Possibly… It could also allow another route for attack… I’m thinking of the Greek siege of Syracuse, where, at various times, they captured the harbor, but were prevented from going further. The Syracusans retreated to the citadel. Still, it meant that part of the city was occupied by the enemy.
Actually, I’m blanking on any other good examples of sieges or assaults on fortified harbors…
Edinburgh Castle and the Alcázar de Segovia share many similarities. Both of them are at the tip/peak of a triangular hill, with two of the castle’s sides standing on high cliffs. Both have wells inside.
The Alcázar of Toledo is less impressive, being a square fortress mostly surrounded by the city and not on top of the hill on which the city is built, but it withstood a 70-day siege in 1936. The survivors were starved but recovered well (they’d run out of food they couldn’t hunt and their allies’ food drops didn’t always hit the target). Again, a well inside.
The most important feature is having a water source that people outside can’t contaminate. One of the smallest, least impressive from afar castles I’ve ever seen wasn’t in a hill (no hills in sight even from its top) and was only a round wall (no towers, moats or anything, just a wall and a gate) and a round tower in the center, but the well was inside the tower and the tower’s tiny door was reachable only via a thin staircase. The place was as defensible as it was unimpressive.
Besalú is not a castle but a fortified village. It stands on a basalt plug in the middle of a basalt canyon; getting siege weapons to the sides of the canyon would be a complete bitch, getting them aimed properly to something that’s lower than said sides is not within the usual parameters of siege weaponry, and neither water nor food are a problem for the villagers. Finding pictures where the setup can actually be seen is difficult, I think getting the angles right would need a helicopter.
Not necessarily a well, though. Constantinople was famously hard to break, and it had underground cisterns-- at least, that’s how I remember it. I did a web search, of course, but only turned up sites about the Basilica Cistern and its value as a tourist attraction.
Yeah but those cisterns need refilling. It can be from wells, it can be from springs, or from rainfall, but it has to be something which won’t run out before the besieger’s patience does. Besalú doesn’t even have any water sources inside (well, nowadays it does, sort of; water gets pumped up from the river rather than drawn by hand), but it’s surrounded by one that can’t be contaminated or cut by besiegers.
Any castle with a Vampire as it’s lord in residence—attacks tend to be successful only in inverse proportions to the number of attackers, although attack likelihood is inversely proportional to the number of attackers as well.
Sorry. 
The siege of Tyre is probably the most famous assault on an island castle/fortification. It succeeded, but proved insanely difficult. The fact that the beseiger was Alexander the Great might have had something to do with the island’s ultimate fall.
Excellent point. This can hardly be overemphasized.
Interestingly, one of the oldest fortifications ever discovered, Jericho, already included all the major elements of castle-buiilding: thick high walls, towers spaced along the walls, and a moat. The most significant later improvement would be to make the towers project forward of the curtain wall so that flanking fire could be directed onto those trying to breach of scale the wall.
I find it fascinating that the very first builders of fortifications seem to have thought of everything important right off the bat.
It’s important not to build the foundations using some trinket that could easily be lost and destroyed.
ISTR that Krak des Chevaliers was about the most defensible castle of medieval times.
That being said, in pre-gunpowder days, most of them were pretty much totally defensible from siege engines and infantry/cavalry attack. This was a result of a couple of things- one, castles are just about never on level ground, unless they’re surrounded by water. They’re always on the top of a hill, preferably with cliffs on 3 sides. Second, they’re almost always built with a defense-in-depth plan. There’s a lower outer wall, and then a flat inner ward with few places to take refuge, and a much higher inner wall that can often fire over the outer wall. Finally, there’s usually a keep inside the inner wall somewhere that’s a fortified building, and the final line of defense. Gates are extensively fortified and have all sorts of murderously ingenious defenses built in.
Sieges were the way to go. Even catapults, mines, etc… were all part of a larger, longer siege. Nobody actually stormed castles for the most part.
Wood palisades would have been around for millenia before stone walls and most of the design flaws would have been noted during that time.
Flippant responses in serious threads irritate me, but I can’t resist adding “Have fun storming the castle!”
Here’s a list of Syrian castles–Crusader & Other. There are some amazing site/sights in that country–maybe some year…
Óbidos in Portugal is a pretty cool place. It’s a walled castle and town that sits on top of a hill. I believe it predates medieval times, dating back to the Roman era. A neat place to visit.
I can’t think of a single Spanish castle which falls into that description. Maybe Castille should ask the mods for a change of name…
In 6th grade, about 34 years ago now, we learned about castles, and the one thing I remember learning that I still remember was the realization that you shouldn’t have convex corners. That’s why most castles have rounded towers where the outside ‘corners’ ought to be.
Corners were more susceptible to being chipped/battered out, which could take place over weeks, under a wooden cover.
The older a castle, the more likely to have a sharp corner.
This sitehas lots of interesting points about castle design, defense, and attack.
Isalnd fortresses are way more useful than some here think. For one, if you have an island fortress you usually include small ships as transports, which can be utterly devastating as mobile raiders at sea or land if used against an invader. Major island forts not yet mentioned include Carnaervon and Malta. Notably, Carnaervon was never attacked (it would have a huge force a VERY long time to deal with it) and Malta, defended by the Knights Hospitalar, wrecked a huge, experienced, and well-equipped Ottoman army. In fact, they inflicted such damage that the Ottomans gave up and pestered Malta again - the blood required to take it would be enough to seriously weaken their hold on Egypt or the Balkans, so the Ottomans decided that they could live and let live. or rather, get raided by Maltese privateers.
Fortifications were less important than speedy infantry or cavalry armies in Spanish warfare. Not that they lacked fortifications, but usually used local defensive terrain over impressive structures.
Some cultures just seemed to prefer the square model, althogh it depended on the enemies they were fighting. The Romans just HATED curves in some aspects of their construction. Their roads, for example, tend to be arrow-straight and cut right through forests and hills. They did build some curved walls and towers in, especially later in the Empire, and Constantinople is a good example of Roman military engineering with curved towers.
Like in the case of the later European militaries, curves were introduced when the military need escalated, partly due to improved technology (more widespread seige technology and know-how in both cases, the trebuchet and later cannon in the Middle Ages). Square towers are also a bit faster and easier to build and fit in the the usual architectural styles of most of the Middle Ages. Rounded towers require more material and more complex construction.
These are some really fascinating replies 
Let me add an additional twist: What about the most defensible castle that was actually attacked? Did it survive? If not, how did it ultimately fall?
That would beFort Eben Emael. The Germans eventually dropped by.
[QUOTE=FlyByNight512]
If you’re looking for the castle that makes military historians go weak in the knees, I have no idea what that would be.
[/QUOTE]
Anything by Vauban.
I see what you did there.
Everything except trumpets. 
I don’t get the “useless” comments above. Folks do appreciate that castles don’t need to be in sight of trade & travel routes to control them, right? Ideally they would be, but in reality they just need to be close enough to them to support sallies forth.
Meaning, in “peacetime” or periods of minimal military threat, even an isolated island castle can still control the countryside if it has enough of a garrison present, i.e. forces protected within and sustained by the castle ride out on a regular basis to patrol & control an area. When the seige trains come to town, those forces ride back to the castle and “button up”.
