Thread by @Hal_Duncan

I need to go to fuckin bed now, like a normal person, but I may have Some Tweeting to do later about how I think I've inadvertently redeemed Ptolemy's bugfuck bonkers map of Scotland. Well, not *entirely*, but to a pretty significant extent for the Lowlands at least.

OK, so this is Ptolemy's map of Scotland. As you may notice, it's... a little wonky. Like WTAF levels of wonky, bent over at a 90º angle (and then some) like someone came at it from behind, stuck a knee in its back & snapped its frickin spine like a supersoldier getting creative.

It's absolutely batshit. You only have to look at it with the remotest idea of what Scotland looks like to think "wildly inaccurate."

BUT...

OK, so here's a good site with a translation of the textual instructions Ptolemy gave for drawing this madness: topostext.org/work.php?work_…

Clearly somewhere along the way something in his sources directions have been mixed up, but translating his directions we can still place the three main tribes of Central Scotland. We just need to read "west" for "north", east for "below", and so on.

That give us: "The Novantae dwell on the side toward the [west], [east of] the peninsula of this name... [To the east] are the Selgovae... From these toward the [north], but more [westerly], are the Damnoni.... Further east are the Otadini."

This is actually pretty accurate. The Novantae are indeed placed in Galloway. The Selgovae gave their name to Solway, it's thought. The Damnonii are placed in Ayrshire and the Clyde Valley. The Otadini or Votadini become the Gododdin of Lothian down into the eastern Borders.

So I was looking at this in relation to the placement of Arthur's battles, wondering if the Lindum in the Damnonii territory could be connected with the region of Linnuis of the Dubglas River battles, with a set of lynns (ponds) along the Douglas Water offering only a weak link.

Looking at the Lindum on there to the west of the Clyde estuary, I wondered if the estuary was drawn too wide but actually accurately reflecting how the Clyde (going upstream) bends south, with the Lanark area northeast of the head of the Douglas Water.

Spoiler: nope.

With the rotation in play, the Lindum and Alauna that Ptolemy assigns to the Damnoni are clearly to the north of the Clyde. Here we have Wikipedia's copy of an attempt to make sense of it. Lindum, Alauna & Victoria north of the Antonine Wall, mostly based on guesswork.

The Wikipedia entry for the Damnonii goes through some of the suggestions, based on looking for river names and town names and suchlike. But one thing about that map above instantly jumps out to me. The Damnonii north of the Forth?!?! en.wikipedia.org/wiki/Damnonii

Granted the Antonine Wall & Pictish incursions might change things, but this doesn't fit with the later picture: the Manua Gododdin in Clackmannanshire. Even the Votadini barely have breathing space with their huge fort on Arthur's Seat virtually overrun.

There's the Maetae too, whose territory looks strongly placeable in the Stirling area, and they're in this timeframe IIRC. Everything I've read makes the Damnonii spreading that far northeast feel Absolutely Fuckin Wrong. The placement of Lindum, Alauna & Victoria there? Crazy.

So, OK, I think. Ptolemy has rotated his Scotland 90º (and then some.) But he's giving us actually precise co-ordinates for the towns he assigns to the tribes. So what if we just plot those, rotate, and superimpose that on an actual fuckin map. So that's what I did. And... damn.

See, we have two pretty fuckin sound anchorpoints. The Novantae's Rerigonium can be identified with Dunragit, and Vindogara can be set on Vindogara Bay, which Ptolemy sets as basically the concavity of the Ayrshire coast from Ardrossan down to Ayr, with Irvine & Troon centre.

The map shows Vindogara inland, so there are wacky suggestions of Paisley and Loudon Hill for the "city"--i.e. fort probably--but the coastal data can be disregarded for now. We're only interested in the spatial relations of the cities as one set of anchorpoints.

So I trace out the pattern of dots on this version of the map, the cities handily marked with wee circles for precision unlike other versions. I then overlay on Google Maps, setting the scale & orientation with Dunragit as Rerigonium and Irvine/Troon as Vindigora. And hoo boy!

The other city of the Novantae--Lucophilia--sits bang on the Gatehouse of Fleet, where the Trusty's Hill hillfort is a recent major discovery, possibly a royal inauguration site. It's virtually sealed the deal on Rheged being centred on Galloway.

We get the Selgovae cities of Carbantorigum and Uxellum sitting bang on the mouth of the Nith and bang on Carlisle. Like, perfect placement. With Corda up Nithdale, where there's a shit ton of Iron Age hillforts. This is exactly the distribution we'd expect.

