Numenorean Defensive Works

[Aldarion]

Numenorean Defensive Works

Numenorean architecture was massive. Tolkien states that in the fashion of Egyptians, Numenoreans built very large things. This is shown almost immediately upon the Fellowship arriving to old borde…

fantasycore.home.blog

Numenorean architecture was massive. Tolkien states that in the fashion of Egyptians, Numenoreans built very large things. This is shown almost immediately upon the Fellowship arriving to old borders of Gondor: Argonath, two massive statues of kings on Anduin, and the Helm’s Gate – a massive fortress said to have been built with hands of the giants. Argonath was built by Minalcar in the 13th century of the Third Age, and that was likely the last such undertaking.

Numenorean architecture however is less Egyptian and more Roman in its nature. Its greatest examples are not religious, but rather imperial. The only time when religious architecture was prominent in Numenor was during the rule of Sauron, a fallen angel: it was then that the big Temple was built in Numenor (whose description makes it a lot like Roman Pantheon, but larger), and many pyramids in Aztec style in Middle Earth, where human sacrifices were performed.

But in all other periods of time, Numenorean architecture is primarily military in nature. We have many examples: Pelargir and Umbar are both large ports, used as shelters for the massive fleets of Numenor. Osgilliath is a port and an administrative centre, while Minas Tirith and Minas Ithil both started out as fortress cities protecting the capital. And in Calenadhorn (Rohan), both major examples of Numenorean architecture are military in nature: Orthanc, a great tower watching the Gap of Rohan, and the fortress at Aglarond.

(...)

 

[Bear Ribs]

City design is one of the few places I think Tolkien actually dropped the ball (not too surprising, the guy did his research but nobody can know everything). He had a tendency to map out cities modern-style, on both sides of a river with bridges in between, but in medieval architecture, this wasn't done. Instead, the city would be entirely along one side of the river with ferries instead of bridges, this way the city itself would be able to use the river for defense along its flank, and bridges wouldn't interfere with sailships going up- and down-river and disrupt trade.

Numenor can kinda get a pass because they built on such a scale that boats could probably travel under their bridges but there's still the question of why they'd want to pass on the extremely high value of having a river along your flank blocking enemy passage.

F'rex here's a map of old Londinium:

All on one side, no bridges, fortifications near the river to block raiders sailing up to it.

And Osgiliath, built on both sides of the river with bridges everywhere including blocking access to its own marina and oddly perfectly geometric walls but no fortifications at the key points of the river where it would protect from river raiders.

 

[Aldarion]

City design is one of the few places I think Tolkien actually dropped the ball (not too surprising, the guy did his research but nobody can know everything). He had a tendency to map out cities modern-style, on both sides of a river with bridges in between, but in medieval architecture, this wasn't done. Instead, the city would be entirely along one side of the river with ferries instead of bridges, this way the city itself would be able to use the river for defense along its flank, and bridges wouldn't interfere with sailships going up- and down-river and disrupt trade.

Numenor can kinda get a pass because they built on such a scale that boats could probably travel under their bridges but there's still the question of why they'd want to pass on the extremely high value of having a river along your flank blocking enemy passage.

F'rex here's a map of old Londinium:

All on one side, no bridges, fortifications near the river to block raiders sailing up to it.

And Osgiliath, built on both sides of the river with bridges everywhere including blocking access to its own marina and oddly perfectly geometric walls but no fortifications at the key points of the river where it would protect from river raiders.

Actually, you are wrong here. There are multiple ancient and medieval cities which were on both shores of the river:

Ancient Rome:

Medieval Paris:

Medieval Norwich:

Ancient Mari:

The only question is how wide the Anduin is at Osgilliath: the cities above had no riverside walls largely because river passage could be blocked by chains. But then again, Byzantines closed off the bloody Golden Horn with a boom chain, so breadth of the river at Osgilliath may not be such an issue.

 

[Bear Ribs]

Actually, you are wrong here. There are multiple ancient and medieval cities which were on both shores of the river:

Ancient Rome:

Pretty obviously, this was all built on one side of the river, with a small later expansion across once Rome got so large and powerful that there was simply no trader planning to bypass it. Even then the fact that 90% of the city is on one side of the river is pretty telling as to how cities were built.

Medieval Paris:

Built All on one side of the river, actually. Your map just leaves off that Le Grands Boulevards goes around Paris and the part bridged is a small tributary leading to the Île de la Cité, allowing them to use the river for cover as usual and take advantage of the island.

