Skip to main content

Full text of "A social history of ancient Ireland : treating of the government, military system, and law ; religion, learning, and art; trades, industries, and commerce; manners, customs, and domestic life, of the ancient Irish people"

See other formats

Fir.. 167A 

Shrink of St. Patrick's Bell, made l>v order of Donall O'l.ochlin, Wing of Ireland idied ira ) 

now in the National Museum, Dublin Sec vol. I., p 374, supra 

(From Miss Stokes's Early Christian Art in Ireland.) 





The Government, Military System, and Law ; 

Religion, Learning, and Art,; Trades, Industries, and Commerce , 

Manners, Customs, and Domestic Life, 

of the Ancient Irish People 



LL.D., TRIN. COLL., DUB. ; M.R.I.A. 
One of the Commissioners for the Pullication of the Ancient Laws of Ireland 




Fir:. 167E— Cormac's Chapel on the Rock of Cashel. 
From Miss Stokes's Early Christian Architecture in Inland 



Printed and Bound 
in Ireland by :: :: 
M. H. Gill &- Son, 
:: :: Lid. :: :: 
jo Upper O'Connell 
Street :: :: Dublin 

First Edition 
Second „ 
Third Impression 


Ornament composed from the Book of Kells. 







i. Marriage, .. 

2. Position of Women and Children, 

3. Fosterage, .. 

4. Family Names, 




i. Construction, Shape, and Size, 

2. Interior Arrangements and Sleeping Accommod 


3. Outer Premises and Defence, 

4. Domestic Vessels, 

5. Royal Residences, 



1. Meals in General, 

2. Drink, . . 

3. Cooking, 

4. Flesh Meat and its Accompaniments 

5. Milk and its Products, .. 

6. Corn and its Preparation, 

7. Honey, 

8. Vegetables and Fruit, . . 

9. Fuel and Light, . . . . 
to. Free Public Hostels 














i. The Person and the Toilet, 

2. Dress, 

3. Personal Ornaments, .. 

4. Rough Classified List of Gold Objects in the 

National Museum, 










i. Fences, 

2. Land, Crops, and Tillage 

3. Some Farm-Animals, 

4. Herding, Grazing, Milking 






i. Chief Materials, .. 

2. Builders, 

3. Brasiers and Founders, 

4. The Blacksmith and his Forge, 

5. Carpenters, Masons, and other Craftsmen, 

6. Protection of Crafts and Position of Craftsmen, 







i. History, .... 


2. The " Eight Parts " of a Mill, 


3. Small Mills, 

• • 338 

4. Drying and Grinding, 


5. Common Property in Mills, 


6. Querns and Grain Rubbers, 

... 34J 





i. Wool and Woollen Fabrics, 

2. Flax and its Preparation, 

3. Dyeing, 

4. Sewing and Embroidery, 

5. Tanning, 

6. Workers in Leather, and the Articles they made, 





1. Length and Area, .. .. .. .. .. 371 

2. Capacity, . . . . . . . . . . . . . . 375 

3. Weight, . . . . . . . . . . . . . . 377 

4. Standards of Value and Mediums of Exchange, . . 380 

5. Time, . . . . . . . . . . . . . . 387 

6. Enumeration, . . . . . . . . . . . . 392 



i. Roads, Bridges, and Causeways, 

2. Chariots and Cars, 

3. Horse-riding, . . . . . . 

4. Communication by Water, 

5. Foreign Commerce, 





1. The Great Conventions and Fairs, .. .. .. 434 

2. The Fair of Carman, .. .. .. .. .. 441 

3. General Regulations for Meetings, . . . . 447 

4. Some Animals Connected with Hunting and Sport, 451 

5. Races, . . . . . . . . . . . . . . 462 

6. Chase and Capture of Wild Animals, . . . . 466 

7. Caman or Hurling, and other Athletic Games, . . 474 

8. Chess, . . . . . . . . . . . . . . 477 

9. Jesters, Jugglers, and Gleemen, .. .. .. 481 






i. Salutation, 

2. Pledging, Lending, and Borrowing, .. 

Provision for Old Age and Destitution, . . 
Irish Poetry and Prosody: Love of Nature 

of Natural Beauty, 
Six Stages of Life, 
Human Temperaments, . . 

7. Blood-Covenant, 

8. Cremation-Ashes thrown into Water, 

9. Something further about Animals, 

10. Animals as Pets, .. 

11. The Cardinal Points 

12. The Wind, 

13. The Sea 

14. Bishop Ultan and the Orphans, 

15. Prophecies of Irish Saints, 

16. Sundry small matters worthy of notice, 










1 Wills, 

2. Funeral Obsequies, 

3. Modes of Burial, 

4. Cemeteries, 

5. Sepulchral Monuments, 









Sculpture on a Capital : Priest's House, Glendalough: Beraneer, 1779. 
(From Petrio's Round Towers.) 






167A. Shrine of St. Patrick's ") Fronlis- 

193. Stone drinking-cup, ... 


Bell. J 


194. Do., do., . . . < 


167B. Cormac's Chapel, Rock "l 
of Cashel. ) 


195. Carrickfergus Castle, 



196. Bronze vessel, hammered, 


168. Clochan or beehive-shaped house,. 


197. Bronze vessel, cast, . 


169. Maynooth Castle, .... 


198. The Kavanagh drinking-horn, 


170. Antique wooden hut, 


199. Ancient Irish wooden vessel, . 


171. Coloured glass plate, enamelled, 


200. Figure of man drinking, . 


172. Coloured glass ornament, 




202. Do 


174. Another 


203. Ancient wooden pail, . . 





205. Glazed earthenware pitcher, . 





207. The Forradh mound at Tara, . 


179. King John's Castle, Limerick 


208. Mound of Dinnree palace, . 


180. Plan of ancient Irish homestead, 


209. Mound of Naas palace, . . 


181. Bunratty Castle in Clare, 


210. Carbury Castle, Kilo are, . . 


182. Plan of interior of ancient Iris] 


211. Rock of Cashel, .... 





213. Small antique table, . . 


184. Great Moat of Kilfinnane, 


214. Bronze strainer, . . 


185. Section of underground hut, . 


215. Ancient bronze caldron, . 


186. Staigue Fort in Kerry, . . 


216. Ancient wooden butter-print, . 

. 137 

187. Dun Aengus in Aran, . . 


217. Firkin of bog-butter, . . 


188. Stone fort of Ballykinvarriga, 


. 164 


219. Do., .... 

. 164 

190. Dundrum Castle, Co. Down, . 

. 64 

. 164 

191. Section of crannoge, . 

. 66 

221. Ancieut hronze lamp, 

. 165 

192. Crannoge village, . . . 

. 6? 

222. Bronze figures of ecclesiastics, 

. 176 



223. Specimen of plaited hair, . . 180 

224. Ancient comb, ..... 181 

225. Do., i8x 

226. Do., 181 

227. Ancient bronze razor, . . .184 

228. Small antique gold box, . . . 188 

229. Angel, from Book of Kells, . . 197 

230. Evangelist, from Book of Kells, . 197 

231. Figures on a book-cover, . . . 200 

232. Mac Murrogh Kavanagh and Earl 

of Gloucester 201 

233. Figures on St. Manchan's shrine, . 204 

234. Bronze pin, 206 

235. Do., 206 

236. Do., • • • • . 206 

237. Do., 206 

238. Do., 206 

239. Do., . . . • . 206 

240. Do., 206 

241. Bronze button, ..... 206 

242. Figure showing trousers, . . . 208 

243. Group of figures, 16th century, . 210 

244. Figures, showing costume, . . 211 

245. Do., do., . . 21 1 

246. Do., do., . . 211 

247. Portion of veil, ..... 216 

248. Ornamented shoe, .... 217 

249. Do., .... 217 

250. Sandals with rosettes, ... 218 

251. Pair of shoes connected ... 220 

252. Gold bracelet 224 

253. Bronze bracelet, .... 224 

254. Gold finger-ring, .... 225 

255. Jet bead for necklace or fastener, . 227 

256. Do., do., . 227 

257. Do., do., . . 227 

258. Gold bead for necklace, . . . 228 

259. Do., do., . . . 228 

260. Gold torque, 2 jx 

261. Do., 232 

262. Gold crescent or necklet of the 

first type, 234 

263. Do., do., . . . 235 


264. Gold crescent or necklet of the 

second type 236 

265. Do., do., . . . 237 

266. Front of gold boss of crescent, . 238 

267. Do., do., . 238 

268. Gold crescent or necklet of the 

third type, 239 

269. Gold Bunne-do-at or fibula, . . 241 

270. Do., do., . . 241 

271. Do., do., . . 241 

272. Do., do., . . 242 

273. Small Bunne-do-at, .... 243 

274. Do., do., . . . 243 

275. Great Bunne-do-at in Trinity 

College, 244 

276. Gold circular ornamental plate, . 245 

277. Bronze brooch 249 

278. Bronze spring brooch, . . . 249 

279. Crowned Irish king 257 

280. Enamelled fillet of a crown, . . 258 

281. Spurious Irish crown, . . . 259 

282. Gold earring, 260 

283. Hollow gold balls 262 

284. Pillar-stone in a rath, . . . 267 

285. A holed-stone, 268 

286. Ancient Irish bronze reaping-hook, 273 

287. Do., do, . . . 273 

288. Yoke for oxen or horses, . . . 275 

289. Brasier's anvil, 296 

290. Do., ..... 296 

291. Inlaid metallic hook, ... 299 

292. Spear-bead, 299 

293. Mould for spear-head, . . . 300 

294. Mould for Celt, 300 

295. Do., 300 

296. Plan of two-chamber smith's 

bellows, 307 

297. Plan of four-chamber smith's 

bellows, 308 

298. Stone hatchet, 312 

299. Do., 312 

300. Bronze adze, 314 

301. Stone hammer, 315 

302. Do., 315 



303. Stone hammer, . 



304. Bronze hammer, . 



305. Bronze chisel, . . 



306. Do., . 



307. Do., 



308. Do., 



309. Bronze gouge, . 



310. Specimen of dry masonry 



311. Clochan-na-Carraige, . 



312. Round Tower, Devenish, 



313. Christian stone oratory, . 



314. Window of Castledermot Abbe 




315. Doorway of Rahan Church, 



316. Front of Killaloe Church, 



317. Ancient mill-wheel and shaft, 



318. Upper stone of quern, . 



319. Quern 



320. Grain-rubber, . . . 


321. Specimen of weaving, . 



322. Do., do., . 



323. Portion of goat-hair web,. 


324. Plaited woollen band, 



325. Portion of woollen dress, . 



326. Two bronze needles, . 



327. Antique specimen of sewing, 



328. Steelyard, .... 



329. Bractcate coin, . . 



3?,X Do., 

. 382 

331. Small bunne or gold ring, 

. 385 


33s. Ancient chariots, . 

. 408 

Horseman, from Book of Kells, . 413 

Do., do., . . 417 

Single-piece canoe 423 

Skeleton of Irish elk, . . . 459 

Otter trap, 47' 

Bone chessman 479 

The colours of the twelve winds, . 522 

Cromlech at Tawnatruffaun, . . 537 

Sepulchral stone circle, . . . 539 

Do., do 542 

Cinerary urn 5+6 

Do., 54& 

New Grange, 555 

King Dathi's grave and pillar-stone, 557 

Cam near Sligo, .... 563 

Duma or burial-mound, . . . 564 
Sepulchral chamber, with carvings 

and sarcophagus, .... 565 

" Giant's grave," .... 567 
Tomb of the Four Maels at Ard- 

naree 569 

The great cromlech at Kilternan, . 570 
The Phoenix Park cromlech, . . 57* 
Decorated lid of stone coffin, . . 572 
Lugnaed's headstone, . . . 573 
An ulad or altar-tomb, . . . 576 
Strongbow's monument in Christ- 
church, Dublin 57* 

Tomb of Fclim O'Connor, king of 

Connaught 579 


Sculpture on Window, Cathedral Church, Glendalough ; Berruiffer, 1779. 
(From Petrie's Round Towers), 



Section i. Marriage. 

ancient Ireland it was a very general custom, 
as it was in Wales, and in Greece in the time 
of Homer, that when a couple got married 
the man was bound to bring the marriage 
portion or dowry, not the woman. Instances 
of this custom are mentioned everywhere in 
literature : and so well was it recognised, that 
the ancient Irish writers — as was their wont in such 
cases — assign a legendary origin for it. The legend is 
found in the Book of Leinster, into which it was copied 
from the still more ancient Book of Dromsnechta. When 
the sons of Milesius arrived in Ireland, they found there 
some Hebrew women who had been driven thither from 
the Tyrrhene or Mediterranean sea by a tempest. The 
newcomers proposed marriage to them : but the women 
answered that they preferred to return to their own 
country ; and that they would not abandon it to marry 
the Milesians unless they got Tinnscra or dowry as a sort 
of compensation : to which the Milesians agreed. And 
the old account goes on to say : — " It is from this circum- 
" stance that in Erin it is the men that purchase wives 
" always : while it is the husbands that are purchased in 
" all the rest of the world."* 

* O'Curry, MS. Mat., 15, bottom : LL, 190, c, 


There were several terms in common use to designate 
dowry : and according to an ancient manuscript in Trinity 
College, Dublin, quoted by O'Donovan,* the several names 
were used for different sorts of dowries. The Tinnscra 
was a gift of gold, silver, copper, or brass : the Coibche 
[cov-ke] consisted of clothes and warriors : the Slabra of 
cattle and horse-bridles : and the Tochra of sheep and 
swine. But there is good reason to believe that these 
distinctions were not rigidly adhered to, and that the 
several terms were in some measure used indiscriminately. 
Thus O'Clery, in his Glossary, explains Tinnscra by 
Coibche : and many other such instances might be cited. 
Moreover, the dowry might consist of other things besides 
those named above, such as land, or houses, or the con- 
cession of some valuable favour or privilege. " Give me," 
said Oengus mac Natfree, king of Munster in the fifth 
century, " your foster-child [Eithne Uathach] as a wife, 
and I will give you land as Tinnscra."^ There were 
other terms, too, for ' dowry,' such as fola and foluch. 
Occasionally time was given for payment of the dowry : 
if it was paid in hand, it was called by the general name 
Ellam,% from lam, ' the hand.' 

