Based on all the research I've been doing, the three "Keelan" townlands (one referencing an individual, and two referencing a surname), in my opinion have been mistaken for the given name "Caolain", and it is in actuality the abbreviation of Ceileachain (either Ceil'ain or Ceilea'n). The "Keelan" surname itself was most numerous along the Monaghan/Cavan border and the Monaghan/Meath border. Close to a millennium ago, all of these areas (even the Cavan side and Meath side) were in the territory of the Airgialla.
Rathkeelan, near Crossmaglen, Co. Armagh, near which in 1659, Seamus Pender wrote "O'Kealloghane, 6 families of the name" in his census; also backed up by the 1602 Fiant of Torlagh McHenry O'Neil's tenants, may really be the homeland of this surname. As the O'Ceileachain spent over a century as Chiefs of the Ui Breasaill Airthir (Eastern Armagh), and yet there are no townlands referencing their original tuaith. It is in this area where "Keelaghan" is said to mostly changed to "Callaghan", and specifically around Crossmaglen, Creggan parish in 1854 were 31 Callaghan housholds, far more dense than any Keelaghan/Keleghan/Callaghan cluster in South Ulster/Northern Midlands. Rathkeelan may in fact be Rath Ceileachain ("The Fortress of Ceileachain"). The 1836 Ordnance Survey map shows an ancient ringfort in the townland on a low hilltop near the Blaney Road. Could this ditched enclosure that is still viewable on Google Earth be the homestead of our eponymous ancestor? Although "Clanbrassill", ie the the territory of the Ui Breasaill Airthir in later records was quite synonymous with Orior and Oneilland baronies, it's likely that The Fews (where Rathkeelan is located, just west of Orior barony) was also included at one time or another.
Crossmakeelan, although it lies in the County Cavan, was in the tuaith of Fidh na Saithne which the O'Reillys took from Airgialla circa 1300. I believe the surname Mhic Ceileachain or Mhic Ceilea'n may have already been present here. There is less information about this townland name, as it doesn't appear on the 1609 Bodley maps, and was probably still unforfeited at the time.
Kilkeelan Co. Meath, being in an area that was heavily anglicized by Anglo-Norman settlement since the late 12th century offers a little more insight. A 1408 jury list in Athboy mentions "Wylliam O'Cwyllane of Kilchewlane". This indicates to me he was a free tenant, and it's very likely that this townland took it's name prior to the 1171 Norman invasion. If this is true, it could be that this surname spread from Armagh to Westmeath within it's first 250 years (as Ceileachain himself was noted to have died in 933) and lost it's internal fricative very early, as this "Keelan" abbreviation was already numerous. Although "Wylliam O'Cwyllane" has an O-prefix on his name, the bulk of the records in the Athboy area suggest the Mac-prefix was dominant. It could be that Wylliam was just simply stating he was a descendant of "Ceilea'n".
MacLysaght wrote that the name "gained a Mac prefix in Westmeath" but I've uncovered that the Mac-prefix probably pre-dated this family's movement to Westmeath, which is also backed up by the townland name of Crossmakeelan in Cavan and a 1663 Monaghan hearth tax record of Bryan McKeelan of Ardragh. Given that this family was among the nobility of Airgialla for at least 4 generations and kings having up to ten sons was quite the norm, there could have been numerous "Sons of Ceileachain" and "Descendants of Ceileachain" spawning both surnames at the same time and we aren't looking at any prefix aliases, but diffferent branches of Ceileachain's many sons and grandsons. And it seems "Keelan" is far more numerous than Keelaghan-variants, and if the Mac-prefix was just as widepsread it also may answer the question of who the 1602 pardoned kerns by the name of "McElane" are, who appear alongside the "O'Kellychan".
I've taken to rootschat and ancestry.com to locate Keelans from the most dense areas of this population cluster to offer to send them a Y-test. Hopefully I get some responses and we can build this genetic tree and either prove or disprove that these surnames area really all one.
Focusing just on the area around Delvin barony, Co. Westmeath reveals a ton of mutations of this name.
I had left several odd ones out in my research just to err on the side of caution, ie mostly the Kelleon (which also coincided with M'Cleon) because I though it could be related to Killeen which was very numerous in Westmeath (O'Cillin). But once I found "Kellyhan" here and thought about how the surname lost it's internal fricative, "Kelly'an" variants started to make complete sense. This occurs very close to my MDKA near Drumcree and one can see how just emphasizing certain syllables could turned it to McClane (MAC-kel-EE-ahn).
- as it seems the Ó Céileachair had some confusion on the basis of their root given-name, Woulfe wrote "companion-dear; spouse-loving", but other sources had ceileach: "wise, prudent", so I couldn't tell which was which as it looked to me to be the same word, but one being derivative of the other definitely makes sense.