Medieval Norwich:

Actually built all one side of the river Yare. What you're seeing running through the city is a very shallow chalk river, the Wensum, which is a small tributary of the Yare, which your map leaves off.

Ancient Mari:

The only question is how wide the Anduin is at Osgilliath: the cities above had no riverside walls largely because river passage could be blocked by chains. But then again, Byzantines closed off the bloody Golden Horn with a boom chain, so breadth of the river at Osgilliath may not be such an issue.

Not built on the river, again it uses a small tributary, actually clearly visible in this map.

It's extremely common for cities to have a smaller tributary running through them for the convenience of having a freshwater supply and dumping sewage and garbage. That's not quite the same as straddling the actual river, which was very rarely done.

 

[Aldarion]

Built All on one side of the river, actually. Your map just leaves off that Le Grands Boulevards goes around Paris and the part bridged is a small tributary leading to the Île de la Cité, allowing them to use the river for cover as usual and take advantage of the island.

Your link actually quite clearly shows the Wall of Philip Augustus encompassing equal areas on both sides of the river. Also, look at Google Maps - Bd. Peripherique follows the line of the walls (though these are early modern walls), and clearly encompasses the river.

Look also at the map of the city limits of Paris:

Even in Middle Ages, it clearly straddled the river. In fact, only the ancient walls did not straddle river. Also seen here:

This is the medieval wall of Philip Augustus. Map:

Wall of Philip II Augustus - Wikipedia

en.wikipedia.org

And the river being bridged is the River Siene itself, not some sort of a tributary, unless all the maps I've looked at are wrong.

 

[Bear Ribs]

Your link actually quite clearly shows the Wall of Philip Augustus encompassing equal areas on both sides of the river. Also, look at Google Maps - Bd. Peripherique follows the line of the walls (though these are early modern walls), and clearly encompasses the river.

If you looked as far as page two instead of quitting on the first page you'd have seen it.

And the river being bridged is the River Siene itself, not some sort of a tributary, unless all the maps I've looked at are wrong.

All the maps you looked at are wrong, or more specifically not zoomed out enough and too modern.

First, you can see how your own smaller map fits neatly over those islands in the tributary while Le Grands Boulevards goes around the city, which is built entirely on one side of the Seine unless you count the small tributary running around Notre Dame.

The second issue you're missing is that you're trying to overlay modern maps over how Paris was back then. Le Grands Boulevards doesn't exist anymore, nor is Île aux Juifs anymore an island due to massive construction projects altering the course of the river and reclaiming land. People have actually significantly modified the course of rivers, and rivers themselves sometimes change course without human participation as time passes. Where a river flows now does not always neatly fit where it was hundreds of years ago.

 

[Aldarion]

If you looked as far as page two instead of quitting on the first page you'd have seen it.

Page two is precisely where I have looked. And then at the rest of the document.

First, you can see how your own smaller map fits neatly over those islands in the tributary while Le Grands Boulevards goes around the city, which is built entirely on one side of the Seine unless you count the small tributary running around Notre Dame.

And both of these are part of the main river itself. River splitting into parts is not exactly unusual. Moreover, Le Grands Boulevards is too far away from the city itself to be considered part of the city. Paris straddles the islands, it is not on the shore of Le Grands Boulevards.

The second issue you're missing is that you're trying to overlay modern maps over how Paris was back then. Le Grands Boulevards doesn't exist anymore, nor is Île aux Juifs anymore an island due to massive construction projects altering the course of the river and reclaiming land. People have actually significantly modified the course of rivers, and rivers themselves sometimes change course without human participation as time passes. Where a river flows now does not always neatly fit where it was hundreds of years ago.

I am aware that they do not exist anymore. But Paris never was on the shore of Le Grands Boulevards. In fact, Ile De La Cite on Siene - Island of the City - is called that precisely because that is where Paris had originally been built. Island was important, Boulevards never entered the equation, at least not until much later.

If you are going to lecture me about history of Paris, you ought to first learn it yourself.

 

[Bear Ribs]

Page two is precisely where I have looked. And then at the rest of the document.

Ah, so you are making a bad-faith argument instead then.

And both of these are part of the main river itself. River splitting into parts is not exactly unusual. Moreover, Le Grands Boulevards is too far away from the city itself to be considered part of the city. Paris straddles the islands, it is not on the shore of Le Grands Boulevards.