In Ireland, as among all the Aryan nations,§ the 
original conception was that the man purchased his 
affianced wife from the father or other guardian, and the 
dowry he brought in was the bride-price. It was usually 
paid over by the bridegroom to the father of the bride. 
Accordingly, Cormac's Glossary interprets coibche as 
meaning cendach, i.e. ' buying.' The bride-price often 
consisted of a yearly payment from the husband after 
marriage : and we find it laid down in the Brehon Law 
that the woman's father was entitled to the whole of the 

* In Hy F, 207, note r : see also Silva Gad., 525, l8 . 
t LU, 54, b, 2 . and Sullivan, Introd., 174, note 29S . For houses as 
dowry, see O'Curry, MS. Mat., 133. J Corm. Gloss., 67, " Ellam." 

§ De Jubainville, Cours de Litt. Celt., VI. 303. 


first year's coibche, to two thirds of the second year's, to 
one-half of the third : and so on, diminishing to the twenty- 
first, when the claim ceased.* In each case, what was left 
of the coibche belonged to the wife.f 

We meet many instances where the dowry consisted of 
a privilege. When Fergus mac Roy, king of Ulaid (Ulster) 
in the first century of the Christian era, proposed marriage 
to the beautiful widow Ness, she refused to marry him 
except on this condition as Tinnscra : — That her son 
Concobar, then a mere boy, should be permitted to reign 
as king, instead of Fergus, for one year : to which Fergus, 
with the consent of his nobles, agreed. But at the end of 
the year — just as the wily widow expected — when Fergus 
claimed his throne, the nobles refused to supersede 
Concobar, who — by the help of his mother — had so com- 
pletely won them over that he remained king for the rest 
of his life. J 

Within late historical times we find a still more 
interesting example of this sort of bride-price. Early in 
the fourteenth century, it happened on one occasion that 
Cormac O'Clery, a learned young ollave or doctor of laws, 
visited the house of Matthew O'Sgingin, professor of 
history to the O'Donnells of Tirconnell. For many 
generations the O'Sgingins had been hereditary historians 
to the O'Donnells : but this Matthew was destined to be 
the last ollave of the name, for his only surviving child was 
a daughter, a beautiful young girl. The two young people 
soon fell in love with each other : and the father consented 
to their union, but demanded from O'Clery as Tinnscra, 
that the first son born of the marriage should be sent to 
study history, so as to succeed to the position held by the 

* Br. Laws, 11. 347 ; in. 315. 

f In remote times the idea of sale and purchase of the woman in 
marriage must have been prominent and familiar : see an instance of 
Concobar mac Nessa buying a woman after the death of her first husband : 
Br. Laws, iv., p. 9, note. 2 . 

X LL, 106, a, 3o: MS. Mat., 274, 636 : Stokes, Lives of SS., xxxv., top. 


O'Sgingins. The young man willingly agreed : and he 
faithfully kept to his promise. His first son, who became 
a historian, was the ancestor of the O'Clerys of Kilbarron 
in Donegal, who succeeded the O'Sgingins as hereditary 
ollaves of history to the O'Donnells.* They were a race 
of scholars who have left us many precious works in Irish, 
illustrating the history and antiquities of Ireland, including 
the Annals of the Four Masters, the greatest and most 
important of all.f 

The fact that the husband paid the bride-price did not 
prevent the bride bringing goods or valuables of her own, 
if she had them. Any number of cases might be cited 
where the young woman brought jewels, or gold, or herds, 
or land : and after the marriage, these continued to be 
her own special property. Sometimes the friends of the 
young couple made a collection for them, which was called 
Tindl (i.e. ' collection ' : pron. tinnole), of which two- 
thirds belonged by law to the man, and one-third to the 
woman.J This custom was common among high and low, 
and we meet with instances everywhere in the tales. Our 
present custom of making a young married couple presents 
is not unlike the old Irish tinnole. 

It was usual that girls should be married in order of 
age, beginning with the eldest. We are told in the story 
of the Boroma that Tuathal the Legitimate, king of Ire- 
land (a.d. 130-160), had two daughters, of whom Eochaid 
[Ochy], king of Leinster, married the elder, though he pre- 
ferred the younger, who was more beautiful. " For at that 
time " — says the story — " it was not the custom in Erin for 
the younger sister to be married before the elder. "§ The 
lady Emer, when Cuculainn sought her in marriage, says : — 
" I may not marry before Fial my sister, who is older than 

* FM, Vol. 1., Introd. Remarks, xx. For Kilbarron, see vol. I., p. 
524, supra. 

f For examples of other sorts of dowries, see Oss. Soc, iv. 299 (head 
of a destructive wild boar), and Bee Fola, p. 175 (a brooch). 

% Br. Laws, 11. 347, 350, 351. § Rev. Celt., xin. 37 : Silva Gad., 402 


I am." It was expected, too, that a girl should marry a 
man of a family equal to her own in social standing.* 

Marriages, as stated elsewhere (Chap, xxix., p. 439, 
infra), formed a prominent feature of the fair of Tailltenn, 
held during the last days of July and first days of 
August. But it would appear from a passage in Cormac's 
Glossary (p. 82, " Gam ") that throughout Ireland in 
general the favourite and fashionable month for getting 
married was November. According to some authorities 
quoted by Sullivan in his Introduction to O'Curry's 
Lectures (p. 240), a tribute had to be paid — at least in some 
cases — to the king, on the marriage of every maiden of his 
people. This tribute was usually a fdinne maighdena 
[fawnya mydena], or ' maiden's ring ' ; for it was often a 
gold ring : but it might be an ounce of gold, or less, or it 
might be the bride's wedding-dress. 

The general custom was to have only one wifef : but 
there were exceptions, for in very early times we some- 
times find a king or chief with two. In the story of the 
Cattle Spoil of Fraech in the Book of the Dun Cow, 
referring to far distant pagan ages, it is related how Fraech, 
a powerful chief of Erris, goes to the house of Ailill and 
Maive, king and queen of Connaught, as a suitor for the 
hand of their daughter Finnabair, though it was well 
known that he had at the time a wife and three sons. 
Cobthach Coel Breg, king of Ireland before the Christian 
era, had more than one queen. % Coming to Christian 
times, Dermot, king of Ireland, a.d. 544-565, had two 
queens.§ In the time of St. Finnchua (seventh century) 
old Nuada the sage, king of Leinster, had two wives {da 
bainchele), who, as might be expected, kept the poor old 
king in hot water by their jealousies and bickerings.|| 

* Br. Laws, II. 347. 

f On Monogamy, the basis of Greek, Roman, and Celtic society, see 
De Jubainville, Cours de Litt. Celt., vi. chap. iv. 
X Orgain Dind Rfg, in Zeitschr. Celt. Phil., in. 9. 
§ See page 255, farther on. || Stokes, Lives of SS., p. 237. 


That chastity and modesty were prized we know from 
many passages, such as that in the Life of St. Finnchua, 
in which he leaves blessings to the Leinster men, among 
them " chastity in their queens and in their wives, and 
modesty in their maidens."* A wedding was called banais 
contracted (from ban-fheis, meaning " woman's feast ") : 
a married couple was lanamain : marriage, lanamnas : a 
widow was fedb and banlrebihach. 

2. Position of Women and Children. 

In ancient Ireland free women (as distinguished from 
slaves) held a good position : and it may be said that as to 
social rights and property they were in most respects quite 
on a level with men. Husband and wife continued to own 
the respective shares they brought in at marriage, such as 
land, flocks, household goods, &c, the man retaining his 
part and the woman hers, each quite independently of the 
other. Of this custom we find illustrations everywhere ; 
and there are many records of married women taking legal 
proceedings on their own account against outsiders, quite 
independently of the husband, in defence of their special 

But notwithstanding this separate ownership, as both 
portions were worked more or less in conjunction, and 
naturally increased from year to year, it was generally 
impossible — even if so desired — to keep them distinct, so 
that a part at least of the entire possessions might be 
looked upon as joint property : and for this state of things 
the law provided. It is from the Brehon Law we get the 
clearest exposition of the rights of women regarding 
property. The respective privileges of the couple after 
marriage depended very much on the amount of property 

* Stokes, Lives of SS., p. 239. 

f For the law on the point, and instances, see Br. Laws, 11. 361, 363, 
379 : O'Curry, Man. & Cust., n. 89 : Stokes, Lives of SS., 165, 235 : 
Feilire, 75 : Reeves, Adamn., 305 : and vol. I., p. 216, supra (the glaisin 


they brought in. If their properties were equal at marriage 
" the wife " — says the Senchus Mor — " is called the wife of 
equal rank," and she was recognised as in all respects, in 
regard to property, on an equality with her husband.* In 
this case all transactions affecting the joint property, such 
as buying and selling, had to be made with the consent of 
both parties : and any contract made by either, for his or 
her own special benefit, without the consent of the other, 
was null and void in a court of law. But if it could be 
shown that the transaction tended equally to the advantage 
of both, the law confirmed it. If profit accrued from any 
transaction (such as selling) it was apportioned to husband 
and wife in the ratio of their respective shares. 

That the husband and wife were on terms of equality 
as to property is made still more clear from the provisions 
laid down to meet the case of separation : and from the 
evident care with which these are set forth, we may con- 
clude that the separation of married couples was, in those 
days, by no means an uncommon occurrence. Sometimes 
they separated by mutual consent and sometimes as the 
outcome of legal proceedings. Seven different kinds of 
injury are enumerated in the law, which if inflicted on a 
wife by her husband, gave her the right to separate 
from him : most of which would at this day lead to a 
decree of " separation from bed and board." If she proved 
her case home, she was entitled to her dowry (or 
that part of it that remained with her after marriage) 
in addition to personal damages as Eneclann or honour- 
fine, f 

If the couple separated by mutual consent, the woman 
took away with her all she had brought on the marriage 
day ; while the man retained what he had contributed. 
Supposing the joint property had gone on increasing 

* Br. Laws, u. 357 : and Preface, lvi. 

t Br. Laws, v. 293 : see also II. 357, 359, 361, 3 8l > 3 8 3 1 and 
Sullivan, Introd., 176. 


during married life : then at separation the couple divided 
the whole in proportion to the original contributions.* 

But all this might be modified by special circumstances, 
which are detailed in the law with much exactness. One 
of these was whether the woman was a " great worker " or 
a "small worker." She was a " great worker " if she had, 
of her own, all the utensils necessary for her feminine 
occupations : such as a mill, a sieve, a loom, a spinning 
wheel, a distaff, spindles, &c. : she was a " small worker " 
if these were supplied by the husband, f In the division 
of property a great worker took a larger share than a small 
worker. For instance, in case of flax at the time of pulling 
and drying : a great worker got one-sixth of it ; a small 
worker one-ninth. J 

If the woman was a " great worker " during married 
life, and helped by her industry to increase the property, 
she was entitled, at separation, to one-ninth of the increase, 
even though she had no property (beyond the utensils) at 
the time of marriage. § And in like manner the man, under 
similar circumstances, if he had been a "great worker," could 
claim just one-ninth of the increase. Here, in the words of 
the law tract (n. 391), " the man goes in the place of the 
woman, and the woman in the place of the man." 

As to household articles manufactured by the woman's 
hand, she was entitled to a part, as in case of other goods 
— even though having no propert}' at the time of marriage : 
but the exact amount depended on the state of advance- 
ment of the work. Take woollen goods as an example, 
which were generally managed exclusively by women. If 
the wool was in the fleece at the time of separation, she 
took one-eighth. If it had been separated into locks or 
flakes ready for combing, she took one-sixth ; if after being 
combed ready for spinning, one-third : after spinning, or 

* Br. Laws, n. 397 : O'Curry, Man. & Cust., rl. 118. 
I Br. Laws, 11. 411. J Br. Laws, 11. 419. 

§ Br. Laws, 11. 391, 393, 395. 


when in cloth, one-half.* So also flax : one ninth-growing 
" on foot " ; one-sixth after drying ; one-third after scutch- 
ing ; one-half after that. Similar arrangements were laid 
down for dye-stuffs in the several stages of preparation ; 
and for various other materials and products. From all 
this, moreover, as well as from many separate passages, 
it may be inferred that great importance was attached 
to hand-work of all kinds. 

Women, as has been said, might take actions at law on 
their own account, and if successful might distrain the goods 
of the defendant. But the distress should be confined to 
such things as were understood to specially pertain to 
women, in their daily life, of which a list is given in the 
Senchus Mor. These included sheep, lap-dogs, and cats ; all 
utensils used in the manufacture of cloth, such as spindles, 
wool-bags, needles, weavers' reeds, &c, and also such 
articles as looking-glasses, sieves, and kneading-troughs.f 

When the wife owned land, it was subject to the same 
law of succession as that of the husband ; viz., if she had 
sons it descended to them, whether there were daughters 
or not : if she had no sons it went to the daughters. Here 
however, there were proper safeguards to prevent the land 
passing from the tribe, in case the woman married a man of 
another tribe. J When land passed by any right to a woman, 
she entered into formal possession, by a process which was 
attended with very curious ceremonies. § When a woman 
came into possession of land, she was bound to send men 
for war-service like male land-owners : but she was set free 
of this obligation on giving up to the fine or tribe half the 
land ; a privilege which was not accorded to men.|| 

* Br. Laws, II. 369, 373 : O'Curry, Man. & Cust., II. 118. 

f General list in Senchus Mor, Br. Laws, 1. 151 ; and see 149. See 
also iv. 9, note 2 . and 13, I9 . 

\ See Br. Laws, iv. 17 and 19, 39 and 41, 45, 47, top : and Introduction, 
cxvi, cxvii. § Ceremonies : for which see Br. Laws, iv. 9, 11, 13. 

|| Br. Laws, iv. 19, 8 to end of par. ; 21, note 2 ; 41, mid par. ; 45; 
47, n ; 49 ; an( i Introd., cxvii., bottom 


Husband and wife stood on equal terms in a brehon's 
court, so that if the husband gave evidence against his 
wife, she was entitled to give evidence against him. For, 
as the Senchus M6r expresses it, " though the Law cedes 
" headship to the man [husband] for his manhood and 
" nobility, he has not the greater power of proof upon the 
" woman on account of it, for it is only contract that is 
" between them."* But her father could give evidence 
against his daughter, whether married or single, and she 
was not permitted to rebut it by her evidence, f 

The testimonies hitherto brought forward are mostly 
legal and historical. But the general popular conception 
of the position of married women may be also gathered 
from the old romantic tales and legends, including those of 
the Dinnsenchus, in which women hold as high a place as 
men. We read of great female physicians, such as Air- 
meda the daughter of the leech-god Dianket ; and of 
distinguished female brehons or lawyers, such as Brigh 
Briugaid, whose decisions were followed as precedents for 
centuries after her death. 