(I won't go into why the Nith makes a whole lot of sense here with the Welsh literature of the Hen Ogledd and Scottish historical records, but the Brittonic names Nudd and Nwython, Pictish Nechtan, myths & pedigrees... the Nith has left its mark in the records.)

So, we have six strong points of correspondence on this basis. I absolutely think it's safe to say that, rotation aside, Ptolemy's map is actually *really fuckin accurate* for the southwest of Scotland--Dumfries & Galloway and Ayrshire. There's one issue, but we'll come to that.

So what does this do for the rest of the Damnonii cities? Well, it places them (mostly) in highly logical places. Caria ends up on the South Calder Water just north of Cambusnethan (*cough* Nethan) extending an arc of forts: Carnwath (Caer Nuadh); Carstairs; Carluke (Caer Lugh).

Calonica ends up just southeast of Biggar, on the very edge of South Lanarkshire, not quite on that smooth arc but absolutely in line with a border between Damnonii and Votadini following the Lanark/Lothian border of today basically, here also bordering on the Selgovae south.

Lindum? Shit, Lindum falls out on the western shores of Loch Lomond, at Luss. Like literally on Luss. Lindum is (proper philologists will tell you this, it's not just my fancy) the Latinisation of the Celtic Lindon, from the "linn" root, meaning The Pool.

Alauna comes out on the east of Loch Lomond, just up from Drymen. This feels sketchier, I have to say, but Lindum at Luss makes for a *way* more logical expansion of the valley-dwelling Damnonii up through the Vale of Leven.

There's a mad outlier in Ptolemy's Victoria, which ends up at the Falls of Dochart, but hey, from the top of Loch Lomond, going through Glen Dochart, guess where that takes you out? An exceptional name signifying conquest kinda fits the geographic exception.

All of this removes the Damnonii from the northeastern turf that makes no sense for them. Ptolemy's mapping of the western Lowlands (with Loch Lomond as a kind of Lowlands bite out of the Highlands) actually holds together *really fuckin well* with later medieval distributions.

Here's the fuckin cherry on the fuckin top though. Of the three Votadini cities, *their* Alauna is also sketchy-it ends up in Crieff, again counter-intuitive, but Coria ends up just south of Peebles in the Borders: Gododdin stomping ground. And then there's Bremenium, baby...

Using Dunragit & Irvine as anchorpoints for scale & orientation, Bremenium comes out smack dab on top of fuckin Edinburgh.

The capital of the Gododdin, of the Votadini, the Otadini as Ptolemy has the. Bremenium lines up with Edinburgh perfectly.

Anyone familiar with Arthur's Battles? With the battle of Mount Agned that's been traditionally sited as Edinburgh, specified as the Castle of Maidens, the Dolorous Mountain?

The one where variant copies of Nennius sometimes name it as Breguion. Brewyn via the gw-/w- shift.

Or: Bremen in the w/v/m blurring that you get in Cumbric. And this is not just my layman fancy again. There was a Bremenium that the Romans had as a fort in Rochester in Northumberland. It's solidly identified as rooted in the native name Breguin/Brewyn attested in poetry.

So we have a corroborating tradition linking Edinburgh, via Arthur's battle list, to the name Breguion. I could make a case for why the Agned becomes attached as a dedication to St Agnes (patron of virgins, maidens) supplanting one to the Celtic goddess Brigid. But another time.

You won't be able to read my writing, lol, but here's my tracing.

No fudging, no adjustments of Ptolemy's spatial relationships, no attempt to fit a city to a place based on speculative place-name similarities and house-of-cards logic. I just traced the city points with the paper on my laptop screen, to keep Ptolemy's co-ordination 100% intact.

I then overlaid it on Google Maps, resizing Scotland on that the best I could to get the scale right, with Rerigonium & Vindogara in the right places, and traced the coastline from that. Not brilliantly, but it still works as an overlay with Google Maps at the right scale.

I've ignored his description of tribal relations in the Highlands and who there has which city, but I *did* include the city point data, and interestingly, they all sit comfortably at points on the west coast that... aren't insane.

Like, one of them comes out on the shores of Loch Linnhe, a good way down from Fort William but kinda sensible in a similar way. None of them end up in the water, lol.

The one definite misplacement is Ptolemy's Trimontium, which we know to belong in Newstead, at the foot of the Eildon Hills. This *is* within what's regarded as Selgovae territory, but Ptolemy has it way too far south. I have a theory here though.

Edinburgh being known as Bremenium has, it seems, cause a confusion between its location and that of the Rochester, Northumberland Bremenium. Someone somewhen in Ptolemy's measurement sources has tried to "correct" the "misplacement" of the northern Bremenium.

