Lawrence of Arabia and the problem of preservation

In addition to all the interesting characters who attacked, defended, or lived in it in the middle ages, Byblos Castle also plays a role, if a small one, in the story of a fascinating figure from the 20th century: T. E. Lawrence, later to become famous as “Lawrence of Arabia.” In 1909, Lawrence, than an undergraduate at Oxford, embarked on a walking tour of Ottoman-controlled Syria, looking at crusader castles as part of the work that would become his thesis, The Influence of the Crusades on European Military Architecture — to the 12th Century. Among the castles Lawrence inspected was Byblos, the keep of which is, of course, the inspiration for our model.

Lawrence (left) in Syria with Leonard Woolley in 1911.

One of the interesting things about Lawrence’s work at Byblos castle is that the upper floor in his photographs looks quite different from the upper floor that we see today.


This photo from Lawrence’s thesis shows Byblos (also called Giblet, Jiblet or Jubayl) standing a good deal higher than it does today; you can plainly see a row of windows that are no longer extant. Another view of this construction appears in a collection called Architectural Lantern Slides of Lebanon:


In this image, you can also clearly see the difference in material between the lower and upper levels. At some point, then, the castle seems to have been being used as a residence, with appropriate remodeling. We don’t really know more about these changes: they were made some time before 1910, and they no longer exist today, but why they were made and when they were demolished remain unclear.


This raises an important issue that anyone trying to “reconstruct” a historical building has to confront: buildings aren’t static. When you’re trying to create a reconstruction of a building — or even if you’re trying to restore an actual historical building — you have to start by deciding exactly what it is that you’re trying to restore. Buildings change constantly, and there’s no “correct” version. You’re just making a choice to prioritize one of the building’s many forms over the others. Once again, we see that the idea of “accuracy” or “authenticity” is more complex than it seems to be.

That is, of course, if you’re even sure what the building looked like at any point in its history. Subsequent rebuilding can make it very difficult to identify earlier phases of a building’s history, and rebuilding is very common. After all, in the late medieval and post-medieval periods, most high medieval castles were considered draughty eyesores at worst and convenient sources of building stone at best. A medieval keep was useless to whoever made the alterations we see in Lawrence’s photograph, so rather than let the building go to waste they modified it to fit their needs.

Changes in the fabric of buildings are inevitable when you’re dealing with structures like Byblos castle that have long, varied histories — but without images like Lawrence’s, we’d know even less about them than we currently do.


Some more Crusades reading

Wargaming in the crusades seems to be in the air at the moment! We mentioned the recent release of Osprey’s Outremer: Faith and Blood, as well as the wide variety of other Crusading games out there. There’s also a series of articles currently running in Wargames Illustrated that should be of interest to our readers.


Steve Tibble is the author of Crusader Armies, a new book about, well, Crusader armies from Yale University Press. To tie in with this new release, he’s contributing a series of articles to WI about his work and how it could apply to wargaming. In the process, he’s challenging some of the assumptions about warfare in Outremer that often appear in medieval wargames.

We’re not going to summarise the articles here — there’s a lot to cover and they’re worth reading yourself — but there are a few points that are very relevant to Gaming a Crusader Castle, particularly in the article that appears in issue 370. In this piece, which is about sieges, Dr Tibble argues (among other points) that the crucial difference between early (mainly successful) Crusader sieges and Muslim strongholds and later (mainly unsuccessful) ones was distance from the sea. Close to ports, the Crusading armies were able to benefit from the supplies and technical expertise of mainly Italian fleets. Gangs of Italian ships’ carpenters are a pretty uncommon site on a wargaming table, but they may well have played a vital role in the success of the early Crusades. Since Italian crusaders play a big role in our project, we’re pretty excited about that fact.

Anyway, it isn’t all engineering and logistics: in the second article, which appears in issue 371, Tibble suggests that wargames err by making Crusader armies act too much like their European contemporaries and suggests that, rather than being an extension of European warfare, Crusading warfare quickly adapted to meet the challenges of a new environment, with the result that Crusader armies that look quite similar to contemporary European armies on paper might behave in quite different ways.