But with all that has been said so far in commendation 
of the position of women, there were some features, which, 
regarded from a moral point of view, were very objection- 
able. These were not indeed peculiar to Ireland, but were 
common enough among European nations at the time ; but 
they were not the less repulsive for that. It is manifest 
from Irish literature in general, and especially from the 
Brehon Laws, that the practice of separation of man and 
wife, either by mutual consent or by process of law, was 
unpleasantly common. Concubinage was very general, 
especially among the higher classes, and does not appear 
to have been regarded by the general public as in any 
degree reprehensible : indeed the Brehon Law provides 
for it as a recognised custom. Female slaves too were 
treated with great grossness, at least till the time of 

* Br. Laws, n. 351. f Ibid., 347. 


Adamnan (at the end of the seventh century), by whose 
exertions some much-needed improvements were effected 
in their position, as well as in that of women in general.* 

It is worthy of remark that the good old lawyers who 
compiled the Brehon Law tracts, while doing full justice 
to the position of women, are sometimes given to a rather 
pompous assertion of the superiority of their own sex. 
We have seen the lofty expression in last page : and the 
commentator, explaining why the Senchus M6r, though 
treating as much of women as of men, is still called the 
" Senchus of the Men of Erin," says "it is proper indeed 
" that it should be so called, so as to give superiority to 
" the noble sex, i.e. to the male : for the man is the 
" head of the woman, and the man is more noble than the 
" woman. "f 

The son was under the father's control till formally 
emancipated : but what was the age or in what the 
ceremony of emancipation consisted I have not found. % 

We have seen (vol. I., p. 165, supra) that, even as late 
as the twelfth century, it was common among the English 
to sell their children and other relatives, especially to the 
people of Ireland. In Ireland illegitimate children were 
sometimes sold, but not legitimate children. It is laid 
down in the Brehon Law that the children begotten 
illegitimately of a woman who has been abducted belong 
to the woman's family, who may sell them if they choose. § 
That fathers also sometimes sold their illegitimate children 
is shown by a story in the Life of St. Brigit, who — accord- 
ing to this legend — was the illegitimate daughter of a 
pagan chief named Dubthach. When she was a girl living 
in her father's house, she was so charitable that every- 
thing in the house she could lay hands on she gave away 

* Trip. Life, Introd., xxii. f Br. Laws, 1. 35. 

Br. Laws, iv. 231, 12; V. 357, „. 439, 5. See De Jubainville, Cours 
de Litt. Celt., vi. 312 : and " Saer, Leicthe " in Atkinson's Br. Law Glos 
As to the duty of the son to support the parents, see p. 495, infra. 

§ Br. Laws, m. 403, 541. 


to the poor and the needy : till her father became at last 
so incensed that he resolved to rid himself of her by selling 
her for a slave. He actually brought her away in his 
chariot with this object but was diverted from his purpose 
by the advice of a friend : and so Brigit was saved from 
bondage.* In the annals and other ancient writings we 
sometimes come across references to times of famine and 
distress so severe that people were driven to sell their 
legitimate children to procure food. But these entries, so 
far from showing that the practice was customary in 
Ireland, prove, as acknowledged exceptions, the very 

Adoption, whether of individuals, of familes, or of 
whole septs, has been already dealt with (vol. i., p. 166, 

A child was called in Irish lenab, now leanbh, pron. 
lannav : an infant was noidiu. A son was mac, corre- 
sponding with the Welsh map : a daughter, ingen [ing-een] : 
a grandson was ua, hua, or haue. A father was athair : a 
mother, mdthair : a brother, dearbhr athair, now pron. 
drahaar : a sister, deirbhshiur, pron. drihoor. 

3. Fosterage. 

One of the leading features of Irish social life was 
fosterage (Irish, altram or altrum), which prevailed from 
the remotest period. It was practised by persons of all 
classes, but more especially by those in the higher ranks. 
The most usual type of fosterage was this : — A man sent 
his child to be reared and educated in the home and with 
the family of another member of the tribe, who then became 
foster-father, and his children the foster-brothers and foster- 
sisters of the child. While young persons were generally 
fostered in this manner, in families, some were put in care 
of distinguished ecclesiastics : and many of the Irish saints 

* Stokes, Lives of SS., 187 


were fostered in this way, whose early training in a great 
measure determined their future life. St. Columkille was 
fostered and educated during childhood by a holy priest 
named Cruithnecan.* For a number of individual ex- 
amples of the fosterage of well-known historical personages, 
the reader may refer to O'Curry's Manners and Customs of 
the Ancient Irish, I. 374, 375. 

The foster-father was denoted by the word aite or 
oide [2-syll.] : the foster-mother by muime [mumme] : the 
foster-child by dalta. Dalta is still in use as a term of 
endearment to denote a favourite or a petted child, but 
is now always applied to a boy. A foster-brother was 
comalte [3-syll.]. Fosterage was subject to stringent 
regulations, which were carefully set forth in the law. A 
special portion of the Senchus Mor — occupying twenty- 
four pages, Irish type, of the second volume — is devoted 
to it ; in which the rights, duties, and obligations of the 
parties are detailed with minute particularity : and it is 
referred to in other parts of the law. I give here a few of 
the most important of these regulations. 

A child might be sent to fosterage at one year of age. 
Boys might be kept till seventeen and girls till fourteen, 
which were considered the marriageable ages : then they 
returned to their parents' house. There were two kinds of 
fosterage — for affection and for payment. In the first there 
was no fee : in the second the fee varied according to rank. 
The fosterage fee (iarrad) sometimes consisted of land, 
but more generally of cattle. For the son of an og-aire or 
lowest order of chief, the fee was three cows ; and from 
that upwards to the son of a king, for whom the fee was 
from eighteen to thirty cows. For girls, as giving more 
trouble, requiring more care, and as being less able to help 
the foster-parents in after-life, it was something higher 
than for boys. The child, during fosterage, was treated in 
all respects like the children of the house : he worked at 

* Reeves, Adamnan, 191. 


some appropriate employment or discharged some suitable 
function for the benefit of the foster-father : and he had to 
be educated in a way that suited his station of life : as has 
been already described (vol. I., p. 441). There were minute 
regulations regarding clothes, food, and means of amuse- 
ment, all of which varied according to rank. How far the 
foster-father was liable for injuries suffered by the foster- 
child at the hands of others, or for his misdeeds, is set 
forth with great care. 

Precautions were taken, in the shape of penalties, to 
prevent the fosterage being terminated before the time by 
either party without sufficient cause. At the termination 
of the period of fosterage the foster-father gave the foster- 
son a parting gift, the amount of which was regulated 
according to rank and other circumstances. If in after-life 
the foster-father fell into poverty, and had no children of 
his own to support him, he had a claim on his foster-son 
for maintenance, provided he had duly discharged all the 
duties of fosterage, including that of the parting gift. The 
foster-mother had a similar claim. It was usual for a chief 
to send his child to be fostered to one of his own sub- 
chiefs : but the parents often chose a chief of their own 
rank. Sometimes a chief had a large number of children 
at fosterage : in the Book of the Dun Cow we are told 
that at one time Ochy Beg, king of Cliach, the district 
round Knockainy in Limerick, had forty boys in his charge, 
sons of the nobles of Munster.* In cases where children 
were left without parents or guardians, and required pro- 
tection, the law required that they should be placed in 
fosterage under suitable persons at the tribe's expense, f 

The children of kings, chiefs, or other distinguished 
persons were eagerly sought after for fosterage ; and in 
order to satisfy such claims, it sometimes happened that 
such children had two or more foster-fathers, with whom 

* O'Curry, Man. & Gust., I. 357. f Br. Laws. 11., Pref. lvii. 


they lived in succession or in turn. Thus the great 
Dedanann chief, Lugh the Ildana, had nine fosterers ;* and 
coming to historic times, Lewy Mac Con, king of Ireland 
(a.d. 250-253), was fostered by Olioll Olum, king of 
Munster, and by Olioll's brother, Lewy Laga.f To be 
fostered by several was considered a mark of distinction. 
Laegaire's daughters, when inquiring from St. Patrick about 
God (who they thought might be some great chief), asked, 
among many other questions, did many foster His Son — 
implying that this would be a sure indication of rank and 

Although it ( was more usual to send boys to be fostered 
than girls, still we often find in the old tales pleasant 
pictures of girls in their homes with their foster-sisters. 
In the story of the courtship of Emer, the young hero 
Cuculainn, going to the house of her father, Forgall Monach, 
at Lusk, north of Dublin, to woo the young lady, found 
her on the lawn of the fort with her foster-sisters, who 
were learning embroidery and hand-dexterity with her.§ 

Fosterage was the closest of all ties between families. 
The relationship was regarded as something sacred. The 
foster-children were often more attached to the foster- 
parents and foster-brothers than to the members of their 
own family : and cases have occurred where a man has 
voluntarily laid down his life to save the life of his 
foster-father or foster-brother. This attachment is noticed 
by many English writers from Giraldus down|| : and 
illustrations are found everywhere in Irish writings both 
ancient and modern. At the great Battle of Moyrath, 
fought in 637 by the monarch Domnall against his 
rebellious foster-son, Congal Claen, the king, both before 
and during the battle, shows himself most anxious for the 
personal safety of Congal, now his mortal enemy, ^f For a 

* Rev. Celt., xn. 89. % Todd, St. Patk., 453 : see also Trip. Life, clxix. 
f Silva Gad., 349. § Kilk. Archseol. Journ., 1870-1, p. 404. 

|| Girald., Top. Hib., in. xxiii. v Tf Moyrath, 135, 155, 161, 305. 



modern illustration, see Carleton's story of " The Foster- 
Brother."* The custom of fosterage existed in Ireland — 
though in a modified form — even so late as the seventeenth 
or eighteenth century. 

There was also a literary fosterage, when a boy was sent 
to be reared up by an ollave or professor, and instructed 
for a degree. The foster-father was " to instruct him with- 
" out reserve, to prepare him for his degree, to chastise 
" him without severity, and to feed and clothe him while 
" learning his legitimate profession." The amount of fee 
was regulated by law. All gains earned by the pupil 
while learning were to be paid to the tutor, and also the 
first fee he earned after leaving him. If the teacher fell 
into poverty in after-life, his foster-pupil was bound to 
support him. The relationship of literary fosterage was 
regarded as still more close and sacred than that of 
ordinary fosterage. On this see also vol. I., p. 423. 

Gossipred. — When a man stood sponsor for a child at 
baptism, he became the child's godfather, and gossip to 
the parents. Gossipred was — and is still — regarded as a 
sort of religious relationship between families, and created 
mutual obligations of regard and friendship, f 

After the Anglo-Norman invasion the people of the 
English colony, from the great lords down, often sent their 
children to be fostered by the Irish : and as might be 
expected, these young persons grew up speaking the Irish 
language, and thoroughly Irish in every way. Mainly for 
this reason the two customs of fosterage and gossipred 
were bitterly denounced by early English writers, J most 
of whom were anxious to keep the two races apart : and 
we know that the Government passed several stringent 
laws forbidding them under the penalty of high treason : 
but these laws were generally disregarded. Gossipred in 

* Irish Penny Journal, 338. 

t For full accounts of Fosterage, see Br. Laws, II. 147, 349 ; v. 97 : and 
for both customs, Ware, Antiqq., chap. xi. X Instance, Spenser, 112. 




a modified form exists to this day all over the empire ; 
and the custom of fostering was formerly common among 
the Welsh, the Anglo-Saxons, and the Scandinavians. 

4. Family-Names. 
Hereditary family-names became general in Ireland 
about the time of Brian Boru, viz. at the end of the tenth 
and the beginning of the eleventh century : and some 
authorities assert that they were adopted in obedience to 
an ordinance of that monarch. The manner of forming 
the names was very simple. Each person had one proper 
name of his own. In addition to this, all the members of 
a family, and of their descendants in the male line, took 
as a common surname the name of their father, with Mac 
(son) prefixed, or of their grandfather or some more remote 
ancestor, with Ua or (grandson or descendant) prefixed. 
Thus the O'Neills are so called from their ancestor Niall 
Glunduff, king of Ireland (a.d. 916), and ' John O'Neill ' 
means John the descendant of Niall : the Mac Carthys of 
Desmond have their surname from a chief named Carrthach, 
who lived about the year 1043. The same custom was 
adopted in Scotland : but while in Ireland was much 
more general than Mac, in Scotland the was very rarely 
chosen, and nearly all the Scotch Gaelic family-names 
begin with Mac. 

Sculpture on a Column, Church of the Monastery, Glendalough. 
(From Petrie's Round Towers, 260.) 

Ornament : composed from the Book of Kells. 



Section i. Construction, Shape, and She. 

efore the introduction of Christianity, buildings 
in Ireland, whether domestic, military, or 
sepulchral, were generally round or oval. The 
quadrangular shape, which was used in the 
churches in the time of St. Patrick, came very 
slowly into use, and round structures finally 
disappeared only in the fourteenth or fifteenth 
century. But the round shape was not universal, even in 
the most ancient period. The great Banqueting Hall of 
Tara was rectangular, as we see by its ruins at the present 
day. The Craebh-ruadh [Creeveroe], a similar hall at 
Emain, was of a like shape ; and the bruden or feasting- 
hall at Dun-da-benn (the fort overhanging the waterfall in 
the Bann near Coleraine) was square : both of these made 
in imitation of the hall of Tara.* And in case of many of 
the ordinary good-sized dwelling-houses, the expressions 
used regarding them show that the walls were straight and 
parallel, and that consequently the shape was rectangular. 
Some of the old lisses or forts still to be seen are of this 
shape : and even where the surrounding rampart was round 
the wooden houses it enclosed were often rectangular, f 

* Tain bo Fraich, 160, 161 : Mesca Ulad, 13. 

t See for example Sick Bed, Atlantis, 11. 105, first line : also O'Curry, 

Man. & Cust., 11. 31. 



The common Irish word for a house is tech, Lat. 
tectum ; whence come the compounds tegduis or teaghdais, 
one of the names for a homestead ; and teaghlach, ' a family 
or household.' A dwelling in general is denoted by drus 
or dross ; a homestead by baile, now generally anglicised 
bally, but used in a more extended sense to denote a town- 
land. The word brug or brugh [broo] was also applied 
to a large dwelling : in Peter O'Connell's Dictionary, taken 
from old authorities, we find : — " Brug, the same as baile, 
a mansion, manor, or farmhouse." But this word brug 
had other applications, which will be found fully set forth 
in Hennessy's Introduction to the Mesca Ulad, p. 7. 

