At the end of last year I fell hard down the rabbit hole that is possible bear-related place-names.
I decided for this blog to look again at the TORA data (introduced in a previous post), breaking down the small sample of bear/björn place-names by type of terrain they refer to, and by province.
I then looked again at the New Place-name Register data, as a point of comparison (as the TORA data is such a small, albeit medieval/early modern sample). I ran the björn- names through a word cloud generator to see which terrain types were the most common - and used this as a base to compare different levels of those terrain types across various provinces.
TORA bear/björn- names
For each of the bear/björn- place-names in the TORA database, I separated these out by province (Småland, Uppland, Värmland, Södermanland, and Öland), and then by word type (vatten/water, terräng/ground, naturnamn/topographical, anläggning/structure, and ägomark/agricultural). Where the name does not contain a terrain type, this is listed as 'none'.
The aim was to see whether these medieval/early modern place-names show any sort of pattern of linking bear/Björn- with certain kinds of places. Only five provinces in the database contained such names, and of these Öland only had one bear place-name: Björnhovda (Bear Head), a topographical name. I will summarise the results from the other four provinces below, in order of (roughly) most northern to least.
Uppland
Of the names from Uppland, 50% were structure related -bo names, which are interesting as this can be translated as "dwelling"/"farm" OR "animal den"! In many ways a terrible and fabulous component to have with Björn-/bear. The ground or landscape names refer to mountains, wetlands, and an isthmus.
Värmland
Once again we find names with -bo that could indicate a farm or a den-place, but the largest proportion of names are related to non-water terrain types however, while not 'lake' or 'stream' all of the 'ground' names from Värmland are actually associated with waterscapes, e.g. island, headland, sand.
Södermanland
With Södermanland we likewise find 50% of the names associated with non-water places, although unlike Värmland there is greater variation in type, with two wood names, 2 hill names, one valley name and only one coastal name. For the first time (in this dataset), we find agricultural names cropping up (meadow and field) and one topographical name (Björnfoten); as well as a similar proportion of water names (pond, mire) to Värmland.
Småland
While the proportion of direct water names are lower than our previous two examples, 57% of the ground names from Småland are waterscape related (islands or coastal names). The other ground names use hult (wood), and the structure names are mixed, with the more unambiguous Björnabygden (built-up place) and Björnstorp alongside a Björnsbo (farm or den).
I do find it interesting that, though this is a small dataset (39 names in total), there may yet be interesting things these names can suggest about regional attitudes to bear-names, if not local experiences. The overwhelming use of -bo names in Uppland can be contrasted with the mixed structure names in the more southern Småland, especially if at least some of these names may refer to animal dens. Water or waterscape names form a large proportion of the data consulted across regions, with wood names cropping up in fewer provinces. That agricultural names only appear in Södermanland so far may also indicate certain relationships with bears/the ideas of bears in this area not found elsewhere, especially in light of the lack of structure names - i.e. bears more often associated with incursions onto agricultural land, rather than the presence of their dens. Alternatively, the lack of 'structure' names in Södermanland may show naming places after the personal name Björn in this area is more likely to be applied to a field or meadow, than to a farm/dwelling in general.
New Place-Name Register
The categories I used above (vatten/water, terräng/ground, naturnamn/topographical, anläggning/structure, and ägomark/agricultural) were ones taken from the New Place-Name Register that classifies its names in these categories. It should be noted here that mosse and myr are labelled under terräng rather than vatten, therefore as a landscape not a waterscape.
When you take the Björn-/bear- names as a whole (a sample of 7399 names), it comes out like this:
which can be compared with the TORA (medieval/early modern) names as a whole:
We see almost exactly the same levels of ground/landscape names, but widely divergent proportions of anläggning/structure names; the agricultural, topographical and water proportions are also smaller in the TORA dataset - perhaps offering the smallest slice of evidence for possible changes in attitudes to place-naming over time.
Mosse, myr, and berg
As an alternate method of looking at the data, as mentioned above, I ran these locations through a word cloud generator, setting the word frequency maker to 260+. This yielded only three locales than appeared more than 260 times in the dataset of 7399 place-names: mosse, myr, and berg.
I then used these key words to filter place-names from seven provinces (6 more northern provinces, and (the most southern) Skåne as a counterpoint. Out of interest, I also included searches for tjärn (lake, woodland lake), and hult (wood):
This is still so much smaller a study than this dataset deserves - a little further than a starting point, but nowhere near what needs to be done. There is scope here for a study that looks in more detail at the terrain of these provinces, for example to see to what extent the amount of mountainous areas in a region correlates to the number of bear-mountain names (i.e. is Norrbotten just that much more mountainous than other northern regions?). I have not done that (yet) - but various things emerge from even this mid-scale study:
bear names and wetlands are consistently common, although more so in the most northern (and then in the most southern!) regions.
tjärn is a common component of these names in northern provinces, but not in the southernmost
bear/Björn- names using mosse are very prevalent in Skåne (as opposed to the other regions)
hult is not commonly used (need to check other wood words, e.g. lund, skog, etc.)
For a wider study, the aim would be to collect concepts together, such as "wood" using hult and lund etc., "wetland" using mosse, myr, etc. and "lake" using tjärn, göl, etc. (also island, bay, etc.).
Ide (winter den)
Another one of the cool things about the vast wealth of data in the place name register is that you can see which regions have the most bear-related names associated with specific place-types, for example, "ide" (hibernating den; and possibly also bol and bo, that can frustratingly mean both farm and animal den).
I used ide as a starting point, with rather dramatic results:
What is it about Dalarna that means there are a prevalence of such names?? Well, it seems that "some of the most popular areas for brown bear sightings in Sweden include the forests of Central Sweden such as Hälsingland, Dalarna and Jämtland", so it might be a) historically more bear dens to name places after, or b) a greater culture of bear-hunting with modern places named after dens, but that doesn't by itself explain such a frequency, when instances in the other counties with bears are so much lower.
Dalarna also cropped up a lot when I did a search for gudfar- names - in Norwegian godfar is a local name for bear, and out of interest I wondered whether this might be similar in Swedish - watch this space for more.
Part 3 of this blog will investigate these gudfar- names, and apply the place-name strand of this project so far to the overarching research questions of the project, as outlined in one of my first posts.
Comments