It’s thought-provoking stuff and it’s very relevant to this project; if you haven’t read it, it’s well worth checking out. If you’d like to discuss it, why not start a conversation here or on our Facebook group?

Building the castle part 2: the upper floor

In a previous post, we went over the decisions that went into the design of the keep’s lower floor. The upper floor is a little simpler in terms of its interpretation, but poses equal challenges for the model builder.

The upper floor of Byblos Castle was a residential space for the family that inhabited the keep. This is a common division of space in castles of this period; indeed, the upper floor is sometimes known as the “noble floor” because of its association with high-status living quarters.

In order to get to the upper level, residents and visitors would use a staircase located within the thickness of the keep’s wall. The entrance to this staircase is located next to the main entrance to the keep, meaning that visitors don’t have to pass through the lower level, with its storage areas and low-status sleeping quarters, in order to visit the lords of the castle or conduct business.

Stairs within the thickness of the wall are also very defensible. Even if attackers managed to get through the main entrance, it could still be possible to defend the upper floor. Indeed, William of Tyre records an instance in which attackers were able to destroy the door to the lower level and enter the keep. However, the defenders were able to prevent them from gaining access to the upper floor. The defenders held out until an approaching relief force made the attackers withdraw.

Although a staircase within the wall is highly defensible, it poses some challenges from a model-making perspective. Most model castles simply omit the stairs or relocate them to the interior of the building in order to simplify the task. This makes sense because in most models, the walls aren’t nearly thick enough — whether MDF of plastic, realistically thick walls would be heavy and expensive. If you’ve ever visited a medieval castle, you’ll understand why: castle walls could be metres thick.

To solve this problem, Supreme Littleness Designs created a castle that includes a double wall. Sandwiched between the inner and outer walls is the staircase. Gamers can position models on the staircase by using removable panels in the inner and outer walls.

The upper floor was far more than just a residential area, of course. Medieval buildings often included living spaces that could be used for several different purposes. In the case of a noble family, this could include activities ranging from dining to judging court cases; public activities like these served as forms of display that helped to reinforce the family’s social role.

The interior space of the upper floor might have been divided into a number of smaller areas, but it’s harder to determine what these might have been. There are small holes in the walls of the upper floor of Byblos Castle, but we don’t know exactly what these were used for. They could have been attachment points for marble cladding on the walls, which would have been yet another visible sign of the family’s wealth. Ultimately, what this means is that gamers can decorate the upper floor of the castle however they choose — with hangings, wall paintings, or panelling — and still be in line with the evidence.

Most wargaming buildings with multiple stories feature simple removable roofs that allow gamers access to the upper floor. The problem with this when designing a keep is that the roof of a keep is in use during combat. Players are going to want to have archer models lining the battlements, making it tricky to remove the roof. In the Byblos Keep model, Michael solved this problem by making only the central part of the roof removable. Because the walls are realistically thick, players can leave troops on the battlements while still having access to the whole upper floor.

The result is a model that represents as many of the features of the castle as possible while also being accessible for wargaming.

Claymore report

We’re back from Claymore, where we ran our first participation game. Although we might be biased, we think it went really well! This was our first chance to actually play a game with the spectacular keep model from Supreme Littleness Designs.

The table, almost completely set up and waiting for players to arrive. 

We got a great response — the game was completely full both times we ran it. Indeed, we had to team people up to fit a larger group in the second game, which was very satisfying!

James runs the second game at a crowded table.

Both the project and the castle were much admired, and we also got the chance to take some more photos — so look out for a post on the keep’s second floor, coming soon!

We definitely learned a few things from the scenario we ran:

  • Lion Rampant seems like it was a good choice. No one who played had played it much, and most had never played or read it, but it was easy to get them started, although the difference between activation values and attack or defense values could sometimes be a little confusing.
  • Some of the random events were a little too random. We had wanted to make them unpredictable, but a small child turning on his captors and apparently punching a man to death might be beyond plausibility.
  • We thought quite a lot about how to use the interior of the keep, but in the end it didn’t happen. The archers made good use of the battlements, but the forces of the scenario were sufficiently well balanced that in both games the attack turned into a slugfest in the village square before anyone could even think of breaking into the keep.