It has sometimes been stated that there were no towns 
or cities in ancient Ireland : but this statement is mis- 
leading. There were many centres of population, though 
they were never surrounded by walls ; and the dwellings 
were detached and scattered a good deal — not closely 
packed as in modern towns. In our old writings, both 
native and Anglo-Irish, we have many records of towns 
and cities. As a comparatively late example — in the 
twelfth century — may be mentioned Downpatrick, which 
Giraldus Cambrensis, in his account of John de Courcy's 
invasion of Ulster, calls the " City of Down " : but it was 
quite open and undefended. Then we know that some of 
the large monasteries had two or three thousand students, 
which implies a total population much larger. Some of 
the provisions of the Brehon Law show that numbers 
of lis-dwellings must have been clustered together : one 
statement, for instance, that a mill or a fishing-weir was 
sometimes the common property of the fine or tribe. 

The dwelling-houses, as well indeed as the early 
churches, were nearly always of wood, as that material 
was much the most easily procured. The ordinary kinds of 
timber were used according to circumstances, but the most 
common were deal, oak, and yew. The custom of building 
in wood was so general in Ireland that it was considered 

•- &»** 
■S , o 

■= c j= 

— O o 

* 5 i 
£ S « 
2 b . 


" o 

^ 3 - 

■^ A) '.£ 

K 1 * S 

— M < 

" * J 

'A £ 9 

B * 

° a ■ 

c « c 


^ "y C 



1 .9 I 

E S * 

j= ? •" 

* U e 
B W B 

1-1 Si c 

.= = a 


S 8. 


a characteristic of the Irish — more Scottorum, " after the 
manner of the Scots " — as Bede expresses it. Yet we 
know that the Britons, Saxons, and Franks, also very 
generally built in wood. When Henry II. was in Ireland, 
1171-2, " he had a royal palace constructed for himself of 
" planed wood, built with wonderful taste, in which he and 
" the kings and princes of Ireland kept the festival of 
" Christmas."* Of course this house was the work of 
Dublin builders and tradesmen. Some of the houses in 
Waterford in 1168 were of wood : and it was by pulling 
one of them down that Raymond le Gros effected an 
entrance into the city. Wooden houses, highly orna- 
mented, continued in use in Dublin, Drogheda, and other 
towns, down to the last century, f 

But although wood-building was general in Ireland 
before the twelfth century, it was not universal : for some 
stone churches were erected from the time of the intro- 
duction of Christianity : beehive-shaped houses, as well as 
cahers and cashels (pp. 57, 58, below), were built of stone, 
without mortar, from pre-historic times : and the remains 
of these primitive structures — churches, houses, and cahers 
— are still to be seen in many parts of Ireland. J In all 
these mortarless buildings, the stones, though in their 
natural state — not hammered or chiselled into shape — are 
fitted to each other with great skill and accuracy : or, as 
Petrie expresses it, " with wonderful art."§ 

The dwelling-houses were almost always constructed 
of wickerwork : tech-figthi or tech-fithi, a ' wickerwork 
house ' : from jigim, ' I weave.' The wall (fraig) was 
formed of long stout poles (slat, ' a pole ') placed in a 
circle, if the house was to be round, standing pretty near 
each other, with their ends fixed deep in the ground, the 

* Hoveden, quoted in Cambr. Ev., 11. 173. 
f Dublin Penny Journal, 1. 89 and 268. 

J For a whole town of pre-historic circular stone houses in Kerry, see 
Macalister's article in Trans. Roy, Ir. Acad., vol. xxxi., p. 209. 
§ See Stokes's Life of Petrie, p. 135. 


spaces between closed in with rods and twigs neatly and 
firmly interwoven ; generally of hazel. The poles were 
peeled and polished smooth. The whole surface of the 
wickerwork was plastered on the outside, and made 
brilliantly white with lime, or occasionally striped in 
various colours ; leaving the white poles exposed to 
view. The residence of O* Murphy at Dunflin in Sligo 
in the thirteenth century is called in HyF (265) " a white 
wattled edifice of noble polish."* When the house was 
to be rectangular the poles were set in two parallel rows, 
filled in with wickerwork. 

Building in wicker-work was common to the Celtic 
people of Ireland, Scotland, and Britain, f It is ver} 
often referred to in Irish writings of all kinds. An 
instance has been already cited in vol. I., p. 10. Adamnan 
(p. 106) relates that Columba on one occasion sent his 
monks to bring bundles of rods (virgarum fasciculos) to 
construct one of the houses of a hospice. The same 
incident is told in the Irish Life of St. Columba ; and 
here the rods or wattles are called by the Irish term 
caelaig (sing, caelach, ' anything slender,' ' a slender wattle,' 
from cael, ' slender,' pron. kail). The cliath [clee-a] or 
hurdle and the wattles or laths for building houses are 
often mentioned in the Brehon Laws. J In some large 
houses the standing poles were very thick and high : in 
describing the construction of Bricriu's house at Dun- 
Rury, the writer of the " Feast of Bricriu " (p. 5) states, 
probably with some exaggeration, that it took seven 
strong men to put each pole in its place. But more 
usually they were of moderate dimensions. 

From the curious details given in the Brehon Law 
tract called Crith Gabhlach,§ it appears that, after the 

* On all this, see O'Curry, Man. & Cust., II. 32 : HyF, 265, 279 : 
Three Irish Homilies, 77 bot. : and vol. I., p. 10, sup*a 

f Bede, Eccl. Hist., III. x. : Rhys, The Welsh People, 199, 200 : Ware, 
Antiqq., xxv. 

J As in vol. iv. 253, 305, 313. Br. Laws, iv. 305. 




poles had been fixed in the ground, the spaces were 
filled up with wickerwork in the following manner, to 
form the fraig, or side wall. Beginning at the bottom, 
a strip of a certain width was woven all round ; another 
strip was woven above that : and so on till the eave was 
reached ; after which a sloping -drip board was fixed all 
round at the junction of each adjacent pair of strips, and 
one at the eave over all. 

FIG. i6cfc 
Maynooth Castle at present : photograph. Erected originally in 1176 by Maurice 
Fitzgerald : but greatly altered and enlarged subsequently. One of the Anglo-Norman 
castles referred to in page 65 farther on. (From Journal of the Kildare Archaeological 
Society, I., p. 223.) 

This last description, and that of Bricriu's house- 
poles above, go to show that the side wall was often 
very high : and this is borne out for other buildings by 
many passages, both direct and incidental. Keating 
(page 333), drawing from old authorities, says that the 
Banqueting-Hall of Tara was 300 feet long by 75 feet 
wide and 45 feet high : and Kineth O'Hartigan, in the 
tenth century, makes a similar statement as to length and 
height.* Now Keating understates both the length and 
breadth, as appears by actual measurement of the present 

* Petrie's Tara, 190. 


existing iiiin,* so we may take it that the height was not 
under 45 feet. Again, three great heroes contend in 
another banqueting-hall, the feat consisting in throwing 
a heavy roth or wheel-quoit upwards towards the roof. 
Laegaire the victorious throws it half-way up the wall : 
Conall Cernach throws it to the ridge pole : but Cuculainn 
sent it right through the roof.f Lastly, in the Battle of 
Rossnaree (p. 5), Concobar, speaking of the devastation 
committed by the Connaughtmen in Ulster, says : — " Our 
" fine dwellings were burned so that they were left no 
" higher than single rooms or outhouses." This passage 
seems to point to two-storied houses, with which other 
passages concur. But in some cases the wall, or part of 
it, was so low that the eave was within reach of the hand, 
like the eaves of some " Swiss cottages " of the present 
day : for the Crith GabhlachJ lays down a penalty for 
taking away any portion of the straw from the thatch of a 
bo-aire's house. When there was more than one apartment 
in a house, each had a separate wall and roof : except, of 
course, where one apartment was over another. 

In the Highlands of Scotland wattled or wicker houses 
were used, even among high-class people, down to the 
end of the eighteenth century§ ; and it is probable that 
they continued in use in Ireland to as late a period. 

In the superior classes of houses, and in churches, a 
better plan of building was adopted, by forming the wall 
with sawed planks instead of wickerwork. The little hut 
erected at Iona for St. Columba's special use was con- 
structed of wood planks (in tuguriolo tabulis suffulto).\\ 
The oratory built at Rahan in the present King's County 
in the year 747 was of boards ; and we are told that it was 
unusually large, so that it took a thousand boards to build 

* For actual dimensions, see p. 85, below. 

t Kilk. Arch. Journ., 1870, p. 438. 

\ Brehon Laws, iv. 313. 

§ Stuart, in Book of Deer, Pref. cli. note I., 

|| Adamnan, p. 54 ; and see 177. 


it.* But the dwelling-houses of monasteries, as well as 
the smaller oratories, were generally of wickerwork : of 
which one instance has already been cited at p. 24 from 
Adamnan. In the houses of the higher classes the door- 
posts and other special parts of the dwelling and furniture 
were often made of yew, carved, and ornamented with 
gold, silver, bronze, and gems. We know this from the 
old records ; and still more convincing evidence is afforded 
by the Brehon Law (iv. 313, 315), which prescribes fines 
for scratching or otherwise disfiguring the posts or lintels 
of doors, the heads or posts of beds, or the ornamental 
parts of other furniture. 

Small square timber houses, consisting generally of just 
one apartment, have often been found deep in bogs, and 
sometimes in clay. They consist of beams and planks of 
oak and other timber, joined together with much rude 
skill by tenon-and-mortise without nails. They seem to 
have served some temporary purpose, as they are too 
small for permanent residence ; perhaps they were in- 
tended to shelter workmen for the time who were cutting 
turf, or for those who attended to cattle when they were 
grazed in the booleys. Or possibly some may have been 
the little buildings connected with ancient mills (see 
chap. xxv.). From the position in which some of these 
houses have been found, it seems plain that they are 
very archaic : belonging probably to a time beyond the 
reach of history, f 

The roof of the circular house was of a conical shape, 
brought to a point, with an opening in the centre for the 
smoke. It was of wickerwork or hurdles supported by 
rafters sloping upwards from the tops of the wall-poles 
all round, to the centre at the very top. From its shape 
and material this sort of roof was often called cua-chlethe, 

* O'Curry, Man. & Cust., 11. 37. 

t Instances : Kilk. Arch. Journ., 1879-82, pp 307, 561; see also 
Wood-Martin, Pagan Ireland, p 223 et seq 


' cup-shaped wicker roof.'* The roof of the quadrangular 
houses was much like that of the common run of houses 
of the present day. If the house was large, the conical 
roof of those of circular form was supported by a tall, 
strong tuireadh or pole standing on the centre of the floorf ; 
in case the house was quandrangular, there was a row of 
such supporting poles, or two rows if the structure was 
very large. The circular building with conical roof was 
in shape exactly like the buildings called ' tholos ' among 
the Greeks. 

Straw was used for roof-covering from the earliest 
times, and its use has continued to the present day. In 
1596 the straw thatch of O'Madden's Castle was ignited 
by a firebrand thrown by the besiegers, so that the roof 
was burned (Hy Many, 150). The word tuga was used 
to denote a roof -covering, whether made of straw or of 
any other material. In the Tripartite Life of St. Patrick 
(p. 157) we are told that a certain person built a house 
in Ess Mace n-Eirc ; but that a rush of the thatch 
(simni tuga : simni, ' a rush ') had not been put upon 
it before it was demolished by another person. Reeds 
were often employed ; and for this purpose they were 
sometimes cultivated in special plots of ground. J We 
have seen that St. Finan roofed his church at Lindis- 
farne with reeds (har undine texit). Whatever the material, 
the covering was in all cases put on with some degree of 
art and neatness, such as we see in the work of the skilled 
straw-thatchers of the present day. 

A better class of roof than any of the preceding was 
what is called in Irish slinn, commonly rendered by 
shingle. The house of Ailill and Maive at Cruachan 
had a roof of slinn (tuga slinned).§ Slinn, in Old Irish, 

* Sullivan, Introd. 299, note 531 : and LU, 19, a, 17. 
f O'Curry, Man. & Cust., II. 32. % Adamnan, 163. 

§ Tain bo Fr., 141 : Ir. Texte, 1. 281, „ . and see " Slind " in Wirter- 
huch, same vol. : also Fled Bricrenn, § 55. 


glosses imbrex, ' a brick or tile,' and it has the same 
meaning in the modern language : but Uiga dinned, 
generally means a roofing of thin boards. For instance, 
in an Irish poem written by Mac Conmidhe early in the 
thirteenth century, the church of Armagh is said to be 
roofed with slinns of oak (slinnlech darach)* ; and in 
another much older authority shingles made of yew are 
mentioned. f The covering was constructed by making 
the small flat pieces overlap as in modern slated or 
tiled roofs. Sometimes, anticipating modern usage, they 
employed materials superior to any of the preceding. 
The Annals of Ulster record that in the year 1008, the 
oratory of Armagh was roofed with lead. 

The thatch of ladies' greenans (see p. 42, infra) was 
sometimes formed of birds' wings. In a poem composed 
by the hero Cael O'Nemnann about the lady Crede [Cray], 
daughter of the king of Kerry, it is stated that her beautiful 
greenan was thatched with the wings of birds, so arranged 
as to form bright stripes of brown, reddish purple, and other 
colours^ : and King Cormac Mac Art, when he visited 
Tairngire or Fairyland, saw people thatching a house with 
the wings of white birds.§ 

Over the top of the principal room on the inside 
extended a ridge-pole or roof-tree called feici [faika], 
from which lamps were suspended to light the apartment : 
whence Cormac's Glossary (p. 81) derives the word from 
feighe r fay], ' illumination.' This last word again is derived 
from feig, which Zeuss (998, 3 8, 39) explains ' bright, or 
illuminated,' and which is connected with the modern 
Gaelic verb, fetich, ' see.' This word feici was also applied 
to the ridge-pole of a tent, and sometimes to the lintel 
of a door. || The feici was used in houses of a rectangular 
shape, and possibly in round houses also. It was supported 

. *0'Curry, Man. & Cust., H. 5*. t Ibid., p. 34. 

X O'Curry, MS. Mat., 309 : Maw. & Cust , 11. 12, 13 : Silva Gad., 120. 
8 Ir. Texte, in. 213. || Moyrath, 200, I3 . 


by the posts already mentioned, which also supported the 
roof. In Cormac's Glossary (34, under ' clii ') a house-post 
is mentioned as tapering from floor to ridge. 

