What are the boundary conditions for assuming that an unusual name is a misspelling of the common form?

Question

Examples from my own ancestry: Evaluating all Laidlaw births recorded in ScotlandsPeople from 1553-1855.

2260 Laidlaw births
171 Laidlay, 131 Ladlaw, 84 Laidla, 55 Laidlow. They can't all be misspellings, so I reckon they must be treated as proper names. Especially as they occur in clusters, mainly in the Selkirk area.

And only ever 1 Laidlaws, 1 Lidlaw. I reckon these unique items are misspellings.

But 2 Laidlae, 5 Laidlah, 13 Laidlau, 3 Laidlew, 2 Laudlaw.

Where does one draw the line?

Answer 1

Before about the 20th century, it doesn't make sense to talk about misspellings of names. There really was no such thing.

Yes, people made mistakes in recording names: they misheard them or misremembered them or made slips of the pen (such as adding one too many humps to an 'm'). However, the concept of "misspelling" doesn't apply, because it presupposes the existence of a correct spelling.

In a society where illiteracy is the norm, spelling doesn't count. It's the sound of the name that matters. How the clerk chose to represent that sound in writing depended on a lot of different factors, such as where he went to school (what style of spelling he learned) and the local habits of spelling and dialect. None of his choices can be said to be "correct" or "incorrect".

Comments

Absolutely, interpretation of writing styles can be very difficult. Your example of an extra hump on a lower case 'm' is one I am very familiar with.

---

Local accents can result in some very different spellings.

The parish clerk recorded a baptism of  John son of John Worster or Wiston. He annotated the entry  with:

"The name of "Wiston" is commonly pronounced as "Worster" & has been written in the Register on several former occasions."

A daughter was baptised as Sarah Wiston but married in a neighbouring parish as Worster and Worster is her maiden name on her children's birth registrations.

As to my Peasland relatives they also appear as Peasnals  and Pasewell When  one arrived in Australia he was transported as Peasland  but became a Peasley.  All, I'm certain due to thick local accents.

Answer 2

Did you look at all the original records on Scotland's people, or just look at the search results?

In many cases the names may be badly written, or not clear and then transcribed differently.

The names are not necessarily misspelled but written as the person who wrote the record thought it should be.

Handwriting has changed significantly since 1553, spelling has also changed. My thought on this question is that the names could all be slightly different spellings or transcriptions of what is now considered to be the same or a very similar name.

Another thought; if you sort the the results by place, do the name variations vary by place.

If priest or registrar A in place A always recorded the last name as Ladlaw or one of the other variations maybe it was that person's interpretation of the spelling.

What do the earliest records use as a last name? Are the earliest records all in one area or are they spread out across several areas?

This isn't a definitive answer, just things to think about.

Comments

As someone who spends considerable time with antiquarian books, M Ross is correct about wide variation in spelling even with what one considers common words. And, there were changes in what appear to be spelling conventions of particular eras. As the language changed so did spellings of names—sometimes a little, sometimes considerable, and, as noted in this thread often related to the level and type of education of the scribe. It’s (to me) an amazing thing to watch language evolve.

---

I sorted the results by location for names beginning with 'Laid' there are differences:

Ashkirk has many of the variations you have given,

Balquhidder has a few, Barony has several, as do other places. Channelkirk has lots, Colinton has almost all births as Laidlay and they are all in the 1600 and 1700s.

I did not look at place names after 'C'

There are other places where there are no variations.

It seems that there are more variations in the pre-1800 records especially in smaller places/places with fewer records.

Which would make sense if a local minister had their own way of spelling the name, or if the local minster had the bad handwriting or the records in that location were damaged in some way.

Or over time the Laidlaw spelling became more standard.

The search could also be done for each of your variations, by date and location.

It would also be possible to put the information into a spreadsheet and then sort in many different ways to look for trends by location and date.

Answer 3

I have found in my family the spelling of the last name sometimes depended on who did the writing, parish priest, minister, census takers. The "clusters" do point to the person who did the writing. I would tend to use the name of the parents for children and use other spelling in other last name(s).

Answer 4

Do your best to determine the "best" spelling, and record all of those other variations in the Other Last Names data field (separated by commas) so that people searching for one of those other spellings will not be frustrated (and will not create duplicate profiles).

WikiTree has standards and guidelines that we should consider when determining a "best" spelling. See Help: Name Fields for the WikiTree-wide standards regarding names. Many of our projects have additional project-specific naming conventions and guidelines; see Scotland - Name Field Guidelines for guidelines specific to Scotland.

Answer 5

My thanks to all who have responded, and please forgive me if I do not answer the individual posts - the result would be too fragmented.

Maybe I failed to put the problem clearly: of course, there have always been spelling variations and plain mistakes. If there is a single example of a "Lidlaw" in a period of 300 years, then to assume that this is a variation of Laidlaw is easy, and one can proceed without much risk of propagating an error.

But names do vary over time, with the result that some variants become established as names in their own right. A fellow student always gave his name as "Clark with an e" (i.e. Clarke). Once upon a time "Clarke" may have simply been an alternative to "Clark", but if we were now to relabel the 21 thousand Clarkes on WikiTree as Clarks (because that is the "standard form", with 73 thousand entries), the result would be chaos, not clarity.

In other words, there is a point at which correction becomes falsification, and I was hoping that someone could give me a guide as to where that point lies.

In my own case, I have established to my own satisfaction that "Laidla" came about as a variation of Laidlaw in the parish of Hawick, being amost entirely local and disappearing about 1730. The detailed demonstration is too involved to be presented here, and I'm not even sure if it would be of general interest.

The question - to which there is no simple answer - is "When does correction become falsification?"

Comments

Document with the primary records that you find. If there are few records, extend to the siblings and children and add variations to the OLN. Keep it simple and make no "corrections." That's how I handle it. 

---

The rule here is that the person's LNAB-last name at birth- as it is shown on their earliest record source is their last name.

You can add a name under other last names, or if there are many alternates you can put them in the biography or I often put them in Research Notes. Also recorded as ....on these sources.

Pick the most common alternate to go in Other Last names.

Occasionally I have put the name that is shown on the most sources as the LNAB and an alternate in other last names.

You have done what many others wouldn't which is to look at all the variations over time, a great idea, but for the majority of other researchers here and on other sites, they are looking for the name the person was called at birth.

It could be as in your example for Clark/Clarke that a variation became the standard form in some areas.