Geir Thorud’s personal notes on Genealogy in Norway

The oldest genealogies in Norway can be found in the Viking sagas, although most of them are not very reliable. Before World War II a popular activity was to construct genealogies going back to Viking kings, all have later been disproved. Many Norwegians are likely to descend from these kings, but no one can tell how (exept that there could perhaps be lines via European nobility??).

Types of genealogy publications

There are several types of genealogy publications. Ordinary family histories was the first to appear, but during the last century a very popular type has been "community history books" (Bygdebøker) that in the beginning contained the history of the major farms and the people living there, but in the last 40-50 years this has been extended to cover all places and all the people living in the community (at least those that can be found in the sources). These books go back to about 1650 and in some cases before that, and end with people living at the time of publication. They are structured according to the farms/properties. Such books have been produced for at least halve the country. One reason for structuring the documentation of families this way, compared to ordinary family histories, is to avoid duplication (overlaps) between families. (As a pioneer project Familysearch has started to transfer the material into databases, unfortunately leaving out the farm history.)

Examples of other types of publications are those that deal with medieval families (incl. nobility) or those that lists thousands of emigrants from a community to North America incl. descendants in America (after Irland, Norway was the European country that had the largest number of emigrants compared to the size of the population). Also, most local communities have history books (incl. genealogy info) that are published each year.


The most important sources are:

Although there are several source types that start before 1700, the percentage of the total population mentioned before 1700 is small (maybe 20 %, but I have not seen any statistics on this). One exception is the 1664-66 census, which covers maybe 40%, but this is my guessing.

From the type of event (type of info) you can in many cases tell the type of source, but e.g. birth dates can in the last 200 years be found in several types of records (e.g. confirmation, burrial) , although the christening record is the primary one. But if you dont have a record of the exact date of birth, you can use age, which occurs in many types of sources. You can in most cases easily determine the repository from the type of source. Many of the pre 1800 sources that are not in a tabular format are difficult to read, and it may require years of training to read them without error.

Digitization of sources

Many of the most important sources for Norway are available free on the internet, scanned from microfilm and/or transcribed. All parish records before 1910-30, land transactions before 1950, court records before 1800, probate records before 1800-50, the 1664-66 censuses, “local administrator/tax collector” records from 1600-1650 are scanned. Most censuses have been transcribed – the 1891 census will be scanned in 2011 (one page per person). Many parish records have also been transcribed. Many other sources will be published in the near future. A service with historic maps, linked to a database with administrative regions, will be available in 2011.

Thus functions in genealogy programs for reference to pages in these sources on the internet are important.

A few special aspects are:


The same comments that Adrian Bruce has on "professional standards" and the format of footnotes in the UK also applies to Norway - no standards. But we have standrads for transcription of parish records and censuses.


There are numerous organizations dealig with history and/or genealogy, there is at least one in each municipality (400+ municipalities) and several on the national level. There are two organisations especially for genealogy, Norsk slekstshistorisk forening" and DIS Norge (focusing, but not limited to, the use of computers in genealogy)..Conclusion: History is important to Norwegians.


Still to be written ....

There is probably more to say about this – maybe later.


greglamberson 2010-12-03T21:42:22-08:00

Thanks for this incredibly useful information. The experiences and special issues you deal with in your research are certainly different from what I am familiar with. Your perspective has already been very valuable, and personally, I have changed my mind completely on some things thanks to your input.

In regard to patronymics, hasn't there been some major work on the use of patronymics in genealogical research lately? I was hoping that you would know about this, because I have heard of this subject being worked on, yet I am unable to track anything down. Perhaps it's because I'm only looking in English?
gthorud 2010-12-04T07:18:35-08:00

I have included a link to Wikipedia about patronymics on the page. Patronymics are more common than one may expect.

I am not aware of any special recent work on patronymics. To me, this is not a big science, you simply use the given name of your father as a basis, and in a very few cases your mother's (matronymics). But I am sure there has been some work on statistics.

One thing to note is that, in my view, special fields in programs for patronymics are not important - the important thing is to handle several surnames. This view seems to be supported by our National archives which do not operate with patronymics in source transcriptions. But I guess there are other views on this .... there has not been much discussion about this lately.