There were windows in the fraig or wall, and often a 
skylight in the roof. A window was called by two names 
semester and fmnneog, the first derived from Lat. fenestra, 
and the second — which is the word now in use — from 
A. -Sax. windeage* A skylight was called by the native 
name forless (' top-light '), from less or les, ' light.' On one 
occasion persons kept a lady in concealment in a wicker- 
work house, door and windows all closed up except one 
seinister and a forleas, ' a window and a skylight. 'f The 
house of Ailill and Maive at Croghan had a shutter of 
brass to each of its windows, and a fastening of brass to 
its for les. I 

Glass was known among various ancient nations from 
the most remote period : the Celts of Britain were well 
acquainted with it : and from constant references to it in 
our oldest writings, it is obvious that it was well known 
to the ancient Irish. § Beads and other small ornamental 
objects of glass, variously coloured, are constantly found in 
Irish pre-Christian graves and crannoges : and in one of 
the Loughcrew graves were found a number of them, one 
about an inch long, and — says Mr. Fergusson — obviously 
shaped by being softened by fire.|| The statement that this 
bead was softened by fire is quite true indeed, inasmuch as 
all the objects of this kind wherever found in Ireland were 
formed while the material was heated to softness. More- 
over the manufacture of these little articles was an art 
requiring long training and much delicate manipulative 
skill, for most of them are made of different coloured glass 

* Stokes, in Lives of SS., Pref. c. f Stokes, Da Derga, 19, 20. 

% Crowe, Tain bo Fr., 141 top. 

§ See, for examples, Miss Stokes, in Trans. Roy. Ir. Acad., xxx. 283 : 
Fled Bricrenn, 208 : Todd, St. Patrick, 222 : and Kilk. Arch. Journ., 
1879-82, p. 532. |i Fergusson, Rude Stone Mon., 218. 



[PART in 

or porcelain — blue, white, yellow, pale red, &c. — blended 
and moulded and beautifully striated in the manner shown 
imperfectly here in the black-and-white figures. They were 
used for ornamentation, very often forming the heads of 
pins, but sometimes made into rings, or strung together for 

One of the testimonies to the use of glass at a remote 
period in Ireland is the fact that it has two native Irish 

names — not derived — viz. 
gloinc or glaine [2-syll.], which 
signifies clearness and purity, 
from glan, ' pure,' ' clear,' 
V^^L^^I irWSr\/B * bright ' : and bus, which is 

explained by O'Davoren as 
'crystal or glass.' We often 
read of copans or 
cups of bus, corns 

Fig. 172. 

Fig. 173. 

Fig. 174. 

Fig. 175. 

Class and porcelain ornaments, full size, now in National Museum. In figures 172, 173, and 175, 
the coloured orn; ments form part of the substance, and were worked into shape while the whole 
mass was softened by heat. Figure 172, made of clear glass, with a yellow spiral ornament. 
Figure 173, of opaque light-green glass, grooved from top to bottom. Figure 174, body of deep 
blue, on which is twisted, and fastened while soft, an ornament of white enamel. Figure 17s, 
pin-head of fine light-red porcelain decorated with wavy stripes, some white, some yellow : found 
with part of bronze pin attached, as shown in figure. 

There are in the Museum many ornaments of coloured glass, with variously coloured patterns of 
enamel on the surface, of which the most beautiful is shown, full size, in figure 171. It is a circular 
disk, half-inch thick, the body of dark blue glass, with a wavy pattern of white enamel, like an open 
flower, on the surface. (AH, both figures and descriptions, from Wilde's Catalogue, pp. 162-165.) 

or horns of bus, &c. These words were also applied to 
natural crystal as well as to glass. Glass was turned to 
various uses. Glass drinking-vessels were known to the 
Irish at least as early as the sixth century. Adamnan 
(p. 147) relates that a druid named Broichan, foster-father 

* Sec Wilde's Catalogue, 162 to 169. 


of the Pictish King Brude, in the time of St. Columba, had 
a glass drinking-cup (vitream biberam) of great value which, 
as he was about to drink from it, fell and was broken into 
fragments : and vessels of glass are mentioned in the 
Lebar Brecc*. In the Tripartite Life of St. Patrick (p. 95) 
there is a legend of a stone altar found in a cave in Con- 
naught with four glass chalices (ceitri cailig glainidi) at 
the four corners. 

Glass and vessels of glass are frequently mentioned in 
the most ancient of the tales. In several passages of the 
Voyage of Maildune we read of lestars or drinking-vessels 
of glass : and in one part of the voyage he sails over a 
transparent sea " like green glass." In a sermon in LU, on 
the Day of Judgment, the six kinds of mercy by which 
heaven is to be attained are called " the six glass doors 
" through which comes the light of eternal life into the 
" church." f Add to all this that the remains of a regular 
glass factory have been found by the Rev. Mr. Ffrench in 
the townland of Moylisha, almost beside the ancient church 
of Aghowle in Wicklow, where great quantities of lumps of 
glass, chiefly of the three colours, blue, green, and white, 
have been — and can still be — dug up. The fuel used in 
the manufacture was charcoal, bits of which are found 
among the fragments (Kilk. Arch. Journ., 1885-6, p. 420). 

Glass was used in England for church windows in the 
seventh century ; and it had been long previously in use 
for this purpose on the Continent : so we may conclude 
that the knowledge of the use of glass for windows found 
its way into Ireland from Gaul, Italy, and England, through 
missionaries and merchants. £ At all events glass windows 
are mentioned in many of the ancient Irish tales, which 
shows that this use of glass was familiarly known to the 
original writers. In the Feast of Bricriu in LU, which was 

* Atkinson, Pass. & Homil., p. 48, „ 65 _ 
f Rev. Celt., iv. 249 : LU, 32, a, 29 . 
J See Petrie, Round Towers, 201. 


copied in noo from earlier books, we are told that Bricriu 
made an apartment for his own special use, with windows 
of glass (senistre glainide) on every side ; and he placed 
one over his own couch in his grccnan, so that he could 
have a full view of the banquet-hall and company through 
it.* In the same tale the house of Ailill and Maive, king 
and queen of Connaught in the first century, is described 
as having twelve windows closed up with glass, f There is 
of course bardic exaggeration in all this : still we are 
forced to believe that glass of some kind was used in Ire- 
land for windows, certainly before noo, and probably as 
early as the beginning of the historical period. 

There was one large door leading to the principal 
apartment of the dwelling-house, with smaller doors, open- 
ing externally, for the other rooms. Generally the several 
rooms did not communicate with each other internally. 
In the outer lis or rampart surrounding the homestead (for 
which see p. 54, below), there was a single large door. The 
doors of some great houses were very large and heavy. In 
the Boroma, Branduff and his companions " went forth 
" outside and shut the great royal doorleaf of the palace 
" behind them " (ro iadsadar in rig-comlai moir in rigthigi 
dara neise), " for the strength of nine men was in each of 
them I " : which implies that the comla or door was very 
massive. But such doors were exceptions, and those of 
ordinary houses were not larger than was necessary. The 
common Irish word for door was, and is, dorus : a single 
leaf of a door was comla. 

Sometimes there was a cairthe or stone column or pillar 
— or more than one — standing at the side of the outer or 
lis doorway. In the Mesca Ulad a person, seeing certain 
white objects in the distance, mistakes them for shields : 
but his companion says : — " They are not shields at all but 

* Ir. Texte, i. 254 ; and Fled Bricrenn (Henderson), p. 5 
f Ir. Texte, 1. 281 : Henderson, 69. 
X Rev Celt., xm. 61 Silva Gad., 410. 


" the [white] stone columns (colomna clock) that are in the 
" doors of these royal raths "* : and in the " Destruction of 
Dind Rig " we are told that at the door of the dun outside 
there was a cairthe or standing-stone. The knocker was 
a small log of wood called bas-chrann, i.e. ' hand-wood,' 
which lay in a niche by the door. It is everywhere 
mentioned in the old tales that visitors knocked with the 
bas-chrann. In rich people's houses there was a special 
doorkeeper (Irish doirseoir, doirside or doirsire), to answer 
knocks and admit visitors. At the bottom of the door 
was a tairsech or threshold. Cormac, in his Glossary 
(p. 161), derives this word from tairis, ' over it,' because 
" people pass inwards over it." It is a derivative from the 
Irish tars or tarsa, ' across,' connected with Latin trans. 

The jamb was anciently called aursa or irsa, but in the 
modern language it is ursa : the lintel was for-dorus (i.e. 
' on the door '), now usually fardorus. A certain student 
was making prostrations near the door of his hut, when he 
struck his head against the fordorus and fell dead.j On 
the outside of the large door of the lis was a porch called 
aurduine (lit. ' front part of the dun '). Cormac's Glossary 
explains aurduine as a structure " at the doors of the duns, 
which is made by the artisans " — implying ornamentation. 
The lis door was always closed at night. A more usual 
name for a door-porch was immdorus (irntn or im, ' about ' : 
' about a door '). In the Vision of Mac Conglinne (90, 91) 
persons are spoken of as carrying offal " from the immdorus 
of the great house to the immdorus of the dun or rampart 
on the outside." In O'Clery's Glossary immdorus is stated 
to be the same as fordorus, from which it would seem that 
the two words were sometimes used one for the other. 

The door was secured on the inside either by a bolt 
or by a lock. We have the best evidence to show that 
locks were used in Ireland in very early times. When 

* Mesca Ulad, 21 

t Fordorus : so in LL : see Stokes, Lives of SS., Pref. xi, note 5. 


St. Columba went to visit the Pictish King Brude, " the 
door of the fortress" — as the Irish Life of the Saint relates 
— " was shut against him, and at once, through Colum- 
cille's prayer, the iron locks (glais iarnaidhi) opened." 
We are told by Adamnan that a certain disciple on an 
important occasion peeped at St. Columba in his hut at 
Iona through the keyhole, and was soundly rated by the 
saint next day for his curiosity. Another similar occurrence 
ended less harmlessly. While Columba was surreptitiously 
copying St. Finnen's Book of Psalms at Drumfinn (vol. I., 
p. 501, supra), Finnen sent a messenger to spy out what he 
was doing, who looked through the keyhole and saw him 
at the work. But the saint's pet crane, happening to be 
with him, walked over to the door and neatly picked out 
the man's eye through the keyhole* 

Fig. 176. Fic. 177. 

Bronze Keys. Figure 176, a very perfect and highly decorated key, 2% inches long, 
with a pipe in the shaft : found in Tory Island. Figure 177, i$n inch long. (From 
Wilde's Catalogue.) 

In the romantic literature notices of locks and keys 
are equally common. In the story of Mongan in the Book 
of the Dun Cow, mention is made of a beautiful chamber 
locked and opened by a key.t And in the story of the 
Demon Chariot of Cuculainn in the same old book, we are 
told that the king's palace in the Isle of Skye had iron 
locks.J Locks were used for other purposes, as, for 
instance, to fasten fetters. In the seventh-century Life 
of St. Patrick by Muirchu, it is related that Maccuil put 
on his feet an iron fetter which he locked, and threw the 
key into the sea.§ The common word for a lock is glas. 

* Adamnan, 226, and note/;. 

t Voyage of Bran, 1. 55 : LTJ, 134, a, last line, and b, first line. 

X Kilk. Arch. Journ., 1870-1, p. 385. § Trip. Life, 288, 28 . 


A key is denoted by echuir or eochuir (gen. eochrach), 
which in Cormac's Glossary (p. 68) is derived from two 
words signifying ' crooked-straight ' : i.e. partly crooked 
and partly straight. A keyhole is poll-eochrach {poll, ' a 
hole '). Sometimes a door had a chain (slabrad) attached, 
which was probably used to fasten it. When Labraid and 
his men were about to set fire to the palace of Dinnree, 
they drew out the chain that was attached to the comla 
or double door, and put it on or round the pillar-stone 
outside, apparently to prevent the escape of those inside, 
who in a little time were all burned to death.* Mention 
is made of the aradh or ladder, which must have been in 
constant use. 

The houses were generally small, according to our idea 
of size. But then we must remember that, like the people 
of other ancient nations, the Irish had very little furniture. 
In the main room there was probably nothing — besides 
the couches— but a sufficient number of small movable 
seats and a large table of some sort, or perhaps a number 
of small tables. On this point it has been remarked that 
the Grianan of Ailech on Greenan-Ely near Derry, a 
circular building of uncemented stones, which was the 
palace of the Ulster kings, " gives a very poor idea of 
the extent of an ancient Irish regal abodef " : inasmuch 
as it was, as its ruins show, only seventy-seven feet in 
diameter. But this was merely the central keep or citadel. 
The dwelling of the king himself may have been within 
this enclosure, which afforded space enough for a respect- 
ably large house. The whole hill is surrounded by several 
earthen ramparts, one outside another, now nearly levelled, 
with broad spaces between (for which see page 91 below). 
In the intervening spaces timber houses were built, in 
which the chiefs and numerous dependents of the king 
lived : and probably the king himself had one or more 

* Stokes, Destruction of Dinnree, Zeitschr. Celt. Phil., III. 13. 
f Hennessy, Book of Fenagh, 63, note 3. 




outside the circular fortress. Many of the English and 
Anglo-Irish square castles, of which the ruins are seen to 
this day all through the country, were small and incon- 
venient to live in — often much smaller than the Greenan- 
Ely fortress : but most of them were merely citadels, which 
were originally surrounded by buildings of a lighter con- 
struction and more convenient size, in which the family 
and dependents customarily lived. 

fig. 178 

Trim Castle, originally built by Hugh de Lacy the Elder, end of twelfth century ; 
bat afterwards rebuilt. One of the Anglo-Norman strongholds referred to at p. 6s 
farther on. (From Cromwell's Tours. Drawn by Petrie.) 

