From Wikipedia, the free encyclopedia
Jump to: navigation, search
Good article Keep has been listed as one of the Warfare good articles under the good article criteria. If you can improve it further, please do so. If it no longer meets these criteria, you can reassess it.
September 1, 2011 Good article nominee Listed
          This article is of interest to the following WikiProjects:
WikiProject Military history (Rated GA-Class)
MILHIST This article is within the scope of the Military history WikiProject. If you would like to participate, please visit the project page, where you can join the project and see a list of open tasks. To use this banner, please see the full instructions.
GA This article has been rated as GA-Class on the quality assessment scale.
WikiProject Architecture (Rated GA-class, Mid-importance)
WikiProject icon This article is within the scope of WikiProject Architecture, a collaborative effort to improve the coverage of Architecture on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.
 GA  This article has been rated as GA-Class on the project's quality scale.
 Mid  This article has been rated as Mid-importance on the project's importance scale.


Made some chanegs to this page. More informative, with several links to monuments which support the definition C. Wilson 17:13, 4 December 2005 (UTC)

C. Wilson removed cleanup note. This page is accurate. CJ DUB 06:59, 16 March 2006 (UTC)

Rewrite on the Development section. Looks like somebody had a try, but didn't do very good on the language or grammar, or even factual accuracy. Fixed it up and gave some good examples. CJ DUB 03:30, 26 February 2007 (UTC)


While the "dungeon" has an additional meaning of "prison", the Dungeon article describes the same concept as keep; the articles should be merged. Duja 08:35, 13 April 2007 (UTC)

Nope. The dungeon article needs to be written different is all. IMO it is poorly written and subject to frequent point edits. Note that the donjon/keep and dungeon are quite distinct. 13:03, 13 April 2007 (UTC)


I've added a see also to reach the section on Japanese castle that touches upon the tenshaku, commonly called a "keep" in English. I was hesitant to add any lengthy discussion of the tenshaku to this here article, as it's a primarily European topic, and that's alright... I'm really quite on the fence about whether this is a WP:CSB issue, or whether it's truly a European topic and the fact that a Japanese structure happens to serve a similar purpose is mostly irrelevant. I think this is the best compromise for now... LordAmeth (talk) 00:52, 9 December 2007 (UTC)


The article was just moved to Keep (tower) without explanation in the edit summary or a discussion I can find. Certainly there is nothing here. What is the point behind this move? Nev1 (talk) 20:09, 25 July 2010 (UTC)

Agree that there is no apparent reason for this move, and that this article is clearly the primary usage of "Keep". PamD (talk) 21:46, 25 July 2010 (UTC)
Have reversed the move, so that the fortification, the primary usage, is now once more at Keep. Note that there are >500 incoming links to Keep, all of which would have to be updated if the page was moved. PamD (talk) 22:23, 25 July 2010 (UTC)
In fairness to MadonnaPenguin (talk · contribs), I can see where they were coming from as in common usage "keep" means to retain possession of something. However, Wikipedia is not a dictionary, so the primary use as far as the encyclopaedia is concerned is that of the fortification. Nev1 (talk) 22:34, 25 July 2010 (UTC)


I've gone through and had a go at expanding the article. I think I've captured the bulk of the French and English literature on the topic, and I think I've got the gist of the German work on this topic. I couldn't find much in Spanish, but I may have been looking in the wrong places. See what you think - it could no doubt use a good copy edit! Hchc2009 (talk) 16:24, 25 July 2011 (UTC)

Wow, you've done great job! That was awesome to see.--Profitoftruth85 (talk) 07:17, 29 July 2011 (UTC)


I took out "a corruption of the Latin dominarium or lordship". *Dominariu would have given *dominaire in French, not donjon. The word donjon is not a "corruption", but a gallo-roman creation with the Latin dominus as root. The evolution of the word is well attested in the different medieval latin texts, see Du Cange [1]. All the etymology dictionary say about the same thing. see the article Dungeon as well.Nortmannus (talk) 10:20, 31 January 2012 (UTC)

Keeping the conversation together, the earlier bits from my talk page:


Please, I am not finished with that. I must take this out, that is a pure non-sense.Nortmannus (talk) 09:30, 31 January 2012 (UTC)