They've decided that Bremenium belongs way south of where it's set, and they've done jiggery pokery with their co-ordinates to preserve the spatial relations while swinging the whole shithouse clockwise by ~135º.

I haven't measured the angle, but turning the paper from landscape to portrait, there's still rotation needed and by eye it's not exactly another 45º, but that's roughly in the zone. I dunno what the pivot point is either or even if it *is* just round a single point. But still.

There could, I guess, be a second Trimontium that's accurately placed on the Scotland/England border in the region of Old Castleton. Or I dunno how solidly that name is attached to Newstead, whether the Eildon Hill fort might have been misidentified by that name?

I'm not inclined to dismiss that identification though without knowing a fuckload more. I'm no expert, so I'll trust those who are, and Trimontium being erroneously moved makes sense to me as part of an erroneous correction.

What I *will* say with conviction is that I think Ptolemy is actually--shock horror--*totally fuckin vindicated*. Compensating *only* for the rotation, his co-ordinates for the location of tribal centres are fuckin bang on target.

So I didn't end up finding Lindum in the place I was looking for to fit a theory. But I think I just made Ptolemy's map work--for the cities at least. And AFAIC, it cements the identification of Mount Agned with Edinburgh. Frickin Ptolemy literally puts a Brewyn on the map there.

Oh, yeah! I meant to link this site too--an interactive map of all the hill forts in Britain and Ireland. If you zoom in on Nithdale, it gives you a good sense of why I think the Selgovae city sitings are solid as you get without an actual find: hillforts.arch.ox.ac.uk

Did I science this? I think this might count as actual science. Clear predictive hypothesis: that scaling & orienting Ptolemy's cities with Dunragit & Irvine as anchorpoints, others would line up with hillfort candidates. Falsifiable: some points might end up in the water, lol.

If I had the skills, the rigorous experiment would be to plot Ptolemy's city co-ordinates in some computer cartographic software to eliminate human error, and then perform the scaling and rotation to superimpose it over a map precisely. You could then compare with known data.

If the co-ordinates are precise enough to identify a small enough region with e.g. the Cambusnethan site, which I don't think has any discoveries mapped, then you'd have a very specific prediction that could be tested with surveying. Proper science, mate.

Still done from a manual tracing, but a more legible application to an actual map. There are points where it's maybe not bang on, but it's hand-traced from a screen and I'm not sure if the dots should really be larger with the degree of precision to Ptolemy's co-ordinates.

Well, well, well! The northerly (Votadini) Alauna turns out to be situated by my reckoning on a tributary of the Tay called the River Almond. Shit. It's literally where Glenalmond shows up on the map. Update yer Wikipedia, mate, I've sorted Ptolemy's Scotland.

The southerly (Damnonii) Alauna is just down the Forth from Loch Ard, due south of the western tip of the Lake of Monteith. Where this forested area looks suspiciously moundy.

And look at my placement of Tuesis being exactly where a small fortification is on record.

Hello, Pinnata Camp. Or "Barrisdale" if you will.

Or *then again*...

Devana & Orrea are in the middle of nowhere here, and all those west coast sites look kinda inaccessible. Not impossible, but I had another look at Ptolemy's description of the coastline--the estuaries & bays--and found a Tuesis River on the east coast. So...

What if, I wondered, Ptolemy got his Highlands co-ordinates flipped around too somehow? Or his sources did? If we reverse the overlay, site Tuesis on the Tuesis River and aim for a similar ~45º angle wonkiness... let's see what happens. And lo and behold!

What happens is two of his towns of the Vacomagi line up from Tuesis with Aberdeen and Dundee. The third comes out on the North Esk in a place with no such obvious import, but still, two out of three ain't bad. And the Vacomagi *should* be on this coast by his description.

This still doesn't sort out Orrea and Devana. The flipped overlay puts them in locations (in green) that again don't instantly compute. And Devana is in the right terrain for the Taezali tribe Ptolemy sets to the north of the Vacomagi, but Orrea? Not so much.

Ptolemy has the Venicones to the south of the Vacomagi. So the purple Orrea is an experiment where I flipped the overlay yet again, took Bannatia/Dundee as anchorpoint and again went for a ~45º angle to with Orrea southward. It lands on Dunfermline.

Identifying Orrea with Dunfermline is absolutely tenuous as fuck. It's based on a single anchorpoint and a rough judgement of the angle by eye. It could absolutely be bollocks. But it does sit better with Ptolemy's description of how the tribes are distributed.

This here being, as best as I understand it, how the tribes are distributed. It's woolly as fuck anywhere you don't have towns to centre a turf on, but this *kinda* seems to make sense.