Still the general run of houses were small in early 
times, in Ireland as elsewhere. Moreover the standard of 
living was in all countries low and rude compared with 
what we are now accustomed to ; a fact that ought to be 
borne in mind by the reader of the account given here of 
the domestic arrangements in ancient Irish houses. In 
England, even so late as the time of Holinshed — sixteenth 
century — hardly any houses had chimneys. A big fire of 
logs was kindled against the wall of the principal room, 
the smoke from which escaped through an orifice in the 
roof right overhead. Here the meat was cooked, and here 


the family dined. In very few houses were there beds 01 
bedrooms, and the general way of sleeping was on a pallet 
of straw covered with a sheet, under coverlets of various 
coarse materials, with a log of wood for a pillow : while 
the manner of eating, which is noticed farther on (page 
in), was correspondingly rude. All this is described for 
England by Roberts.* 

It is not easy to understand the statements given in the 
Brehon Laws as to the size of houses. For instance, the 
text of the Crith Gablach (iv. 311) says that a brewy or 
public hospitaller had " a house of 27 feet," and a backhouse 
or kitchen of 17 feet. But this is obviously a partial and 
imperfect statement — like so many others in the Brehon 
Laws ; for elsewhere we are told that he should be provided 
with all the necessary appliances to accommodate numer- 
ous guests, including, in case of one high class of brewy, 
100 beds. It is probable that the " house of 27 feet "was 
his own special residence, rectangular in shape, 27 feet 
wide : the length undetermined. In this case the little 
apartments for the family beds (see p. 46, infra) might be 
along one or both side walls : but if it was a circular house 
27 feet in diameter, some at least of the family beds must 
have been in separate houses outside. As for the beds 
for the brewy's guests, there must have been a number 
of separate houses for these. 

St. Patrick, in laying out the ecclesiastical buildings in 
Armagh, imitated the ancient fashion of the country, as he 
wisely did in most other things : for we are told in the 
Tripartite Life (p. 237) : — " In this wise then Patrick 
measured the ferta, namely, seven score feet in the lis 
[i.e. the circular rampart enclosing the whole establish- 
ment], and twenty-seven feet in the great house, and 
seventeen feet in the kitchen, and seven feet in the 
oratory : and in that wise it was that he used to found 
the congbala [ecclesiastical homesteads] always." 
* See Roberts, Soc. Hist., p. 318, 




We know that many of the great houses were very 
large. The present remains of the Tech Midchuarta or 
Banqueting-Hall of Tara measure 759 feet long and 
46 feet wide : and Petrie states that it must have been 
originally about 90 feet wide.* In the " Wooing of 
Emer " (p. 69) we are told that the measurement of the 
hall of Emain was " fifteen feet and nine score " : which 
refers to a square shape. 

FlC. 179. 

King John's Castle in Limerick. Erected in the beginning of the thirteenth 

century by one of the Anglo-Norman chiefs. Stanyhurst states that it was 

built by the order of King John. One of the Anglo-Norman castles referred 

to at p. 63 farther on. (From Mrs. Hall's Ireland.) 

We may form some idea of the better class of dwellings 
from an enumeration, in the Crith Gabhlach, of the various 
buildings in the homestead of a well-to-do farmer of the 
class bo-aire, who rented land from a chief and whose 
property was chiefly in cattle. His dwelling consisted 
of (at least) seven different houses, each as already observed 
with a separate wall, door, and roof : — 1. Dwelling-house 
(tech), at least 27 feet in diameter : 2. Kitchen or cooking- 

* Petrie's Tara, p. 185. 




house (ircha, or cuchtair, or cuile), at the back of the 
dwelling-house : 3. A kiln (aith) for drying corn : 4. A 
barn (saball) in which corn was stored : 5. A sheep-house 
(lias* cdirach) : 6. A calf-house (lias laeg) : 7. A pigsty 
(muc-foil or muccdl, from muc, ' a pig,' and foil or fail, ' a 
house ' — fail, ' dcmus,' Z., 5, 43 ).f These were all in one 

s ^H«k*»<!<.>f, 



FIG. 180. 
Conjectural plan of homestead of a well-to-do farmer of the bo-aire class, 
constructed from the descriptions given in the Brehon Laws. " Dw," family 
dwelling-house, of wickerwork, 27 feet in diameter, with three outside 
sleeping-rooms (which might be either round or rectangular) : " Kit," kitchen : 
"K," kiln (chiefly for corn-drying): " B," Barn: " C," calf-house: " P," 
pig-house: "S," sheep-house. The whole group surrounded by a circular 
rath or defensive entrenchment, with one entrance. The cows and horses 
were kept outside this enclosure. 

group close together ; and each generally, though not 
always, consisted of the usual round-shaped wicker-house 
with conical roof, except the barn, which was oblong : the 
whole group surrounded by the lis or rath, described farther 
on (p. 54) . In all houses of the more comfortable class the 

* This word lias (leece) was in very general use to denote a hut for the 
smaller animals — calves, sheep, lambs, &c It must be distinguished from 
les, lios, or lis, ' a rath or fort." | Brehon Laws, iv. 309, 311, 


kitchen was separate from the dwelling-house and placed 
at the back : and there was a separate pantry for provisions, 
called in Irish scallad, a word which glosses cellarium in 
some Old Irish documents.* 

From a fanciful derivation of sabhall, given in Senchus 
Mor (Br. Laws, I. 141), we may infer that a barn was 
oblong and had one side quite open, with the roof sup- 
ported at that side on posts. From the same derivation, 
and from other passages, f it would appear that while in 
some cases the barn belonged to the owner of the home- 
stead individually, in others it was common to the several 
families of the same fine, each householder using his own 
compartment for storing his corn : but in this case it must 
have been very large and detached — not situated within 
the enclosure of any private homestead. This arrange- 
ment could be adopted when the lisses and houses were 
near each other, forming a village or hamlet. 

The women had a separate apartment or a separate 
house in the sunniest and pleasantest part of the home- 
stead. This was called a grianan [greenan], which 
signifies a solarium, solar, or summer-house : a diminutive 
derivative from grian, ' the sun.' The women's greenan 
is constantly mentioned in Irish writings : and sometimes 
the master of the house had a greenan for himself, to 
which he could retire when he pleased. In Cruachan the 
greenan was placed over the fordorus or lintel, as much as 
to say it was placed in front over the common sitting- 
room : and probably it occupied some such position in 
most houses. 

In great houses there was one apartment called the 
house of conversation (tech immdcallamae) , answering to 
the modern " drawing-room," where the family often sat, 
especially to receive visitors. Prince Fraech, when he 
visited Maive at the palace of Cruachan on important 

* Stokes, Ir. Glosses in Tract on Declension, No 741. 
•f And see also Brehon Laws, iv. 305, mid. 




business, was always brought into the " house of con- 
versation " to discuss matters when an interview was 

Sometimes there was a small side-room beside the 
principal apartment, from which a door opened directly 
into it : having no separate outside door. This was called 
erdam or erdomh or aurdom, as it is given in Cormac's 
Glossary (p. 3 : ' on or by a house ' : dam or dom, ' house '). 
This plan was often adopted in Christian churches where a 
small apartment was placed at the side of the church.f 

Fig. 181. 

Bunratty Castle m the south of Clare, on the Bunratty river, where it joins the Shannon : built 
about the end of the thirteenth century by Thomas de Clare, an Anglo-Norman lord. One of the 
Anglo-Norman castles referred to in sect. 3 below. (From Kilk. Archa:ol. Journ., 1890-91, p. 292.) 

The privy was called fialtech, i. e. ' veil-house ' {fial, ' a 
veil ' : tech, ' house ') : the urinary was fualtech, from fual, 
' urine.' In the Rule for the Culdee Monks, both houses 
are said to be the abode of demons ; and whoever goes to 
them is enjoined to bless them and also to bless himself — 
i.e. to make the sign of the cross. £ The fialtech and 

♦Tain bo Fr., 143, 145. 

t See theErdamh or Erdam discussed at length in Petrie's Round Towers 
(Index), + Reeves, Culdees, 91. 


the fualtech are often referred to incidentally, but most 
often in connexion with monasteries. 

Maigens or Sanctuaries. — The plot of land around the 
house of a person of rank was a sort of asylum. This was 
called a maigen or precinct : and within it no man should 
break the peace without the consent of the owner. The 
higher the rank the larger the maigen. The maigen of a 
bo-aire, the lowest rank entitled to the privilege, was the 
smallest : it extended the cast of a spear all round his 
house. That of an aire-desa extended two casts. The 
extent doubled for each rank upwards to the king of the 
tuath, whose maigen extended sixty-four casts round his 
residence. The maigen of a provincial king or of the king 
of Ireland included tho whole plain on which the palace 
stood. There was also a maigen — varying according to 
rank — round the dwelling of an ecclesiastic, and also round 
a church : the sanctuary of a church was often called 
Termon land (i. 358, supra). The Archbishop of Armagh 
had the same extent of maigen as the king of Ireland. It 
will be mentioned farther on (page 173) that every bruden 
or first-class hostel was also an asylum. 

A fugitive, no matter what his crime, and also whatever 
property he had with him, whether belonging to himself or 
to the pursuer, once he entered on a maigen, were safe for 
the time, provided the regulations were complied with. 
The following conditions and formalities were necessary 
to ensure his safety : — 1. The owner, or some member of 
his family legally entitled to act for him in such cases, 
should give permission to the fugitive to enter on the 
precinct, and should persist in claiming asylum for him. 
2. The owner or his deputy should inform the pursuer that 
the place was a precinct. 3. The owner or deputy should 
guarantee that no loss should accrue to the pursuer or 
aggrieved party by the temporary shelter afforded to the 
fugitive — that the original claim should hold good — that 
the fugitive should not be enabled to finally escape from 


justice. If any one of these failed he might be arrested 
on the maigen. A person who committed any act of 
violence within a maigen — provided he knew it was one, 
and that the necessary formalities were observed — had to 
pay damages to the owner, the amount depending on 
honour-price, on the extent of the violence, and on other 

This law of sanctuary in and around a house existed 
also in early times in England, and in a form almost 
identical with that laid down in the Brehon Law.* 

This is a proper place to observe that there was an 
all-important distinction between the asylum-right of a 
private residence and that of a church or a hostel. The 
right accorded to the maigen of a dwelling was for the 
protection of the owner against scenes of violence on his 
premises by outsiders — not primarily in the interest of the 
fugitive ; and as it depended on the will — or caprice — of 
the owner, it was uncertain. It was indeed not an asylum 
at all in the proper sense of the word. But the sanctuary 
of a church or the asylum of a bruden was absolute and 
inviolable, depending on no conditions and on no man's 
will or caprice. 

2. Interior Arrangements and Sleeping Accommodation. 

It will be shown farther on (p. 52) that in large houses 
there were separate sleeping-rooms. But among the 
ordinary run of comfortable, well-to-do people, including 
many of the upper classes, the family commonly lived, ate, 
and slept in the one principal apartment, f as was the case 
in the houses of the Anglo-Saxons, the English, the 
Germans, and the Scandinavians of the same period. In 
the better class of houses in Ireland there were, ranged 
along the wall, little compartments or cubicles, each con- 

* Brehon Laws, III., Introd. by Richey, ciii. For the whole law of 
Precincts see Brehon Laws, iv. 277. See also HI. 119 to 145. 

t For examples, see Mac Conglinne, 58 ; and Tromdamh, 51, 55, 61. 




taining a bed, or sometimes more, for one or more persons, 
with its head to the wall. The wooden partitions enclosing 
the beds were not carried up to the roof ; they were prob- 
ably about eight or nine feet high, so that the several com- 
partments were open at top. A little compartment of this 
kind, whether open or closed overhead, was called an imda. 
The primary meaning of imda is a ' bed,' as is clearly 
indicated in Cormac's Glossary (p. 6), where it is stated that 
the adhart or ' pillow ' is so called because it is higher than 







FIG. i8a. 
Conjectural plan of a good-class house, where the family lived, ate, and slept 
In the one large apartment : constructed from the descriptions in Tales and 
Brehon Laws. (House here made quadrangular, but might be round or oval 
Eight imdas, cubicles, or sleeping-places, each with one bed : some beds for 
one person, some for two, some for three. Four low, small tables and a number 
of seats are shown, all movable. Seats at ends of cubicles outside are fixeu. 
Five supporting posts (shown by little circles) : fire near middle. The openings 
or windows in walls arc not marked ; neither are the doors in doorways of house 
and cubicles. 

the rest of the imda or bed. But by a natural extension 
of meaning the word is often used to denote the whole 
compartment or cubicle with its bedstead. Sometimes the 
word imda was applied to a large room : for example, in 
Kineth O'Hartigan's poem it is stated that in Tara there 
were fifty imdas, each with fifty men in it.* But the usual 
meaning was either a bed or the little room containing a 
bed. In the Bruden Da Derga, Ingcel goes to reconnoitre 
the hostel that his party are about to attack. He sees 
many imdas of different sizes with men in them varying in 
number from one up to nine. From the whole context it 

* Petrie's Tara, p. 190 


is plain that these imdas were not couches but little com- 
partments open in front, so that all those in them could be 
seen sitting or reclining on couches : and accordingly the 
persons are described all through, not as on the imda, but 
as in it. Let us add that Dr. Stokes, in his edition of this 
story, always translates imda by ' room.' 

At the foot of each imda outside, and projecting into 
the main room, there was a low fixed seat, often stuffed 
with some soft material, for use during the day. Besides 
these there were on the floor of the main apartment a 
number of detached movable day couches or seats — all 
low — with one or more low tables of some sort. 

The preceding description of the disposition of the beds 
applies to the better class of houses. The lower classes of 
people probably slept, like those of Wales and Scotland of 
those times, on beds or pallets ranged along the wall with 
little or no attempt to screen one from another. Giraldus 
describes the Welsh as sleeping in this manner with their 
heads to the circular wall and their feet towards the fire. 

The fire was in or near the middle, and the people sat 
or reclined by day all round it ; while the smoke escaped 
through an opening in the roof : a custom which, as Scott 
records, existed in Scotland down to 200 years ago.* In 
England also, down to the time of Elizabeth, before coal 
was brought into domestic use, and when wood was the 
general fuel, there were hardly any chimneys, and the fire 
was lighted— as in Ireland and Scotland — in the centre of 
the single big room or hall, or up against one of the walls, 
the smoke escaping through a hole in the roof.f 

That these arrangements for living and sleeping were 
in general use in Ireland is abundantly plain from many 
passages in the old writings. We find the expression, 
so many " imdas from fire to side wall in the house all 

* Rob Roy, chap, xxviii. 

t Roberts, Social Hist. pp. 325, 348 ; see also Mr. P. H. Newman in 
" Social England," 1. 225. 


round "* constantly used in the tales. That there were 
seats distinct from those at the ends of the cubicles appears 
from a passage in the Crith Gabhlach describing the house 
and furniture of an aire-tuisi. chief, in which we read that 
there were " eight imdas with their proper furniture [namely, 
" bedsteads with beds and end-seats], besides six couches 
" [brothrach, ' a couch ' : pi. brothracha], with their proper 
" furniture both pillows and sitting-skins '* (i.e. skins stuffed 
with feathers). f All this shows that the ancient Irish of the 
higher classes had two distinct kinds of couches : a couch 
or bed for sleeping on, and another sort for sitting or re- 
clining on at meals, or on other occasions during the day ; 
just as the Romans had their two kinds of lecii or couches 
for the same two purposes. 