Overall, the games went well, and it was great to meet and talk to lots of gamers and hear their enthusiasm for the project. We’ve got a lot to think about over the next few months!

To cap it all, we won a prize:


Third place is not bad at all, given the high standard of games on display. We even got invited to put on the game at another wargames show — Hereward in Peterborough — although sadly we won’t be able to make it this year. Still, it’s really encouraging to see people responding well to the project.

In our next post, we’ll be taking a more in-depth look at the keep itself, including the upper floor and the stairs.

Building the castle part 1: the gallery

At the centre of the Gaming a Crusader Castle project is the new crusader keep model from Supreme Littleness Designs. The creation of this model provides an interesting look at the challenges of turning historical and archaeological evidence into a model usable for gaming.

People often talk about the tension between historical accuracy and ease of play, but that presumes a starting point. In medieval history, the problem is often knowing what historical accuracy is. Byblos Castle, which we chose because it represented a common type of Crusader keep, demonstrates some of these challenges.

These images of Byblos Castle come from 1934 book Les Châteaux des Croisés en Terre Sainte (Castles of the Crusades in the Holy Land). As you can see, the standing remains of the keep include a cellar, a mostly-complete first level, and a partially-complete upper level. We also don’t have a lot of information about the internal layout of the levels. That doesn’t mean we have to guess, though; we can use our knowledge of other castles and our understanding of the social structure of medieval society to make some educated interpretations.

Level 1 of Byblos Castle shows a row of square holes running along the walls. These probably served as supports for some kind of construction, but we don’t have definitive proof what it was. Based on other examples in medieval castles, Gianluca concluded that these were probably supports for a gallery of some kind. If, as was typical, the upper floor was the residence of the castle’s owners, it would follow that the lower floor served as living quarters for servants and troops, as well as for storage.


This photograph shows the gallery as Michael from Supreme Littleness implemented it. It’s a roomy space that could accommodate a pretty sizable staff or garrison. But even with the basic idea, there were still lots of decisions to make in the design. For instance, Michael had to choose where to put the staircase — would it run along the wall from the door up to the gallery, or would it be under the gallery? In the end, he chose to place the stairs at the rear as a security measure. Michael explains: “this was partly influenced by a visit to Edinburgh Castle that has a small mezzanine in the great hall – probably used as a minstrel gallery. There is no obvious way up to it at all!”


Like all wargaming models, the keep does contain some concessions to playability. The stairs are wider, with more space between them than real stairs would have; this makes it possible to place models on individual bases up to 3mm thick on them.

Working to create a historically authentic model like this crusader keep means dealing with more than one of this kind of question. In the next Building the Castle installment, we’ll be talking about another authentic but technically challenging element: the keep’s internal staircase.

If you’re interested in learning more about crusader fortifications, here’s some further reading:

Kennedy Crusader Castles
Crusader Castles by Hugh Kennedy
Nicolle crusader castles
Crusader Castles in the Holy Land by David Nicolle

Nicolle’s book combines these two titles, which you can take a look at if you’re interested in just one part of the period:

Nicolle crusader castles 1097-1192
Crusader Castles in the Holy Land 1097-1192
Nicolle 1192-1302
Crusader Castles in the Holy Land 1192-1302


Help us plan a scenario, part 3: playtest!

August 2018 will see Gaming a Crusader Castle’s first appearance at a wargames show. We’re going to be hosting a game at Claymore 2018, using the scenario we’ve been developing for Osprey Games’ Lion Rampant.

Jolanda Contarini and her retainers prepare to defend the village.

Last week, we had an opportunity to playtest the scenario at Heffers Books in Cambridge. Lion Rampant creator Daniel Mersey was in attendance and was able to offer some advice on scenario development, which was very welcome.

The previous week was spent making terrain and painting the last of the models; it was a bit of a sprint, but in the end we got it all done (This isn’t the terrain we’ll be using at Claymore itself, where we’ll have the splendid new keep from Supreme Littleness Designs to show off).

Panicked servants try to hide plunder from the raiders.