I am aware that they do not exist anymore. But Paris never was on the shore of Le Grands Boulevards. In fact, Ile De La Cite on Siene - Island of the City - is called that precisely because that is where Paris had originally been built. Island was important, Boulevards never entered the equation, at least not until much later.

If you are going to lecture me about history of Paris, you ought to first learn it yourself.

The point --->





Your head --->

You're just nitpicking details in bad faith at this point for some reason. Was the city butted right up onshore along every part of Le Grands Boulevards? No, but I didn't say that. Did the Seine go around ancient Paris instead of the city straddling the river? Yes, which is what I said cities on rivers usually did. Could ships continue upriver because the river went around the city? Yes, which is also what I said. So what point do you think you're making here?

 

[Aldarion]

Ah, so you are making a bad-faith argument instead then.


The point --->





Your head --->

You're just nitpicking details in bad faith at this point for some reason. Was the city butted right up onshore along every part of Le Grands Boulevards? No, but I didn't say that. Did the Seine go around ancient Paris instead of the city straddling the river? Yes, which is what I said cities on rivers usually did. Could ships continue upriver because the river went around the city? Yes, which is also what I said. So what point do you think you're making here?

I'm not making bad faith argument, you are just lying. City was straddling the river, or rather, it was straddling one segment of the river which at that place had split into two channels. Fact that it may or may not have been the main channel is completely irrelevant - what you call "tributary" is actually a channel, a.k.a. part of the river itself. It is not a tributary, and it is not a minor flow either.

 

[Bear Ribs]

I'm not making bad faith argument, you are just lying. City was straddling the river, or rather, it was straddling one segment of the river which at that place had split into two channels. Fact that it may or may not have been the main channel is completely irrelevant - what you call "tributary" is actually a channel, a.k.a. part of the river itself. It is not a tributary, and it is not a minor flow either.

And now you're making up facts along with the bad faith. The fact is you already acknowledged you knew the river went around the city and presented maps with that part outside the borders as well as maps that represented it after the channel was filled to deceive readers.

Of course the fact that it wasn't the main channel is relevant and I explained why. Cities didn't straddle rivers so that boats could go up and downriver easily, and so that the river could be used for defensive purposes. Having Le Grands Boulevards going around Paris gave it those options, and the river undeniably went around the city in that path. If you prefer the American term for Tributary, "Fork" so be it, it doesn't change the actual facts on the ground that the Seine went around ancient Paris instead of it being on both sides of the river.

 

[Husky_Khan]

Guys your both pretty and I'm sure the Vikings would be happy to sack both of your cities, no matter how close to straddling both sides of the waterway you are.

 

[Aldarion]

And now you're making up facts along with the bad faith. The fact is you already acknowledged you knew the river went around the city and presented maps with that part outside the borders as well as maps that represented it after the channel was filled to deceive readers.

Of course the fact that it wasn't the main channel is relevant and I explained why. Cities didn't straddle rivers so that boats could go up and downriver easily, and so that the river could be used for defensive purposes. Having Le Grands Boulevards going around Paris gave it those options, and the river undeniably went around the city in that path. If you prefer the American term for Tributary, "Fork" so be it, it doesn't change the actual facts on the ground that the Seine went around ancient Paris instead of it being on both sides of the river.

Yes, majority of cities did not straddle rivers, but you claimed that it never happened, and that is simply false. There were a few cities which did, and many more which had a small portion of the city across the river. And while it is indeed logical to use river as a defensive channel, and thus have one side of a city face the river with nothing acorss, your statement that city would block the river traffic is completely illogical: if that was part of the reason, we would have had absolutely no cities straddling the river, even with a minor portion of the city (which, contrary to half the city being across the river, is actually seen quite often). But as I have already shown, that is not the case, and I can provide you quite a few more such examples:

Buda:

Osijek:

(granted, it is only a fortress on the opposite shore, city itself is all on one shore - but you can see the bridge across the river)

Paris:

Lyon:

Mostar:

Maestricht:

Once you have a city straddling a river, it does not matter how many houses are on the other side: if there are enough houses to warrant a bridge over the river, if there is enough anything to warrant a bridge over the river, then you have an obstruction to passage of ships, regardless of how the city is set up. And before you bring up "tributaries" in reference to the maps above again, note that Buda/Pest, Osijek and Mostar all sit on major rivers with no major canals nearby, in other words, no way to bypass the bridges between the two shores. Lyon, Paris and Maestricht have/had a way to go around the main river, but agin, not an universal rule.