The bedstead within the imda, in the best class of 
houses, consisted of four pillars connected by rails, with 
a canopy overhead, and curtains running by rings on 
copper rods. J Such a bed was designated lige cumtachta, 
i.e. a ' protected,' enclosed, or testered bed : and this 
designation occurs so often that such beds must have 
been pretty common. Near the foot of the bed and 
within the imda there was a rack with pins or hooks 
for hanging clothes or other articles on.§ Lige or ligi 
[lee] was a usual term for a bed, cognate with Lat. lecius : 
but the commonest name was lej>ad, which, in the form 
leaba or leabadh [labba], is the term in use at the present 
day. This word was also used to denote a couch for day 
use, which had generally a little table beside if for food 
and drink. St. Patrick, when at Tara, was summoned to 
King Laegaire's kpad in the banquet-room to have some 
food. |! Both lige and hpad were applied to a grave. The 
word sceng [skeng] also means ' a bed,' though not often 
used : and hence the enclosure round a bed or couch was 

* Tain bo Fr., 139. \ Silva Gad., 120. 

t Brehon Laws, iv. 326, 6; 327, ,,. § Brehon Laws, iv. 75, bot. 
[1 Trip. Life, 55 


often called imscing [im, ' about '). In Cormac's Glossary 
(p. 98) we have " imscing, a little house [or apartment] 
in which a bed [imdae] fits " : and again (p. 150), " sceng, 
i.e. iumdha, a bed, whence imscing, a small both or tent 
which surrounds a bed." Sceng and imscing, like lefiad, 
were applied also to day-couches. King Domnall, at the 
banquet of Dun-nan-gedh, sat or reclined at the head of 
the table in his golden imscing (Moyr. 29) : i.e. an imscing 
ornamented with gold. 

A bedframe or bedstead is often called tolg. We are 
told in an ancient book of Irish annals, and also in the 
story of the Boroma, that Feredach, king of Ossory, in the 
sixth century, falling very sick, had to lie abed : and he 
caused to be brought to him all his treasures, which he 
kept beside him in his tolg : — " For it was [then] the 
" custom of kings to have couches (tolgs) of yew around 
" them, in which they had a collection of their bars and 
" ingots of silver, and their [valuable] cups and vessels, and 
" their chessmen and chessboards, and their camans or 
" hurleys."* The fierce old warrior Cellach, unable to 
walk or move about on account of his great age, had a 
brass tolg as his bed, in which he always remained : and 
his only treasure and consolation was his sword, which 
he kept unknown to all, hidden under the bedclothes, f 
The practice of keeping a sword in bed must have been 
common : Dalian mac Moire, chief poet to Cerball king of 
Leinster (a.d. 885-909), in his poem addressed to Cerball's 
sword, asks : " from the day that Cerball departed, with 
whom shalt thou have bed-fellowship ? " {lepthanas) .% 

As distinct from the imda and bedstead, the bed-tick or 
mattress was called dergud [dergu], a distinction clearly 
pointed out in this passage from the voyage of Maildune : — 
" There were seventeen canopied imdas in the house, with 

* Three Fragments of Annals, 9 : Silva Gad., 416. 
t Moyrath, 43. 

I Kuno Meyer in Rev. Celt., xx., p. 12 : LL, 47, b, 13 from bottom. 



good der guds {dagdergudhaib) set in them."* The word 
colcaid (a loan-word from Lat. culcita) was also often applied 
to a bed or bed-tick : O'Donovan always renders it ' a flock- 
bed ' : but whatever sort of bed it was, it must have been 
regarded as a luxury, for we are told in Cormac's Glossary 
(p. 44) that it was used by nobles. This word was also 
often applied to a quilt or other bed-covering ; having 
undergone a change of meaning like the English word 
quill, which also comes from culcita. The blanket (setigi) 
and other bed-covering were brought out by day to be 
aired and sunned, f White linen sheets were used, and in 
grand houses they were often embroidered with figures. J 

Beds of the best class were stuffed with feathers. 
St. Columba is made to prophesy of a certain king that 
he would not be killed in battle, but that he would die 
on his own feather-bed (filumaliuncvlam) .§ Some of the 
beds in the guest-house of the Cork monastery were made 
of feathers. Straw was sometimes used : Mac Conglinne 
(p. 14) growls by way of contempt that the attendant who 
waited on him in Cork monastery had no better way of 
lighting the fire in the guest-house than by pulling a wisp 
of straw from the bed. The Book of Aicill,|| defining the 
penalty for stealing straw, lays down a double fine if 
the straw was intended to be put as beds under people ; 
which indicates that it was subjected to some sort of 
preparation. Rushes were sometimes used for beds — as in 
Wales^f — especially in cases of emergency or for tempo- 
rary use. When Cuculainn and Ferdiad had finished their 
day's fighting, their attendants prepared beds of fresh rushes 
for them.** When the Fena of Erin were out on their 
hunting excursions, they put up hunting-booths each 
evening, after which — to use the words of Keating : — 

* Rev. Celt., x. 65. t Mac Conglinne, io, 22 . 

% Leahy, Courtship of Ferb. 29 : LL, 256, b, top line: Three Fragm., 11, 17 . 
§ Adamnan, 44. ^ Giraldus, Descr. of Wales, 1. x. 

)| Brehon Laws, 111. 151, IS> ^ ** O'Curry, Man. & Cust., 11. 439. 


" Each man constructed his bed of the brushwood of the 
" forest, moss, and fresh rushes. The brushwood was laid 
" next the ground ; over that was laid the moss ; and the 
" fresh rushes were spread over all : which three materials 
" are designated in old books tri cuilcedha na Feine, ' the 
" three bed-materials of the Fena.' ' The people often 
used beds of hides stuffed with some soft material : or 
perhaps they simply spread the skin on the top of straw 
or rushes. The Senchus Mor mentions " a poor sick man 
lying on the hides."* 

1'IG. 183. 
Castle of Athlone : erected by John de Grey, Lord Justiciary, or Governor, or 
Ireland, 1210-1313. One of the Anglo-Norman castles referred to at p. 65, infra. 
(From Mrs. Hall's Ireland.) 

A pillow was used for the head. The most common 
word for a pillow was adart [ey-art], which is used to 
this day by speakers of Irish. A fanciful derivation of 
the word given in Cormac's Glossary (p. 6) indicates 
clearly the nature of the article : — " Adart, i.e. ath-ard, 
" additional height, because it is higher than the rest 
" of the bed," Sometimes frithadart was used ; and a 

* Brehon Laws, i. 195. 


passage in Fiach's Hymn, in which this word occurs, 
also indicates the distinction between the bed and the 
pillow : — " He [St. Patrick] slept on a bare flagstone, a 
pillar- stone was his pillow" (frithadart) * Another name 
for a pillow was cerchaill or cenncherchaill (where cenn is 
' head '). Cormac's Glossary (p. 38) defines cerchaill as 
" head-protection." From the same passage we learn that 
a pillow was filled with feathers, and that the case was 
[sometimes] made from the skin of a wild deer (" It is of 
" his hide [the hide of a wild deer] that the case for the 
" feathers is made."). 

Often two, and sometimes three, persons slept in the 
same bed. St. Patrick placed the youth Aed, the king of 
Leinster's son, in charge of Cascorach the minstrel, say- 
ing : — " Let the king of Leinster's son be in one bed {in 
" aeinlebaid) and in one condition with thee till we reach 
" Leinster."f When St. Caillin visited the O'Cahalans of 
Connaught, they received him so well that he blessed them 
and prophesied that there should be always among them 
some chief who would be [so much esteemed as to be] 
selected as a king's bed-fellow. % It was a mark of distinc- 
tion to set apart a bed for one. Maildune and his men 
came to a certain house in which were a number of bed- 
couches, one intended for Maildune alone, and each of the 
others for three of his people§ : and in another place was 
a house with a number of large beds, each for three of 
the household, and one smaller bed for the master of 
the house. || One of the complaints of the unreasonable 
demands of the poets who were on a visit to Guaire 
king of Connaught was that they insisted on a separate 
bed for each.^T 

In great homesteads there were sleeping-houses or 
apartments distinct and separate both from the sitting- 

* Trip. Life, 408-9. § Joyce, Old Celtic Romances, 141. 

f Silva Gad., 205. || Ibid., 125. 

J Book of Fenagh, 179, 185. ^ Tromdamh 41, IS; 109, verse. 


or banquet-room and from one another, each probably 

circular and having a conical roof of its own : often called 

tech-leptha, i.e. ' bed-house.' When the three Red Branch 

champions came to the palace of Cruachan, Ailill and 

Maive gave them the choice of a house (tech) for each, 

or one house for the three : and they selected three 

houses, in each of which was a bed.* " We have distinct 

' statements in our ancient records " — says O' Curry — " that 

' different members of the same family had distinct houses 

' (and not apartments) within the same rath, dun, lis, or 

' cathair : that the lord or master had a sleeping-house, 

' his wife a sleeping-house, his sons and daughters, if he 

' had such, separate sleeping-houses, and so on, besides 

' places of reception for strangers and visitors. "f But 

this applies to the great houses belonging to people of 

rank. Even in high-class houses, however, it was usual to 

put two or three in the same room, with a bed for each.J 

People of the upper classes sat upon seats covered with 

skins. St. Patrick's chariot-seat was covered with the skin 

either of a cow or of a seal : both are mentioned in the 

Tripartite Life (p. 75) as in use. Skins for sitting on 

(Gaimniu suidi) are noticed in the Crith Gabhlach§ as in 

the house of an aire-tuisi chief. In Kuno Meyer's Liadain 

and Curithir (p. 23) is mentioned a couch covered with 

white fleeces of [sheep-] skins ; and Cormac's Glossary 

(p. 81) quotes this verse from a poem, ancient in his time : 

" It is delightful for me to be [sitting] on a yearling calf's 

skin in Garbhan's house." 

It was a common practice in the better class of houses 
to strew the floor with rushes : and when distinguished 
visitors were expected, the old rushes were removed and 
fresh ones supplied. The women-servants always managed 
this business. When Murkertagh of the Leather Cloaks 

* Fled Brier., 69. f Man. & Cust., II. 70. 

% See Silva Gad., 52 mid. and 102, 9; and Hyde, Lit. Hist., 295. 
§ Brehon Laws, iv. 326, 8 , 


was approaching Ailech his home, with many royal captives, 
after his circuit round Ireland, a.d. 941, he sent on a page 
the day before with directions that women should be sent 
to cut rushes for the floor.* The use of rushes for this 
purpose was so well understood that there was a special 
knife for cutting them ; and such a knife is enumerated 
among the household articles in the house of a brugh-fer 
or brewy.f Sometimes the floor was covered with soft 
green-leaved birch-branches with rushes strewn over them 4 
We know that this custom of covering the floor with rushes 
also prevailed in England, where it was continued down 
to the time of Elizabeth. In some of the inferior apart- 
ments of Irish houses, straw was used : for example, it 
was expected that the kitchen of a bo-aire chief should 
be kept strewn with fresh straw,§ which one would think 
a dangerous practice. 

3. Older Premises and Defence. 

The homesteads had to be fenced in to protect them 
from robbers and wild animals. This was usually done 
by digging a deep circular trench, the clay from which 
was thrown up on the inside. This was shaped and faced ; 
and thus was formed, all round, a high mound or dyke 
with a trench outside, and having one opening for a door 
or gate. Whenever water was at hand the trench was 
flooded as an additional security : and there was a bridge 
opposite the opening, which was raised, or closed in some 
way, at night. The houses of the Gauls were fenced round 
in a similar manner. Houses built and fortified in the 
way here described continued in use in Ireland till the 
thirteenth or fourteenth century (see Westropp's " Anc. 
Forts of Ireland," p. 624). 

* Circuit, 53 and note : see also Mesca, 13 top : and O'Curry, Man. & 
Cust., 11. 13, „, f Brehon Laws, iv. 311. 

X Leahy, Courtship of Ferb. 8 ; LL, 253, b, 23. 
§ Brehon Laws, iv. 315 top. 


These old circular forts are found in every part of 
Ireland, but more in the south and west than elsewhere ; 
many of them still very perfect — but of course the timber 
houses are all gone. Almost all are believed in popular 
superstition to be the haunts of fairies. They are now 
known by various names — lis, rath, brugh, mur, dun, moat, 
caiseal [cashel], and cathair [caher] : the cashels, murs, and 
cahers being usually built of stone without mortar. These 
are generally the very names found in the oldest manu- 
scripts. The forts vary in size from 40 or 50 feet in 
diameter, through all intermediate stages up to 1500 feet : 
the size of the homestead depending on the rank or means 

Fig. 184. 

The great " Moat of Kilfinnane," Co. Limerick, believed to be Trtda-na-Rec, the 
triple-fossed fort of the kings, one of the seats of the kings of Minister. Total 
diameter 320 feet. (From a drawing by the author, 1854.) 

of the owner. Very often the flat middle space is raised 
to a higher level than the surrounding land, and sometimes 
there is a great mound in the centre, with a flat top, on 
which the strong wooden house of the chief stood.* 

Forts of this exact type are still to be seen in England, 
Wales, and Scotland, as well as in various parts of the 
Continent ; and the figure of an existing one near Geisel- 
berg in Germany, given by Borlase (p. 1128), might be 
mistaken for a drawing of some of those in Ireland. 
Round the very large forts there are often three or more 

* On this point see the instructive letter of the Welsh antiquary, Mr. 
Geo. T. Clarke, in Stokes's Life of Petrie, p. 216 ; showing that the same 
custom existed in England and Normandy : and see also Mr. Westropp's 
Essay (Ancient Forts of Ireland), p. 585, in which are given, from th? 
Bayeux Tapestry, representations of houses on the tops of forts. 




great circumvallations, sometimes as many as seven* 
The " moat or fort of Kilfinnane," figured above, has three. 
A dim, sometimes also called dind, dinn, and dingna, 
was the residence of a Ri [ree] or king : according to law 
it should have at least two surrounding walls with water 
between. f Round the great forts of kings or chiefs were 
grouped the timber dwellings of the fudirs and other 
dependents who were not of the immediate household, 
forming a sort of village. Any great fortified residence of 
this kind was often called port', in Cuimmin's Poem on 
the Saints of Ireland, Armagh is called Port Macha.% 

Fig. 185. 