Scenario-wise, we learned some lessons, but it seems like the fundamental idea of the scenario itself is sound. We wanted to communicate several key ideas in play:

  • Castles weren’t just military or residential structures, but places of refuge in times of emergency for the civilian population.
  • Small-scale raiding was a significant problem for local landowners.
  • Groups nominally on the same side still competed politically, socially, and even militarily.
Jolanda’s archers take cover as enemies approach.

The scenario, which consists of two sides of two players each, fulfilled these roles well. Within the sides, each player had objectives that conflicted with those of their partner, meaning that it wasn’t as simple as a two-on-two battle, and the objectives of capturing loot and prisoners meant that the tactics were more complicated than just charging in.

Ghazi volunteers make their way up a flank.

There were a few things we’ll want to change — we’re working on simplifying the briefings and making the way objectives work more consistent, for instance, and we probably overdid it with the terrain — but we’re looking forward to seeing you at Claymore. We’ll be in the Atrium Hall at Table A1, right next to the Supreme Littleness booth. If you’re interested in playing the final version of the game, you can register your interest at this Doodle poll!


Help us plan a scenario, part 2: historical inspiration

In an earlier post, we talked about choosing a set of rules for our first game. We’ve spoken to a lot of people, here, on Google+, on wargaming forums and on our Facebook page, and the overall response seemed to be that people were recommending Osprey Games’ Lion Rampant for the kind of mid-sized scenario we were aiming to start with. This is good news for us, because we’re already reasonably familiar with the rules, cutting down on the preparation work for our playtest.

The defenders of Lodi Vecchio from the 2017 game.

Now that we know what rules we’ll be using, it’s time to answer the next question: what is our scenario going to be about? One of the aims of this project is to use wargaming to present the history of the Crusades, so we want to be sure that we don’t just have two sides ranked up and fighting on a rectangular battlefield. We want something that represents the distinctive nature of the period — and, of course, something that involves the castle.

When you’re looking for inspiration for a historical scenario, the best place to start looking is, of course, the primary sources. We’ll be doing an upcoming post about some of the available sources, but for now here are some of the incidents we considered when creating our scenario:

  • In 1170, an attack on the Frankish castle at Darum (Dair al-Balah, near Gaza) was almost successful: the Saracens broke their way into the tower, but the defenders mounted a spirited defense of the upper floor and held them off until reinforcements arrived and lifted the siege.
  • In 1124, the inhabitants of Ascalon raided the Frankish-controlled town of Mahumaria, capturing it. However, some of the inhabitants were able to retreat into the castle and were saved.
  • In 1182, the village of Buria was surrounded by a Turkish force. The villagers retreated into the castle, but the attackers undermined it within just four hours(!), forcing the defenders to surrender.

Most of these incidents come from the history written by the 12th-century archbishop William of Tyre, but wouldn’t be out of place in a 13th-century setting. They seem like good inspiration for a mid-sized scenario: relatively short engagements with a focus on raiding and protecting or capturing civilians. An emphasis on plunder gives us something for the attacking side to do other than simply fight, and something for the defenders to do other than simply wait behind their walls.

The basic structure for the scenario therefore looks like it’s going to be an attack on the settlement near a small keep. The defenders will need to hold the attackers off and get civilians and valuables into the tower while the attackers try to snap up as much of value as they can before reinforcements arrive. In keeping with the examples here, we’ll have Muslim forces on the attack and the Franks as the defenders.

The scenario is therefore going to feature a Muslim raiding party attacking a keep and its village while most of the defenders are away. As a small group of defenders try to get the civilians and valuables in the town to safety, the remaining defenders must hurry back to stop the raid.

We’ve thrown together some force lists and roughed out a map of a possible board setup; now we just have to source and paint a few figures for some of the characters in our scenario and write up briefings for the commanders. We’ll upload these to the blog as soon as they’re complete, so watch this space. If you want to give the scenario a go yourself and tell us what you think of it, including any improvements you can suggest, we’d be grateful!

New recommend reading page!

Part of the Gaming a Crusader Castle project is designing not only the castle itself but also scenarios for historical wargaming. We’d like these scenarios to illustrate the various roles castles played in the warfare of the crusading period.

