt that. I did ask in a Mailing List whether it was a Lowlands or Highlands thing but I got Highlanders dismissing such practices and Lowlanders swearing by them.

But how do you know which scenario (follow or not) you have? You can't, until you've found the children by "proper" research. So that's why I have zero faith in these conventions. Nice to use as broad hints or to review after the event but that's all.

As far as lines 4, 5 and 6 go, all the explanations that I've seen say something along the lines of "Some people claim that after the first three children, it's X, Y and Z, while others say something totally different". So it's not worth expending any time on lines 4, 5 and 6, is my view. (Unless, as Esther Rantzen used to say, "Unless you know better...")

I have never seen line 3 altered as per Robb. But perhaps the answer is that his sample just happened to do that... (And of course, there is a good chance that the father is named after his paternal GF anyway...). So Robb's Ancestral Conventions, IMHO, isn't just rare, it's downright wrong except by chance.

I've just taken a quick look at the paper and he says

And there is this rule, which applies to the Parental pattern as well [as the Ancestral]: where a given name has already been used, one slips down the list to the next designated ancestor.

No, no, no! The repeat of given names is exactly why Highlanders and Islanders ended up with with distinguishing names like "Big Eric", "Little Eric", and "Wee Eric"... Of course, I'm coming close to hoisting myself with my own petard here - I should say that some people have found some families where names were repeated.

Now, why on earth is a man of logic looking at these ideas?

[LornaCraig]

I haven’t come across the Robb’s convention before but I agree that it is not entirely logical. However, having done extensive research in my husband’s Scottish family I’ll offer some personal observations to add to Adrian’s.

The number of different forenames in use in Scotland before the 20th century was quite limited. So even when families followed the ‘parental convention’ (and I have found quite a number which did) the recognisable pattern quickly disappeared because there were so many matching names. It was very likely that in the group of 7 men comprising father plus 2 grandfathers plus 4 great grandfathers there would be a couple of pairs of matching names. Similarly for the women. So it’s hard to see how anyone could claim to discern a pattern which goes as far as ‘Father’s paternal grandfather’ etc because that increases the number of men to 15. I’m sure I have never found a Scottish family (pre 20th century) which used that many different male forenames or female forenames. (And of course they didn't have the benefit of a genealogy program to help them work out all the routes in a diagram!)

Resorting to names of uncles and aunts didn’t help much, because most of them would have been named after someone in the same ancestral pool.

Many families don’t seem to have known, or used, more than about half a dozen different names for each sex. If a family had more than that number of boys or girls they would resort to ancestral surnames (if they knew them) or biblical names, on an ad hoc basis.

Having said all that, I have occasionally struck gold when parents gave their children two forenames. In one example instead of naming one of their daughters 'Isobel' the parents named her 'Isobel Scott'. This led, eventually, to discovering where the child's father had come from, in a different part of Scotland. His mother was Isobel, maiden name Scott.

[davidf]

Now, why on earth is a man of logic looking at these ideas?

Well if that is pointed at me ...!

I have in a genetic line (as opposed to an adoptive line - so I have to be a bit discrete since I think the genetic family are unaware that one of their members "played away") a family that seems to strongly follow the "Parental Convention" and I have wondered for some time if I might discover more about my ancestor's line in County Down by trying to predict possible parental first names. The danger of course is you then chase evidence to support your conclusions ...

I am also looking at the families of a man who appears to have first married a woman sharing my surname and then (as a widower) married another woman with the same first name who according to the (English) census was born in "Scotland" and with a maiden name of Graham, that is not much to be going on! I also suspect that I have not accounted for all the children, plus the census shows that there are two couples sharing the same pre-marriage surnames married within a year of each other in the same parish. Use of the convention might help ensure (make that "give me a better chance") that I have children allocated to the correct families.

In both these cases if tentative work gets me to a possible living distant cousin I might then find that they have corroborating or contradicting evidence information.

Clutching at straws? Yup - but don't we all at the brick-walls?

I am wondering how easy it might be to write into a diagram text scheme functions to "predict" names based on ancestors (display of the results depending on which "diagnostic" flag I have set). Comparison with actual names would show whether there is the remotest chance that a convention is being followed and if it is highlight the possibility of:

• Missing siblings (predicted names being "offset"), or
• Siblings who died before the birth of a subsequent sibling
• Names of as yet untraced ancestors*

* Do that with enough families and you might get a level of consistency that points very strongly toward the probable name of a Great-Grandparent?

Always of course subject to the caveat that you are not searching for evidence to support a conclusion!

[davidf]

Resorting to names of uncles and aunts didn’t help much, because most of them would have been named after someone in the same ancestral pool.

Which is why if the object was to preserve or honour the names of recent ancestors, viewed as a whole, it does not matter whether a "4th son" is named after a particular uncle or great grandparent, because that uncle was probably named after a grandparent, so which convention (parental or ancestral) was followed is irrelevant because the choice of type of convention really is just a question of how quickly names get "passed down" to which cousin. (But to us genealogists that probably is relevant!)

It was very likely that in the group of 7 men comprising father plus 2 grandfathers plus 4 great grandfathers there would be a couple of pairs of matching names. Similarly for the women. So it’s hard to see how anyone could claim to discern a pattern which goes as far as ‘Father’s paternal grandfather’ etc because that increases the number of men to 15.

But a child's ‘Father’s paternal grandfather’ is already one of "the 7" because he would be the child's 'great grandfather' (one of four)? The somewhat clunky description, ‘Father’s paternal grandfather’ is used because we have two great grandfathers on our paternal side.

