You can receive the posts of this weblog by email.
| Mo | Tu | We | Th | Fr | Sa | Su |
|---|---|---|---|---|---|---|
| 1 | 2 | 3 | 4 | 5 | ||
| 6 | 7 | 8 | 9 | 10 | 11 | 12 |
| 13 | 14 | 15 | 16 | 17 | 18 | 19 |
| 20 | 21 | 22 | 23 | 24 | 25 | 26 |
| 27 | 28 | 29 | 30 | 31 |
I have now completed the reworking of the original 2003-4 study of place names and Roman roads, and have uploaded the results in A Pilot Statistical Study of Potential Roman Sites by Place Names I have also now included links to the source data.
I so admire this study Ian. I remember reading 'Roman Roads (Shire Archaeology Series)' by Richard W. Bagshawe when I was trying to understand the landscape round here in my part of Wiltshire. A really fascinating subject. I went to a talk at my local Museum yesterday (Warminster) and they were showing us a box of old wills they had just been given - including one from the time of Charles II that included an entry for a property in the area no one knew about. I love history.
 | cishanjia [Member] 29/08/07 @ 18:21 |
Thank you. I am surprised that I still cannot find any similar studies online. I would expect others must have conducted them, but perhaps are not publishing them? The only remotely similar one I have found since the original 2003-4 study is the one I have included in a link to from mine.
I remain convinced, though have not proven, that "Eccles", "Street" and now "Town" place names would be shown to have a significant association; but at a greater average distance. Unfortunately, I am not at all sure how to go about studying that.
Ian
You've probably seen these Ian:
http://www.wansdyke21.org.uk/wansdyke/wanart/mcdougall.htm
http://www.brrp.bham.ac.uk/project-info/project-info.html#1994-01
| cishanjia [Member] 30/08/07 @ 16:17 |
I had the last one, but not the others. Thank you for these.
Ian
 | Munzly pro http://poppycock.blog.co.uk/ 29/08/07 @ 18:30 |
Looking back on one's earlier work with a critical eye is always a difficult task, twenty-twenty hindsight being what it is.
Presenting that work to the view of informed critics, is also an act of bravery, however I'm relieved that you have including an extended warning on the accuracy of the results.
I think, that with a more focussed set of keywords, you could rerun the analysis and probably get some meaningful results. However, getting the right keywords would be very important.
For instance "Cold Harbour" and variants is a strong contender (meaning a "cold weather shelter" of which there were a regular supply along all important Roman roads.) Also "Casta" and variations if later castle foundations can be filtered out.
Another factor is hight (as in contour). Roman roads almost always ran along the top or near the top of hills and ridges, hence the term "High-road" - later roads, which did not have the military imperative of all-round vision, stuck to the valleys for shelter, hence "Low-road".
I would also predict that almost all actual occurrences of names including "Steet" (or variations) that turn out to be Roman, will actually lie on the actual line of the relevant Roman road.
| cishanjia [Member] 30/08/07 @ 16:31 |
Thank you for comments, as always.
I actually considered including Cold Harbour (or Coldharbour) in the original study - among several others that I quickly disregarded as absurdities. Research at the time concluded that there is some controversy over this entry; further supported by the only statistical study publicly and freely published that I have yet been able to find: http://keithbriggs.info/distance_to_roman_roads.html
Although I remember reading one suggestion that these were later namings (unfortunately, I don't recall where), I agree this is something that should be included. Keith Briggs's page reminds us, however, to include controls.
I had also considered Casta, but as you say I couldn't see any reliable way to filter this.
The most surprising result does indeed seem to have been hamlet and village names which include "Street". That's something that I am at a loss to explain, given the 2 in 11 probability of chance association (0.22). My hypothesis is that these were founded at a greater average distance from the Roman road than would show up as "in the same grid square"; several lie within 1 km of the road, but because of chance lie in a neighbouring grid square. It seems unlikely that the latter alone would account for this relatively large 'error'. The other possibility, that not all hamlets and villages containing "Street" relate to a Roman road, had also occurred to me. However, I find that problematical, not least because I have no alternate testable explanation for inclusion of "Street" (AS from via staete, "paved way"). If anyone has any suggestions in this regard, I would be grateful.
Ian
| Munzly pro http://poppycock.blog.co.uk/ 30/08/07 @ 17:18 |
The Street Called Straight (Roman Via Recta) was one of the main streets of ancient Damascus, however in the original text is just "The Straight" which is also relevant to this discussion. It seems very likely that the road was called "the straight" before the settlement appeared. Hence "Long Stratton" would be the town on the long straight, not the long straight town.
So, taking the vaguely triangular area of Norfolk bounded on two sides by Cromer-Norwich-Yarmouth and following the coast on the hypotenuse, I challenge you to find a settlement name including the word Street which doesn't have some arguable evidence of being connected to a possible Roman road.
Bearing in mind that not every Roman road would have been of military specifications for width, quality of finish and drainage.
The content of this website belongs to a private person, blog.co.uk is not responsible for the content of this website.
29/08/07 @ 17:04