Happy to help if I can. The current information comes from Robert Liddiard, who's a specialist in the field, but there may well be other variants as well (there usually are with the origins of words!). What's the source of the alternative you're suggesting? (e.g. a book, an article etc.) We can easily note that there are other views if there's another decent source suggesting the alternative. Hchc2009 (talk) 09:34, 31 January 2012 (UTC)
I am sorry but Robert Liddiard who is probably a good historian (I do not know him), never studied linguistics. If he did it, he would not have commited such a mistake. All the etymology dictionnary say about the same, except those who prefer a Germanic etymology, that is less probable.Nortmannus (talk) 09:46, 31 January 2012 (UTC)
Could you give a book that we could cite alongside him as an alternative sourcing? Hchc2009 (talk) 09:55, 31 January 2012 (UTC)
(Sticking my oar in as this affects at least one article I work on...) For what it's worth, I don't think Liddiard is on his own in describing donjon as a "corruption" from Latin as I think D. J. Cathcart King has stated something similar. That said, as Nortmannus pointed out I don't know if either has a background in linguistics. I'll try to track down Cathcart King's book to see if he gives a reference. I wouldn't object to noting that other sources suggest something different, as long as those sources are provided. Nev1 (talk) 09:56, 31 January 2012 (UTC)
Agree - I know there are others who follow a similar line. The key is always having good quality sources for each of the interpretations. Hchc2009 (talk) 10:11, 31 January 2012 (UTC)
  • Liddiard (and some others) state "The 12th century French came to term them a donjon, a corruption of the Latin dominarium or lordship, linking the keep and feudal authority."
  • The webpage dictionary says "Issu du lat. vulg. *dominio, -onis subst. masc. attesté au xies. au sens de « tour maîtresse » sous différentes formes (domnionus, donjo, dangio, etc. ds Du Cange, s.v. dunjo; Nierm., s.v. dominionus et Hollyman, p. 97), dé", which I don't think is in contradiction to Liddiard, whose just being more specific, as dominarium comes from domininio? (NB: my French/Latin isn't perfect though!)
  • Is the gallo-romance bit from Hoad? Hchc2009 (talk) 10:20, 31 January 2012 (UTC)


  • NB: the original phrase in Liddiard was "donjon was derived from the Latin dominarium, meaning "lordship"". By context, I'd assumed corruption, but that could well be my mistake. He seems quite clear on the Latin wording though. Hchc2009 (talk) 10:28, 31 January 2012 (UTC)

(further from my talk page)

gallo-romance is in other sources (related to the fact that the stem dominio / -ono does not exist in other Romance languages). Yes, there is a contradiction : the suffix -arium is another thing and shifted to -aire in French. Moreover dominarium : where does it come from ? It is in no Latin dictionary. Otherwise most phonetic changes are regular : saying "a corruption of the Latin dominarium" is wrong.Nortmannus (talk) 10:43, 31 January 2012 (UTC)
It's coming up as a vulgar Latin or late Latin term in my searches. How about ""The 12th century French came to term them a donjon, a derivation of the late, or vulgar, Latin dominarium or lordship, linking the keep and feudal authority."? Hchc2009 (talk) 10:55, 31 January 2012 (UTC)
That is better Hchc, but Webster explains the etymology of danger, not of donjon. Take care : Webster gives *dominarium with a star, that means "unattested" (that is the reason why it is not in the Latin dictionaries), that has given danger in French[2] and later in English, not donjon, that is another word. In this case, the evolution of the suffix -ariu is regularly -ier > -er because it became masculine, if it had become feminine it would have been -aire (as I wrote above). It is true that the first meaning of the French "danger" was "domination, empire", but the historian concludes himself, that it was the case in Late Latin, when we do not have any trace of this word and how can an historian cite an unattested word without a star and base a theory on a ghost word ? It is clear that donjon does not derive of *dominariu, but of domin(us) + suffix -o(nem). *dominariu would derive of domin(us) too. It never had the same meaning as donjon, that is a different word. If you begin to make such families of words in an article about medieval architecture, it is not the subject anymore. You can mention dominus "lord", dominate, domination, that is more relevant than *dominariu. Regards Nortmannus (talk) 05:03, 1 February 2012 (UTC)
Isn't Webster on that link also describing the Middle English word daunger meaning power or domination, and originally from dominio? (thus danger, the power of a lord to do harm) (NB: you clearly know more about linguistics than I do, so apologies if these questions seem daft!) Hchc2009 (talk) 15:55, 1 February 2012 (UTC)
Don't worry Hchc, I do not judge you, but the historians writing books : they always try to use the so-called etymologies of the words to justify all kinds of historical theories. That is something usual among them and something very widespread. Concerning daunger, it is just a variant form of "danger", that had this meaning "power, domination, empire", with the accent of western France : aun in Norman and Angevin corresponds to an in French, as explained in this famous French sentence (sorry) « Vrèi et qu'en Normandie, é ancous en Bretagne, an Anjou, é an votre Meine...iz prononcet l'a devant n un peu bien grossement é quasi comme s'il i avoèt aun par diftongue ; quand iz diset Normaund, Aungers, le Mauns, graund chose. » (it is true that in Anjou, an a before a n is pronounced as a diphtong aun....). Good evening.Nortmannus (talk) 20:16, 1 February 2012 (UTC)
  • How about... ""The 12th century French came to term them a donjon, which has been explained by castleologist Robert Liddiard as a derivation of the late Latin term dominarium or lordship, and by etymologist T. Hoad as deriving from the earlier Latin stem dominus, lord; in both explanations, the origin of the word links the keep with the exercise of feudal authority."?
Sure, it is much better, but dominarium with * please : *dominarium. Dominarium without a star is a mistake.Nortmannus (talk) 09:36, 3 February 2012 (UTC)
I'm not sure that the star will mean much to the non-specialist reader (it will look like a typo) - how about if we capture that bit in a footnote? Hchc2009 (talk) 09:38, 3 February 2012 (UTC)
Ok Hchc, good idea, the footnote. Good luckNortmannus (talk) 15:35, 6 February 2012 (UTC)
Will sort it out in a couple of days (am about to be travelling with work) - but I think we've got the solution here; thanks Nortmannus. Hchc2009 (talk) 21:22, 6 February 2012 (UTC)


