No, this has nothing to do with the French. It has to do with leap-frogging castles in order to get to your main goal.
This is fairly basic but maybe someone will find it useful.

Generally people agree that you should have one castle for every 4-10 cities, depending who you talk to. I agree with the middle of that scale, one castle per 6-8 cities if you want to survive (ie dominate) economically, the major factor being the strength of your factions militia troops. Better militia equals fewer castles. Italians have awesome militia and can get by with 0-2 castles. The Milanese need none, Venetians and Sicilians a few, north europeans a few more, and eastern europe has crappy militia (thats right, your Town Militia are even WORSE than Town Militia) and needs castles for their cavalry, I haven't had much exp with Muslims. So to achieve the aforementioned ratio, you will be converting a lot of captured castles to cities for the economic benefits.

What I want to point out is that in the early game often times it is advantageous to keep a captured castle as such, temporarily, before converting it to a city. Such a castle (usually at the Wooden Castle level) can be an invaluable resource for retraining the core of your conquering hordes, provided you aren't Italian and taking over the world with Italian Spear and Pavise Militia. You probably will have to upgrade it to a Castle level...castle..., but spending a few thousand florins for walls and a barracks or stables will be worth it.

It basically works like this: You march into enemy territory and take a settlement (happens to be a low level castle), maybe you fight a few battles along the way and your stack is starting to look a little worse for the wear. Keep the castle and build only military buildings. Retrain you mid level troops (Mailed Knights, Light Cav, HA, Armoured Srg, etc) and use the newly conquered castle as a staging point for the next target which is your TRUE goal. Either a really nice city or a Fortress work well as targets (Fortress is recommended because you are able to train your elite units there, and are denying your enemy the ability to do the same). Retrain your all-conquering army, march on to the next target, and convert your staging castle to a city ASAP. It is now behind the front lines and you can put a few militia in it and start reaping the economic rewards of it being a city. Also, while you are staging your next attack, the new castle basically has 1 level higher walls than the corresponding city level, so it is better to withstand any counterattacks while it is on your front line.(the CRUCIAL difference being between converting a Wooden Castle (walls you can stand on) to Palisade walled town(every wall can be knocked by rams))

Example in my Byzantine Campaign:
I take Sofia, Wooden Castle I believe, I don't really want it as a castle because I have Corinth Castle two provinces south, and Bran Fortress (Hungary) is directly north. But I keep it as a castle for a few turns to train up some Byz Cav and Skythion, which I use to take Bran Fortress. If I converted to town immediately, i would have to wait 3 turns while my forces came up from Castle Corinth. Anyway, with my new fortress at Bran secure I can convert Sofia to a city.

Another Example from Byz Campaign:
I want to take the Holy Land area to build my economy. I send my forces through Turkish lands, fighting a couple battles, and end up at Adana. My true goal is the 'Castle' castle at Acre from which i will launch attacks against Jerusalem etc, however I keep the (Wooden) castle at Adana to repel any Turkish attacks and to retrain my horses. At this point if I converted to town i would have to march forces all the way across Anatolia from Smyrna. I bypass Antioch and head straight for Acre, as soon as it falls Adana becomes a town because I can now produce good units in the area to repel any attacks.