So the only reason why you would not build a city in a manner of Osgiliath is defensibility. But Osgiliath was not a fortress, it was an administrative centre and a port, and considering the multiple references to bridges over Anduin, passage over the river was obviously important.

 

[Bear Ribs]

Yes, majority of cities did not straddle rivers, but you claimed that it never happened, and that is simply false.

Now you're just attributing to me strawman arguments I didn't make, and reaching ridiculous levels of pedantry in hopes of snagging a gotcha. I'm unimpressed that you're now reaching for "You're only 99% right."

There were a few cities which did, and many more which had a small portion of the city across the river. And while it is indeed logical to use river as a defensive channel, and thus have one side of a city face the river with nothing acorss, your statement that city would block the river traffic is completely illogical: if that was part of the reason, we would have had absolutely no cities straddling the river, even with a minor portion of the city (which, contrary to half the city being across the river, is actually seen quite often).

It was in fact a problem, it's well known that bridges tended to interfere with river traffic.

We think of bridges as links, providing access from one side of an obstacle to the other. But they can be obstacles themselves: hindering river traffic, for example.

There's a reason cities were built all along one side of the river, to avoid hindering river traffic and use it for defense.

Once you have a city straddling a river, it does not matter how many houses are on the other side: if there are enough houses to warrant a bridge over the river, if there is enough anything to warrant a bridge over the river, then you have an obstruction to passage of ships, regardless of how the city is set up. And before you bring up "tributaries" in reference to the maps above again, note that Buda/Pest, Osijek and Mostar all sit on major rivers with no major canals nearby, in other words, no way to bypass the bridges between the two shores. Lyon, Paris and Maestricht have/had a way to go around the main river, but agin, not an universal rule.

Well no kidding, that's why cities straddling a river are absurdly rare and weren't done. Now go ahead and try to pull another gotcha about how I didn't include a paragraph of clarifications and hemming and hawing about a couple of exceptions over several centuries because obviously I'm a Sith and deal only in Absolutes.

You've already been caught multiple times presenting falsified map evidence in this thread. I'm particularly unimpressed that you tried to sneak Paris (rotated with north facing to the right so that a casual viewer might not realize it was the same map already shown) in a second time after I already proved the river went around it in previous posts, and tried to repeat your folly. That's some impressive levels of bad faith you're showing.

As for your others, your understanding continues to be as low as your honesty. Osijek? Built in the 17th century, well after when we're discussing and had to be destroyed soon after, but I'm not surprised your history knowledge is so spotty. Buda and Pest? Your picture is of the famous chain bridge, which was a series of pontoons that could be pulled out of the way of river traffic, no permanent bridge was built until the 19th century. You can even see the individual canoe-like pontoons in your picture, it did not block river traffic nor was it able to support major land traffic itself (and didn't work in winter). As for Mostar, rather legendarily it took advantage of the terrain and massive embankments and fortifications to build a bridge so high river traffic could go underneath it, and these circumstances were so rare and the bridge so difficult to make that it's considered a UNESCO heritage site and was called the greatest architectural achievement of the entire Ottoman Balkans, which pretty handily proves how rare and difficult said achievement is. Every single one of your examples uses a unique workaround to avoid blocking river traffic, or your grasp of history was poor enough to pick bridges from far later in the timeline.

So the only reason why you would not build a city in a manner of Osgiliath is defensibility. But Osgiliath was not a fortress, it was an administrative centre and a port, and considering the multiple references to bridges over Anduin, passage over the river was obviously important.

Remind me again how many times Osgiliath got sacked, invaded, and conquered?

And Imma note, the referenced to the bridges was to Faramir and Boromir destroying the last one, and keeping on making sure no bridges existed, because they needed the river to act as a defensive barrier. That merely highlights why medieval societies liked being on one side of the river instead of straddling it.

 

[Aldarion]

Now you're just attributing to me strawman arguments I didn't make, and reaching ridiculous levels of pedantry in hopes of snagging a gotcha. I'm unimpressed that you're now reaching for "You're only 99% right."

It is nowhere close to 99%. And you were the one who claimed that it simply didn't happen. All I needed to show, technically, was one example to prove you wrong.

It was in fact a problem, it's well known that bridges tended to interfere with river traffic.

We think of bridges as links, providing access from one side of an obstacle to the other. But they can be obstacles themselves: hindering river traffic, for example.

There's a reason cities were built all along one side of the river, to avoid hindering river traffic and use it for defense.