I've reverted the latest changes, because they're either unreferenced or seem to be running counter to the cited references. In particular:

  • Conisborough: As per the reference, Brown describes this keep as a "most striking" instance of a "polygonal keep" (pp.52-53).
  • Trim Castle: I've strengthened this with an additional reference to Lise Hull, who calls Trim "cruciform" in design.
  • Gaillard: the website you're citing seems to agree with the point made by Liddiard: the keep at Gaillaird has no well. That's what Liddiard says, and the website says that there was one in the enceinte and one in the outside yard - but no mention of one within the keep.

If you think there's a problem with these, happy to chat further.Hchc2009 (talk) 07:14, 14 March 2012 (UTC)

Misuse of the term "Tower Keep" wrong century...[edit]

Tower keep has been misused in this article to describe the 14th cen. generation of large residential keeps, to differentiate from other military structures.

The term "tower keep" is a term in the literature, and institutions such as English Heritage, to describe older 12th century keeps. R.A. Brown describes it as a Norman keep where the lords hall is stacked above the retainers hall. The contrasting arrangement, an earlier design, is called the "hall keep" where the retainer (great hall) and lord's hall are placed side by side. The archetype for the tower keep would be Rochester, whilst for a hall keep the best example is the White Tower, at the Tower of London. IIRC Brown considers all later military keeps after the 12th century, Rouen, Skenfrith; etc to be tower keeps.

Ingleton, 2012 provides an updated reference of the terminology:

Here is another by Gravett, 2001 in reference to Colchester (hall keep) and Hedingham (tower keep):

If you look in the literature, you'll find most great 14th century keeps to be referred to as keeps, residential keeps, or sometimes tower houses. tks-- (talk) 22:00, 6 September 2014 (UTC)

I've reverted for now, and replaced with a "Dubious tag" pending discussion, as the changes don't match up with the cited sources in the article.
The term is used in much of the literature this way, as cited in the article; Pounds talks about tower-keeps throughout the late medieval period, for example; Allen Brown discusses the tower keep through the 15th century (including Warkworth etc.) English Heritage use the term as well: their official listings for Warkworth as an ancient monument and as a listed building describe it as such. The main tower at Raglan is similarly often described as a tower keep. The main focus of these historians is to draw continuity between the 11/12th century, and the 14/15th, although I'd note that the later examples given do also meet the definition you've linked to above (they are taller than they are wide).
You're right to say that the classic hall keep / tower keep distinction is definitely one most typically used in discussions of the Anglo-Norman period, although - as noted above - the tower keep is term used in the later medieval period as well.
I've not seen the term "residential keep" used very much at all in the literature, though, and a quick search I did just now didn't throw up much evidence for it. I'd agree with you, however, that I have also seen the term "tower house" used by some authors (Emery, for example, favours it) for these buildings, and of course some authors just use the term "keep".
My suggestion would be that we build in a referenced comment that "tower house" is an alternative term here, and perhaps add an explanatory footnote. What do you reckon? Hchc2009 (talk) 07:53, 7 September 2014 (UTC)