In order to do that, we’re going to be basing our scenarios on a range of accounts from primary sources, combined with the best and most relevant scholarship on the topic. If you’re interested in doing some of your own reading on the Crusades, we’ve started a recommended reading page with links to some works we think could be helpful.

Right now, the recommended reading page has some introductory works on the history of the Crusades as well as some overviews of the military history of the period. As the project progresses, we’ll be updating it with translations of primary sources and works on more specific subjects, from the role of the Byzantine Empire in the Crusades to castles, specific campaigns, selected sections of the period, and more.

If you have a favourite work on the Crusades to recommend, why not let us know in the comments?

Help us plan a scenario!

Or: “Looking at Games with Gianluca and James.”

As we get started on developing the gaming side of the project, we want to ask our readers and the wargaming community at large for their advice.

Our first event, Claymore, is coming up in August, and we want to have a game ready for it. A lot of factors come into play: what forces do we have painted? What terrain do we have available? How much room will there be, and how much time? How many players are we aiming for?

For Claymore, we’re aiming for a game that should be playable in a pretty short amount of time, or that can be divided into phases, so that playing in it doesn’t mean committing to spending all day doing so. Two “sides” are fine, although it might be nice to split those up into sub-commands to allow more players.

We’re fortunate in that we have a pretty good selection of both Frankish and Islamic forces already painted, although we may have to rebase some of them and add a few models called for by the scenario. And with a terrain maker as partner, we’re probably all right for terrain — our goal is to suggest a small settlement near the keep itself.

We’ll be reusing some of the models from the 2017 Lodi Vecchio game.

Logistics and setup time, though, suggest a pretty small model count. We want something that has enough figures to look visually interesting, but not so many that we have to spend ages setting them up. Dozens per side, maybe, but not hundreds.

With that in mind, we sat down and looked at the big pile of books we’d acquired!


There are a lot of them, and just going through them and comparing them was an interesting exercise. Obviously we haven’t had time to try them all out, but the list so far includes:

Clash of Empires (with its expansion, Kingdom of Heaven): intended for larger battles, it seems. We might look into this for later games, though!

Deus Vult: explicitly intended for the crusades, which is nice. Has options for smaller games, although it seems mostly focused on larger ones.

Impetvs: again, mainly focused on larger battles. It seems popular, though. Like Deus Vult, we might consider it for next year’s larger games.

Kings of War Historical: it only makes sense that Mantic’s Warhammer challenger/replacement would spawn a historical game just like Warhammer did. Has a reputation for simplicity, and if a lot of players are familiar with it that could help make participation games easier.

L’Art de la Guerre: looks like a game designed for big battles using multi-based figures in units. We do have quite a lot of multi-based cavalry, so games like this one that use that system are appealing — rebasing them all might be a lot of work.

Lion Rampant: last year’s Lodi Vecchio game used this system, and it seemed to work nicely for a medium-sized battle. It’s designed for armies that are quite similar to one another, but with a little modification (maybe by importing some of the special rules from Dragon Rampant), that could be fixed — and anyway, how much complexity do we want in a one-off game? That’s not a rhetorical question.

Lords and Servants: a skirmish game, apparently designed for play by more than two players, which is always handy. We haven’t had much chance to look at it in detail yet, though.

Outremer: Faith and Blood: a warband skirmish game with lots of detail for individual models. Probably not right for a medium-sized game, but we have toyed with the idea of having a big battle simultaneous with an individual-scale game as one of our later projects.

SAGA: a SAGA warband is a pretty good size for the game we plan to run, and the game’s distinctive battle boards make armies characterful. On the other hand, they might result in too much extra paraphernalia on the table.

Soldiers of God: our Saracens are already based for this game, which is handy. It also explicitly addresses both raids and sieges, which we’re interested in. We don’t have experience with it, though, so that’s all we know.

Swordpoint: with sides in the hundreds, this game is probably too big for our early effort, but could still be worth looking at for a later game.

Whew! That’s a lot of games. Since we don’t have a lot of experience with most of them, we’re looking for recommendations. If you were planning to run convention games partly intended to show off a castle and partly intended to get people excited about the history of the crusades, which one would you choose?

Blog at

Up ↑

Create your website at
Get started