And, there may be other experiences in other countries. Living in Europe, you quickly learn that there are several ways to do things.
gthorud 2010-12-04T08:03:32-08:00
Let me add that my thinking on "separate field for patronymic" is that proper handling of two surnames, eg. a name with two surnames should appear twice in a sorted list, would be a big step forward for patronymics - AND - it will be useful for other cases with two surnames.

I have suggested a solution that may satisfy all views in my long list of issues on names - see
greglamberson 2010-12-04T10:44:33-08:00

Somewhere somebody did some new work on patronymics in genealogy this year. I believe it had to do with properly sorting names, anticipating the proper suffix and/or automatically generating a child's patronymic name from a parent. Something like that.
gthorud 2010-12-04T16:59:53-08:00

The sorting issue is likely to be of interest.

I am not sure what the issue with prefixes are - what is a prefix in this context - son/datter? Here you will find three variants of son - son/sen/søn - no rule for what is correct.

In my view, there is some distance between on one hand designing an evidence/conclusion model where one may need to enter a lot of names from sources for one person - and on the other hand generating names automatically. What is then left for genealogists to do? For me automatic generation of patronymics is technology looking for a problem to solve.

There are periods of time here - from say 1850 to about 1930 - when you have no way of telling by any rule if a person ever used a patronymic.

There may be other needs in other countries, but I don't see the need here.
GeneJ 2011-05-07T08:22:02-07:00
Geir wrote, "separate field for patronymic, is that proper handling of two surnames, eg. a name with two surnames should appear twice in a sorted list, would be a big step forward for patronymics - AND - it will be useful for other cases with two surnames."

As Geir knows, my mother was a second generation born to Norwegian immigrants. I actually spun her family research into a separate project largely because of patronymics.

It effects my Dutch lines in New Netherlands here on this side of the pond, too (so that in once case, we have two brothers, each opting to use a different surname).

Geir touched on two surnames in his posting; it may describe the situation on my fathers side where, probably with a little help from town clerks along the way, a family name Presbury became Presson and ultimately Preston.

Hope this helps as you discover good and simple approaches for BetterGEDCOM to support these common surname issues.
dsblank 2010-12-04T13:07:23-08:00
BTW, Gramps has some of this

Gramps 3.3 (coming in March 2011) has multiple surnames, farmnames, nicknames, callnames, patronymics, and matronymics. In addition, there are plugins for cultures to have surname guessing that follows the right rules.


gthorud 2010-12-04T18:06:17-08:00

I had seen that page, but did not note that it was for the next version, so I was sort of disappointed when I did not see the multiple surnames after I installed Gramps.

I have not gone through all my issues with names and compared them to the Gramps solution, but it solves some of the issues.

I note that there is no sorting on multiple surnames in Gramps, only on the primary one. Sorting on several would solve the need for having a primary surname, or am I wrong?

Also, I have yet to understand the need to classify a surname as a patronymic. What is it used for?

When we started to transcribe church records, we used a standard that had two surname fields - one patronymic and one family name (=same surname for parent and child, and grandchild, inherited). Five years ago a new standard changed that to a single surname field - although a database may split that into several surnames since prefixes between surnames are very rare. The problem with the old way was that it was often not possible to tell if eg. Olsen was a patronymic or a family name.

Genealogy programs will have the same problem, and since I do not see any use for the classification - the classification is not printed in any reports I have seen - why should we have it? It also complicates user interfaces.

But this is in Norway, so I am not saying that no one, anywhere, has a use for it.

I was not able to find any info about the plugins.
dsblank 2010-12-04T21:30:08-08:00
Gramps does have many different types of plugins, but I misspoke about the surname guessing being one. Instead it is built-in, and set from menu -> Preferences -> Display -> Surname Guessing.

There are 4 built-in surname guessing styles types:

1) None - no surname is guessed
2) Father's surname
3) Combination of Father's and Mother's surnames
4) Icelandic

This could easily be turned into a plugin, or a new style can be added by adding a new function.

The multiple-surnames is brand new (and as you point out, and isn't released to users yet); I suspect functionality (such as sorting) will follow. New reports are created all of the time. The important thing is to get the data into the database. Reports can be then developed as people design or request them. I don't use the patronymic, so I can't comment further.