Again, once you have a bridge across the river, it hardly matters how many more there are and whether there is a half of the city across the river. Obstruction is there. If it cannot be navigated, it cannot. If it can be, so can other bridges.

The only reason cities were built all along one side of the river was because river was used as a massive moat. The end. But even then, you often had a part of the city across the river - I do not know why it was done, but it may have been in order to ensure the ability to block the river with boom chains.

Romans were quite happy to "block off" Tiber with Milvian bridge. Both Trajan and Constantine bridged Danube, which was a massive traffic artery, and, well, Romans built stone bridges over every single major river in the Empire except for the Euphrates (which was a defensive border) and Nile. This included Rhine, another massive traffic artery which had at least four different bridges over it.

BTW, Romans used pontoon bridges as permanent bridges on rivers which were too fast / powerful to be bridged by a conventional stone bridge. Which rather brings into question the assertion that pontoon bridges in Osijek, Buda etc. were built in such a manner for ease of removal.

As for your others, your understanding continues to be as low as your honesty. Osijek? Built in the 17th century, well after when we're discussing and had to be destroyed soon after, but I'm not surprised your history knowledge is so spotty.

Pot calling the kettle black. Original bridge at Osijek was built in 16th century, not 17th (specifically, 1566.). It was also not destroyed by the Ottomans, it was burnt down by Habsburg troops in 1663. And for the purposes of the topic it simply doesn't matter what century it is, whether it is 6th or 16th or 19th, so long as it is before the invention of the steam engine. So long as you had animals pulling carts instead of the steam engine, naval and riverine traffic continued to be massively more important than land traffic, and cities would face exact same physical challenges when it comes to questions of supply and river traffic.

Buda and Pest? Your picture is of the famous chain bridge, which was a series of pontoons that could be pulled out of the way of river traffic, no permanent bridge was built until the 19th century.

Proof it was regularly done? The link gives impression bridge was only removed during winter. And pontoon bridges were often in fact permanent structures, built in lieu of stone bridges merely because they were cheaper.

Remind me again how many times Osgiliath got sacked, invaded, and conquered?

And Imma note, the referenced to the bridges was to Faramir and Boromir destroying the last one, and keeping on making sure no bridges existed, because they needed the river to act as a defensive barrier. That merely highlights why medieval societies liked being on one side of the river instead of straddling it.

And that is the reason. Nothing to do with blocking traffic. But Osgiliath was supposed to be protected by Minas Anor and Minas Tirith - it was noted in a couple of places that these two cities were mere fortresses of Osgiliath. So they may well have thought that securing bridges over the river with a city would be more important than securing a city with a river. And keep in mind that there was no known major threat extant when the three cities had been built: designs of Minas Anor and Minas Ithil actually seem rather paranoid, considering the fact that Sauron was believed to have died and the greatest known threat were the wild men.

EDIT: Also, you are clearly unfamiliar with the term head of navigation. That alone renders most of your complaints irrelevant - with most large cities (e.g. Rome, London, Paris) they didn't have to worry about bridges blocking traffic by large ships, because large ships would have been unable to proceed anyway. And now that I think about it, considering how far inland Osgiliath is, it is almost certain it is located at Anduin's head of navigation. So there is no point in not building a bridge there, considering that ships would have been unable to proceed anyway.

 

[ShadowArxxy]

EDIT: Also, you are clearly unfamiliar with the term head of navigation. That alone renders most of your complaints irrelevant - with most large cities (e.g. Rome, London, Paris) they didn't have to worry about bridges blocking traffic by large ships, because large ships would have been unable to proceed anyway. And now that I think about it, considering how far inland Osgiliath is, it is almost certain it is located at Anduin's head of navigation. So there is no point in not building a bridge there, considering that ships would have been unable to proceed anyway.

Just to step in on this point -- head of navigation is super important in early cities.

The confluence of the Columbia and Willamette rivers is pretty much an inherent place for a port city, but the *exact* location of the city is very much a matter of navigation. And historically, there were actually multiple early cities founded in that region, all of which certainly wanted to become the key port. Historically, the earliest and largest city in this region was not Portland, but Oregon City. Oregon City was also the capital of the Oregon Territory and the endpoint of the Oregon Trail, which under most circumstances should have all but guaranteed that it would become the dominant settlement.

