* Surname Prepositions / Prefix

[simongatward]

Hi all,

What does the collected wisdom suggest for entering surnames with prepositions? For example, I have a number of ancestors with such names as:

Michael de la Pole
Gilbert de Clare

Do I enter as 'Thomas /de Hoo/' or 'Thomas de /Hoo/'? I have tried right clicking the name and adding a Surname Prefix of 'de', but this seems to hide the prefix so it does not show anywhere when doing reports, viewing standard properties etc. If I use '/de Hoo/', all my prefixed names then get lumped together when viewing the individuals list, rather than appearing where you would expect (e.g under H for Hoo).

I would be grateful for any suggestions on the best convention to use for entering these names.

Simon

[tatewise]

Hi Simon. First a little housekeeping please, to set your FH Version in the profile on the right of your postings.
Left-click your username simongatward top right, choose Profile, and then Edit profile.
Select your Family Historian Version from drop-list, then scroll down to click Submit.

You are correct, those prefixes are a bit of a problem, and none of the extra Name sub-fields on the All tab will help.

Generally I think they should be considered part of the /Surname/.
In particular, cases like van Dyke can morph into Vandyke so need to be kept together.

In the Records Window, the Filter searches for All Surname Parts when made up of more than one word (hyphenated or not).
So if you start searching for H then all Surnames with a word starting with H will be listed, including de Hoo.

The only other option I can think of would be to enter the Primary Name as /Hoo/ without the prefix anywhere, but add an Alternate Name using /de Hoo/ which can be easily included in Reports, etc.

Sometimes using a no break space between such surname words is useful. See how_to:entered_accented_and_other_special_characters|> Enter Accented and Other Special Characters.

[AdrianBruce]

Interesting.... The "de", "de la", "van", etc, should go into the "Surname Prefix", that much is clear from the GEDCOM specification. I remain unclear whether the GEDCOM guys thought that the "de", "de la", "van", etc, should go inside the // that denote the surname. Remember(?) things like the "Surname Prefix" repeat what is already in the full name - it's just clarifying what bit is what. Or that's the way it reads to me.

What is clear from the way that FH behaves, is that putting the "de" etc inside the // seems to have effects that are less than desirable normally. For instance, I tried
Robert de /Brus/
Robert /de Brus/

The latter ends up collated in the Ds and with capitalisation including the "de", thus in the Individuals tab, I have:
BRUS, Robert de
(followed by C-names then some DA-names, then)
DE BRUS, Robert

I didn't get deep enough into reports to spot any differences beyond those obvious ones, so it's far from clear to me what use the "Surname Prefix" has in FH.

So I have the feeling that .... it depends. If you think / know that your "van Dyke" will transmogrify into "Vandyke" or similar, and you want them all grouped together then you should put the "van" inside the // - but realise that you'll see "VAN DYKE". If, like me, you would want the "de Brus" names to appear with the "Brus" names, then leave the "de" outside the // - and I'd get "de BRUS", which is nicer to my eyes.

[tatewise]

My interpretation of the GEDCOM spec is different. The Name Pieces do not define the NAME but are an alternative.
The sub Name Pieces are "optionally for systems that cannot operate effectively with less structured information." (p.34)
i.e. They are the structured solution where each part of the name needs a separate named field.
The "less structured" option is the one NAME field using the <TEXT> /<TEXT>/ <TEXT> format (p.48)

If you don't want Surnames capitalised then in Tools > Preferences > General untick Display surnames in upper case.
Then they are shown as recorded in the underlying GEDCOM database in mixed case.

As I said before, if you enter B into the Records Window Filter you get Brus, Robert de near de Brus, Robert without intervening C-names and D-names.

However, I agree it is not ideal, and is a value judgement which format is better.

[simongatward]

Thanks for the replies. As I thought, not a straightforward question to answer.

Given your comments I think I am leaning towards entering the name as Name de /Surname/ rather than Name /de Surname/, and perhaps adding the /de Surname/ version as an alternate name as suggested above.

If I was to look up one of these names in the phone book of the time (a bit of a stretch in the 1200's-1400's I must admit) I would expect to see the names recorded as say 'FERRERS, Thomas de' etc, so that all the FERRERS are grouped together alphabetically.

Also, most of these surname prepositions were gradually dropped through subsequent generations of the same family, so again it makes sense to have them all collectively grouped by surname e.g

FERRERS, Michael de
FERRERS, Thomas
FERRERS, William de

Are there likely to be any issues when exporting to other genealogy softwares or websites when using this methodology?

Simon

[tatewise]

I would not expect any problems exporting those Names to other products, other than similar ones to those experienced with FH.

[AdrianBruce]

My interpretation of the GEDCOM spec is different. The Name Pieces do not define the NAME but are an alternative. ...

You could well be right - it's not wholly clear to me, not least because the GEDCOM spec'n teeters over into defining processing at this point when it should stick to defining data.

If you don't want Surnames capitalised then in Tools > Preferences > General untick Display surnames in upper case. ....

Thank you Mike - I'd forgotten that option existed.

And yes, filtering is undoubtedly the way to pull up "de Brus" and "Brus" together (I rarely scroll to find entries these days, but just filter) but in a sense you have to remember to filter first rather than just read the data in front of you. I guess other software would use "Sort names" behind the scene to sort "de Brus", "Brus" and "Bruce" together but I tend to dislike "Sort names" - the result of losing stuff in iTunes too often and discovering an oddity in the "Sort name" or equivalent.