I am pondering whether to get into the probability work that Robb quotes (Robb J., 2012, The Scottish Onomastic Child-naming Pattern pages 4 & 5). If it stands up (and it is not down to "chance"), it might support work looking at the parental convention and the names for the "first 3", and the names of the two respective grand-parents and the respective parent.

[AdrianBruce]

... I am pondering whether to get into the probability work that Robb quotes (Robb J., 2012, The Scottish Onomastic Child-naming Pattern pages 4 & 5). If it stands up (and it is not down to "chance"), ...

I seriously don't follow his logic. He's just taken the probability of two children being named John and Margaret - I'd agree that's 5%. So how does this follow through to a conclusion about names and parentage? This needs Bayesian Inferencing, IIRC, i.e. the probability of the 3rd son being John, and 3rd daughter being Margaret, given that the parents are John and Margaret. I really can't remember the formula but it would need to involve a lot more than just the probability of those two names being used.... So far as I can see he's just taken two numbers not quite out of thin air. On that basis, there must be a system behind my first name being Adrian because it's so rare... But that's not so.

Maybe I'm missing it all given that it's so long since I did probability but....

[AdrianBruce]

... Clutching at straws? Yup - but don't we all at the brick-walls?...

Yes - to be clear, I don't have any problem with using structures like that to prioritise investigations.

... Always of course subject to the caveat that you are not searching for evidence to support a conclusion!

Well, exactly... Naming patterns don't support Genealogy - Genealogy supports Naming Patterns.

[LornaCraig]

But a child's ‘Father’s paternal grandfather’ is already one of "the 7" because he would be the child's 'great grandfather' (one of four)? The somewhat clunky description, ‘Father’s paternal grandfather’ is used because we have two great grandfathers on our paternal side.

Sorry yes, I think I mis-read it as paternal great-grandfather. But the fact remains that in practice it's difficult to distinguish between the 'parental convention' and the 'ancestral convention' because they both fall foul of the problem of duplication of names. The two conventions are the same for the first two sons and the first two daughters but for many families even getting that far was problematic because of duplicated names. And beyond that, in my experience most families just thought "whose name haven't we used yet?" without applying any recognisable order of priorities.

[davidf]

Maybe I'm missing it all given that it's so long since I did probability but....

Me also, which is why I am trying to see what other's think before digging out my probability notes.

Looking at Robb's reporting of IGI frequency (In the Appendix to his paper: "The 20 Most Frequent Given Names in the IGI for Scotland" - presumably over the whole of the IGI - when a restricted time period covering say the previous 2-3 generations - may be more appropriate), the chance of matching must heavily depend on which end of the distribution you are at.

IGI - First Names - per Robb

Screenshot from 2022-07-21 22-43-46.png (80.89 KiB) Viewed 1610 times

So if you are called "John" (26% of all male names are "John" per Robb), the chance of the relevant grandfather or father also being John by chance (rather than by convention) must be quite high. But if you were called Joseph (>1% of all male names per Robb), the chance of the relevant ancestor also having that name by chance must be quite small.

Which makes me a touch nervous of trying to give any statistical measure!

[davidf]

... in my experience most families just thought "whose name haven't we used yet?" without applying any recognisable order of priorities.

That may well be true, particularly in more recent times when "convention" is less important. It also will apply to later children - which may point towards the difference between the two conventions being theoretically correct but possibly rare in practice.

In respect of duplication, I am intrigued by Adrian's comments about "Big Eric", "Little Eric", and "Wee Eric" etc. I have not got any families where a "baptismal name" is repeated except where an earlier sibling had already died. I have found "apparent" slipping down to the "next in order" when there is duplication (which often causes further duplication!).

[AdrianBruce]

...
In respect of duplication, I am intrigued by Adrian's comments about "Big Eric", "Little Eric", and "Wee Eric" etc. I have not got any families where a "baptismal name" is repeated except where an earlier sibling had already died. I have found "apparent" slipping down to the "next in order" when there is duplication (which often causes further duplication!).

Yes, in the back of my mind, I remember reading that some areas in Scotland followed the pattern absolutely, including creating duplicates, as I implied above. As a result, they had to employ nicknames. In fact, I even have "eiknames" in my mind as the word. I don't think that this was just down to a small pool of names meaning that (say) John Campbell in one family was duplicated by a John Campbell in another...

The closest I can find to anything supporting my notion is in Scottish Customs: From the Cradle to the Grave by Margaret Bennett where someone was interviewed and said that some families in her knowledge had two or three Johns in various forms - e.g. John, Iain and a John Norman.

[AdrianBruce]

In fact the example that I have remembered isn't exactly that relevant - in Helen Castor's book Blood & Roses, she recounts the story of the Paston family during the Wars of the Roses. There are two brothers called John Paston, both very much alive at the same time. So far as I remember, there is no known reason for the duplication - but it must have seemed like a good idea at the time.

So not Scottish, not really our period, and possibly not pattern driven, but definitely concurrent duplication...

[AdrianBruce]

The best that I can do is show that someone else thinks the same as me. See http://homepages.rootsweb.com/~steve/namingse.htm In particular, towards the bottom where the writer says

In situations where the father and grandfather had the same name, the pattern may be altered to avoid duplication, but it is also not at all uncommon to find two sons with the same given name, i.e. altho' the repeat of a name will often mean that the eldest died in infancy, this cannot be assumed without checking the naming pattern possibilities. There are also rare occurences of 3 sons with the same name as the father and both grandfathers had the same name

Three sons with the same name! So, basically, anything is possible, if not likely.

[davidf]

In fact, I even have "eiknames" in my mind as the word.

Quite Possibly:

Sarajevo haes haed mony eiknames. The earliest is Šeher, whilk is the term Isa-Beg Ishaković uised tae descreive the toun he wis gaein tae big.