The only reason it didn't? Navigability. The site for Oregon City was selected to take advantage of Willamette Falls for water powered lumber mills . . . but that ideal site was past the Clackamas Rapids, making it difficult for ocean traffic coming down the Willamette River from the Columbia. The other rival site was Milwaukee (now merely a suburb of Portland), but Milwaukee's position south of Ross Island again made navigation a challenge. While all three city sites were technically navigable, Portland -- just one mile further north on the west side of the Williamette, but still ten precious miles downriver from the Columbia -- was just that little bit easier to get to, and the rest is history.

Consider this: if the Clackamas Rapids didn't exist, then Oregon City would probably have remained the dominant city of the region. If, more dramatically, Williamette Falls didn't exist and the river could be navigated further south, then Newberg, OR would probably be the regional megacity instead of the comparatively small place that it is to this day. There would still be a city where Portland is -- the river is just too good of real estate to not settle -- but the big port is always going to be as far south as you can push navigation down the river, because every mile you can take the river down is time and money saved hauling cargo from the Willamette Valley basin.

Edit: Note that because the Williamette River flows "backwards" from south to north, moving downstream is going north and moving upstream is going south. This can be very confusing if you're not familiar with the area, since the vast majority of north-south rivers in the United States flow southwards. Oregon City -- and the former Linn City on the other bank of the river -- are both just to the north of Williamette Falls, but this means they are downstream of the falls and are, physically, located at the bottom of the falls rather than at the top.

 

[ShadowArxxy]

As for "cities that sprawl across rivers" -- Portland is an example of such, as it exists on both sides of the Willamette River.

On the other hand, the original settlement of Portland was strictly on the west side of the river, with the settlement on the east side of the river originally being the separate city of East Portland. The east side was considered less desirable since the land was marshy with many small creeks and sloughs; as a result, development was much slower. As a result, East Portland did not become an incorporated city until twenty years after Portland did (1871 versus 1851), even though their earliest settlement claims were almost at the same time (1846 versus 1843), and it only lasted twenty years as an independent city before being merged into its bigger neighbor.

None of the cities in this region were fortified, so that element of this entire argument is moot. As to the argument on bridges, every single one of the many bridges spanning the Willamette is built to accomodate river traffic, dating all the way back to the Madison Street Bridge of 1890 and the original Burnside Bridge of 1892, both of which were swing-open bridges.

 

[Scottty]

Moving on from the "building on both sides of the river" question... would the king send people to demolish your house if you built on the "wrong" side or something?... the real issue is how planned a city is.

Did Isildur et al draw up a map for where everything would be?

BTW - I love the detail of there being a feature named for "Tulkas Victor".

 

[ShadowArxxy]

Planned cities are historically very much the exception rather than the rule. They usually existed when a ruler decided to make a glorious new capital city, where planning was possible because *no one lived there* until the new city was built.

 

[Bear Ribs]

Moving on from the "building on both sides of the river" question... would the king send people to demolish your house if you built on the "wrong" side or something?... the real issue is how planned a city is.

Did Isildur et al draw up a map for where everything would be?

BTW - I love the detail of there being a feature named for "Tulkas Victor".

There's a few possibilities. The King probably wouldn't demolish your house but there are fair odds you'd be in another noble's territory, rivers were very popular as borders for good reason, so if the land charter said "East of the river is Count Blowhard's territory" you just gave a free house to Count Blowhard. Regardless, there's a fair chance there will be some folks living on the other side, just to take advantage of the land, but due to the difficulties of crossing the river, one side would generally be smaller and poorer, and reliant on temporary pontoon bridges, ferries, or other boats to cross, at a higher expense than living on the right side of the river.

You see something similar in more recent works with the "Wrong side of the tracks" idiom, where a city has one side of a railroad track be super-poor, due to a similar effect from the railroad owning all the land along the tracks and thus creating a similar (albeit weaker) barrier.

 

[ShadowArxxy]

There's a few possibilities. The King probably wouldn't demolish your house but there are fair odds you'd be in another noble's territory, rivers were very popular as borders for good reason, so if the land charter said "East of the river is Count Blowhard's territory" you just gave a free house to Count Blowhard.

Another example from firsthand experience: the Charles River forms the border between the cities of Boston and Cambridge. In fact, Cambridge -- originally "Newe Town" -- was settled specifically as a more defensible site than Boston, because the sheer size of the natural harbor made it impossible to fortify and defend Boston from sea-based raids. Cambridge was therefore built slightly upriver and on the other side of the Charles, across the nearest practical fording point. Both cities then proceeded to grow in parallel along the river border, connected by ferries; by the time the first bridge was built, the ford that was originally right between them was an inconvenient eight mile detour.