[michele]

Personally I'd prefer to keep a family tree compliant with the GEDCOM standard, particularly to export the data into other programs.

In this case it's hard because of two reasons:

1. The GEDCOM 5.5.1 specification is not perfectly clear on the role of name pieces, including SPFX, the "surname prefix".
2. FH doesn't provide a solution to manage the common need of sorting prefixed surnames.

About the GEDCOM specs my interpretation is:

• In the NAME tag the surname must be included between slashes: /de Hoo/.
• Optionally SURN can repeat exactly the same surname de Hoo.
• The role of SPFX, that will be de, is to indicate what to exclude from the surname to obtain the indexable string Hoo.

About FH, I'm using the version 5 and I didn't find anything that helps to sort by surname excluding the prefix.
I don't know if that has been improved in release 6 or 7. If not, it should!

[AdrianBruce]

I'm reasonably certain (i.e. not totally certain) that sort-order of prefixed names is not consistent across different cultures.

According to https://academia.stackexchange.com/ques ... bliography, the International Federation of Library Associations and Institutions has this to say about the "van" prefix:

• If the person is Dutch, "van Beukering" should be sorted under B
• If he or she is Belgian, sort it under V (but note the small print that says Belgian libraries aren't consistent across the country)
• If they're from the US, sort it under V

There is a link to a fuller version in that thread.

Editted:
The Surname-Prefix (if it were available) could then be the bit that is removed from the Surname (as Michele suggests), to give a "Sort-Surname". Quite how you handle van Beukerings from both the Netherlands and Belgium without confusing the heck out of yourself I'm not sure.

There is actually a good reason for having Sort-Names independently of van, de, von, etc prefixes - and that is when surnames are spelt differently in different sources but are clearly just variations on a theme. I once entered William Healer and Thomas Healow, because those were the baptism names but I got confused because, in my head, they are all Healow but William comes before Thomas. Currently I edit the surnames to be Healow in all cases, whereas setting the Sort-Name to xxx Healow might be better.

[tatewise]

Hi Michele,

Please see related thread How to list Names that begin with Van --- (17179).

Although I agree that keeping your data compliant with the GEDCOM standard is a good idea, it may not help much with exporting data to other programs, because few of them support the full standard. For example, most popular products do not support all the name pieces.

If you are using FH V5 or V6 then they support GEDCOM 5.5 whereas FH V7 supports GEDCOM 5.5.1.
However, the Export Gedcom File plugin converts your data into a GEDCOM that matches the target product.

Did you notice that both GEDCOM 5.5 page 34 and GEDCOM 5.5.1 page 38 advise against using name pieces:

PERSONAL_NAME_STRUCTURE
The NPFX, GIVN, NICK, SPFX, SURN, and NSFX tags are provided optionally for systems that cannot operate effectively with less structured information. For current future compatibility, all systems must construct their names based on the <NAME_PERSONAL> structure. Those using the optional name pieces should assume that few systems will process them, and most will not provide the name pieces.

In FH V7 there are new functions that allow text fields to be manipulated. So for example, if a SPFX field exists then its length in characters can be removed from the start of a prefixed Surname leaving the non-prefixed surname. Then that resulting surname can be sorted in the Records Window or Query Result Sets.
The expression is a bit cumbersome, but it works:
=RightText( %INDI.NAME:SURNAME%, NumberIf( Exists(%INDI.NAME[1].SPFX%), TextLen(%INDI.NAME[1]:SURNAME%) - TextLen(%INDI.NAME[1].SPFX%) - 1, 99 ) )

Therefore, you can either sort by full surname which is the default for the 1st Records Window column for Individuals, or by non-prefixed surname using the above expression, simply by clicking the desired column heading.

[Martin Tolley]

At the risk of causing folk even more grief the ACCR - "Anglo American Cataloging rules" - used by "most" English-speaking libraries has a lot to say about organizing authors alphabetically and therefore how the prefix should sit:

"If a surname includes an article or preposition or combination of the two, enter under the element most commonly used as entry element in alphabetically-arranged directories in the person's language or country of residence ...

Dutch: Enter under the part following the prefix unless the prefix is "ver":
Aa, Pieter van der
Brink, Jan ten
Wijngaert, Frank van den
Ver Boven, Daisy

English: Enter under prefix:
Van Buren, Martin
Le Carré, John
De la Mare, Walter

French: If the prefix consists of an article or of a contraction of an article and a preposition, enter under the prefix. Otherwise, enter under the part of the name following the prefix:
La Bruyère, René
Des Granges, Charles
Musset, Alfred de

German: If the prefix consists of an article or of a contraction of an article and a preposition, enter under the prefix. Otherwise, enter under the part of the name following the prefix:
Vom Ende, Erich
Zur Linde, Otto
Goethe, Johann Wolfgang von

Italian: Enter under prefix:
Da Ponte, Lorenzo
Della Volpaia, Eufrosino
Lo Savio, Niccolo

Spanish: If the prefix consists of an article only, enter under it. Otherwise, enter under the part following the prefix:
Las Heras, Manuel Antonio
Casas, Bartolomé de las
Soto, Hernando de

So there you have it!?