Section of an underground heehive<shaped liut. (From Wood-Martin's 
Pagan Ireland, p. 205.) 

In most of the forts, both large and small, whether with 
flat areas or with raised mounds, there are underground 
chambers, commonly beehive-shaped, which were probably 
used as storehouses, and in case of sudden attack as places 
of refuge for women and children. In the ancient litera- 
ture there are many references to them as places of refuge.§ 
The Irish did not then know the use of mortar, or how 
to build an arch, any more than the ancient Greeks ; and 
these chambers are of dry-stone work, built with much 

* See Kilk. Arch. Journ., 1870-71, p. 387, verse xi. 

t O'Curry, quoting Brehon Laws, Man. & Cust., 11. 3, 4. 

X Stokes, in Zeitschr. Celt. Phil., 1. 72. 

§ Miss Stokes, Early Chr. Art, 3 : Kinahan, On Luscas [or caves] in 
Raths, in Kilk. Arch. Journ., 1883-4, P- ll '• ar *d Westropp, Ancient Forts, 
p. 666. 


rude skill, the dome being formed by the projection of 
one stone beyond another, till the top was closed in by 
a single flag. 

Where stone was abundant the surrounding rampart 
was often built of dry masonry, the stones being fitted 
with great exactness. In some of these structures the 
stones are very large, and then the style of building is 
termed cyclopean. Many great stone fortresses of the 
kind described here, usually called caher, Irish cathair, 
still remain near the coasts of Sligo, Galway, Clare, and 
Kerry, and a few in Antrim and Donegal : two charac- 
teristic examples are Greenan-Ely, the ancient palace of 

Fig. 186. 

Staigue Fort in Kerry. Of stones without mortar. External diameter 114 feet ; internal, 
88 feet: wall 13 feet thick at bottom, 5 feet at top. (From Wood-Martin's Pagan Ireland, 
p. 180, and that from Wilde's Catalogue, p. 120, where a further description of this fort will 
be found.) 

the kings .of the northern Hy Neill, in Donegal,* and 
Staigue Fort near Sneem in Kerry. The most magnificent 
fortress of this kind in all Ireland is Dun Aengus on a 
perpendicular cliff right over the Atlantic Ocean on the 
south coast of Great Aran Island (see next page). 

At the most accessible side of some of these stone 
cahers, or all round if necessary, were placed a number 
of large standing stones firmly fixed in the ground, in 
no order — quite irregular — and a few feet apart. This 
was a very effectual precaution against a sudden rush of 
a body of assailants. Beside some of the existing cahers 
these stones, or large numbers of them, still remain in their 
places (shown in figs. 187 and 188). 

* For which see sect. 5 of this chapter below. 


[PART ill 

The caiseal or cashel was a strong stone wall round a 
king's house, or round a monastery ; of uncemented stones 
in pagan times, but often built with mortar when in con- 
nexion with monasteries. The caher was distinguished 
from the cashel by being generally more massive in 
structure, with much thicker walls. The cahers are almost 
confined to the south and west of Ireland.* Buildings like 
our cahers are also found on the Continent, as mentioned 
by Borlase (pp. n 26-1 129). 

That the wooden dwelling-houses were erected within 
the enclosing lios, les, or rath, is abundantly evident from 

FIG. 187. 

Dun-Aengus on the great Island of Aran, on the edge of a cliff overhanging the 

sea : circular Firbolg caher : without mortar : the standing-stones were intended to 

prevent a rush of a body of enemies. (Drawn for Dr. Wilde : published in Arch. 

Cambr., 1858 : and subsequently in Wilde's Lough Corrib.) 

the records. Queen Medb (or Maive) Lethderg (not Queen 
Maive of Croghan) is recorded to have built the rath near 
Tara, now called from her, Rath-Maive : " and she built a 
choice house within that rath."f There were often several 
dwelling-houses within one large rath: inside the great 
rath at Emain there were at least three large houses, with 
others smaller! : the Rath-na-Righ at Tara had several 
houses within it : and in the romantic story of Cormac in 
Fairyland, we are told that he saw " a very large kingly 
dun which had four houses within it." 

* On these fortified residences, see the valuable article in Stokes's 
Life of Petrie, p. 216, et seq. 

f O'Curry, MS. Mat., 480. \ O'Curry, Man. & Cust., II., II. 


There is good reason to believe that originally the 
word rath was applied to the surrounding embankment or 
rampart, and lios or les to the space enclosed. Thus a 
person who was making his way towards the palace, leaped 
over the three raths, until he was on the floor of the les, 
and from that until he was on the floor of the royal house 
(for lar r?g-tkige)* : a passage which moreover affords 
additional testimony that the houses were built within the 
enclosure. Again, in the tale of the sons of Usna, it is 
told that the child shrieked " so that it was heard all over 
the lis."^ But these distinctions have long since ceased 
to be observed : and the words rath and lis are now 
applied to the whole structure. 

The rampart enclosing a homestead was usually planted 
on top with bushes or trees, or with a close thick hedge, 
for shelter and security : or there was a strong palisade on 
it : — Co n-accai in liss m-bileach m-barrach : " so that I saw 
a liss topped with trees. "} Tuittid cnci cuill cainmessa 
dobilib rath : " the fair-fruit hazel-nuts fall from the trees 
of the raths. "§ Lisses and raths such as we see through 
the country are generally round or oval : but they are 
occasionally quadrangular or square. J| Vitrified forts, i.e. 
having the clay, gravel, or stone of the rampart converted 
into a coarse glassy substance through the agency of 
enormous fires, are found in various parts of Ireland as 
well as in Scotland : and similar forts are still to be seen 
in several parts of the Continent. ^J 

Sometimes outside the outer earthen or stone rampart 
there was a timber palisade of strong stakes, enclosing a 

* Voyage of Bran, i. 47, I7i and 51, , . see also Kilk. Arch. Journ., 
1868-9, P- 223 : and 1870-1, p. 447. 

f O'Curry, 3 Sorrowful Stories, Atlantis, in. 399. 

% Mac Conglinne, 68. § LL, 118, a, 16. 

|| Kilk. Arch. Journ., 1849-51, pp. 23, 24 ; 1867, p. 4 : and Mr. West- 
ropp's Essay on the Ancient Forts of Ireland, p. 583. 

If See Wood-Martin, Pagan Ireland, 183 : Kilk. Arch. Journ., 1879- 
82, p. 756: Borlase, 1126: Stokes's Petrie, 223, 357: Proc. Roy. Ir. 
Acad., v., p. 69 ; and Adamnan, 151, note b. 




large area : this palisade was called sonnacJi (and sometimes 
tonnach), from sonn, a cuaille or ' stake.' Aed Guaire, king 
of Connaught in the sixth century, built, in preparation for 
a marriage-feast for his wife, a new [wooden] house {tedi) 
within a dun or double circular rampart ; and round the 
di'in again, that is, outside all, he made " a sonnach of red 
oak." Just as it was finished Aed Baclamh, the king of 
Ireland's spear-bearer, made a circuit round the kingdom 
to test obedience and discipline among the provincial and 
minor kings : and he demanded that all doors should be 


Fig. 188. 

Ballykinvarga Stone Fort (mortarless) near Kilfenom in Clare. Shape, oval ; 155 feet by 135 feet. 
Showing chrvaux de/rise of standing-stones, to prevent a rush. (See Mr. Westropp's description of 
this fort in Journ. Roy. Soc. Antiqq., Ire!., for 1897, p. 121 ; and p. 57, supra.) 

broken open wide enough to permit him to enter freely 
with his long spear held crosswise. Things went well with 
him till he came to Aed's new house. Aed went so far as 
to break an opening in the outer palisade : but when the 
ill-grained visitor demanded that a corresponding opening 
should be made in the elaborately carved door of the 
house, he found to his cost that he had met his match ; 
for Aed, flying into a rage, struck off his head with one 
blow of his sword.* 

* O'Grady, Silva Gadelica, 70. 




Immediately outside the outer door of the rath was an 
ornamental lawn or green called aurla, a name often varied 
to urla and erla, which was regarded as forming part of 
the homestead : " then queen Maiye went out through the 
" door of the liss into the aurla [isin n-aurlainn], and three 
" times fifty maidens along with her."* So also prince 
Cummascach, when he visited Branduff, king of Leinster at 
Baltinglass, pitched his tent on the erla of the king's haile 

Fig 189 
Carlow Castle in 1845 : believed to have been erected by Hugh de Lacy, who was 
appointed Governor of Ireland in 1179 One of the Anglo-Norman castles referred 
to at p. 6s, infra, (From Mrs. Hall's Ireland.) 

or homestead, f Beside the dun or lis, but beyond and 
distinct from the aurla — and outside the sonnach if there 
was one — was a large level sward or green called a faithche 
[faha] — commonly Latinised ftlatea or plateola% — which 
was chiefly used for athletic exercises and games of various 
kinds : it was sometimes called blai.§ Some idea of its 

* Ir. Texte, i. 280 : Fled Brier., 69. f Silva Gad., 408. 

\ Adamnan, 98 (f?) ; 360 ; 450. 

§ See Windisch, Wurterbuch, Ir. Texte, 1., " Bla," land 


size may be formed from the statement in the law that the 
faithche of a brewy extends as far as the voice of a bell 
{i.e. of the small bell of those times) or the crowing of a 
cock can be heard.* Finn mac Cumail when a boy, coming 
one day to a dun, found a number of youths hurling (oc 
imdin) on the faithche.] The law lays down certain 
regulations regarding the striking of the ball on the 
faithche in hurling. % When not formally measured and 
enclosed, the four fields nearest the house were understood 
to constitute the faithche.^ A visitor was free to go upon 
the faithche and could not be sued for trespass, " for every 
faithche is free " [to all comers] ,|| The faithche was not to 
remain profitless : animals, commonly sheep, were kept 
grazing on it.*[ The haggard for grain-stacks, which was 
always hear the homestead, was called ithla (gen. ithlann), 
from ith, ' corn.' The ithla, like the barn, sometimes 
belonged to an individual, and sometimes to the fine or 
clan, of which each householder had his share** : but in 
this case the ithla was very large, standing apart, and 
unconnected with any one homestead. A garden or 
enclosure at the back, fenced in for general purposes, was 
often called airlis.]] The lubgort, or ' kitchen-garden,' 
will be spoken of at p. 148, infra. 

At a little distance from the dwelling it was usual to 
enclose an area with a strong rampart, into which the 
cattle were driven for safety by night. This was what 
was called a badhun (bawn], i.e. ' cow-keep,' from ha, 
pi. of bo, ' a cow,' and dun%% : and sometimes bo-dhaingen 
[bo-ang-in], which has the same meaning (daingen, ' a 
stronghold '). This custom continued down to a late 

* Brehon Laws, iv. 195. As to ^ Br. Laws, iv. 311 bottom, 
these measurements, see pp. 374 ** Ibid., 1. 125, 141 : in. 285. 

and 375, infra. ft Ibid., iv. 313 bottom. 

f Oss. Soc, iv. 295. XX See Moylena, 183 (whole page): 

X Brehon Laws, in. 253. and Joyce, Irish Names of Places, 

§ Corm. Gloss., 78. 1. 308. 

|| Brehon Laws, III. 253. 


time : and was adopted by the English and Scotch settlers. 
One class of the planters who were settled in Ulster in the 
time of James I. were required to build a castle and a 
bawn. The ruins of many of these settlement-bawns still 

The outer defence, whether of clay, or stone, or timber, 
that surrounded the homestead was generally whitened with 
lime. When Nuada the druid built a dun on Almu, now 
the Hill of Allen in Kildare, " he rubbed the sund or sunn 
" or outer rampart with alamu (ro colmed alamu dia sund) 
" until it was all white. "f The text does not tell us what 
this colouring stuff alamu or almu was. Stokes (Acallamh 
283) makes it alum ; and we know that alum is a native 
product, with which people have been familiar from early 
times. J It is indeed probable that alum was in the 
writer's mind ; for the impossibility of procuring so much 
of it as would whiten a whole immense rampart would be 
no difficulty to an etymologist who invented the episode 
to account for the name. The dun was made white at 
any rate ; and it is pretty certain that lime was the real 
material ; which seems borne out by an old verse relating 
to Almu, quoted in the story in the Book of the Dun 
Cow : — 

" All white is the dun of battle renown 
As if it had received the lime of Ireland." 

That the outer rampart of duns or homesteads was 
often whitened with lime is shown by many other passages. 
Maildune comes to a little island with a large dun on it 
surrounded by a white wall (mur gel), as white "as if it 
" had been built of burnt lime or carved out of one unbroken 

* See the article on Bawns, Ulster Journ. Archasol., v. 125 ; and a 
particular bawn described by Dr. J. P. Mahaffy, in the Athenasum of 
10th August, 1901. 

f See LU, 41, b, 26 and 29 : Hennessy in Rev. Celt., 11. 89 ; Silva Gad., 
!32, % See Kinahan, Geol., 358 ; and p. 357, below. 





" rock of chalk."* " The colour of the dun of the lady 
Crede "—says another story—" is like the colour of lime."f 
The great ramparts of Tara must have shone brilliantly 
over the surrounding plain : for it is called " White-sided 
Tara," in the " Circuit of Muirchertach mac Neill " : but 
this was a memory only, for when the poem was written, 
Tara had been deserted for centuries. 

Fig. 190. 

Dundrum Castle, ne»r Newcastle, County Down. Built at the end of the twelfth 
century by John de Courcy, on the very site of the old Irish fortress called Dun 
Rury, which covered the summit of the rock. The great earthworks belonging to 
the original dun still remain at the base of the rock at one side, but are not seen in 
this figure. (From Kilk. Archaeol. Journ. for 1883-4, P- 158.) 

The treatment of forts here is necessarily very brief. 
Those who wish to study the subject farther may consult 
Mr. Westropp's essay in Trans. Roy. Ir. Acad, on " The 
Ancient Forts of Ireland," in which the whole subject is 
examined for the first time scientifically and in considerable 
detail, and the similarity of the ancient Irish forts to those 
of Greece, Thessaly, Italy, France, Austria, Germany, and 
other parts of the Continent, is pointed out. 

In modern times, when the native knowledge of Irish 
history and antiquities had greatly degenerated, and the 
light of our own day had not. yet dawned, many writers 
attributed the ancient Irish raths and duns to the Danes, 
* Joyce, Old Celtic Romances, 131. f O'Curry, Man. & Cust., 11. 13. 


so that it became the fashion to call them " Danish raths 
or forts '" : but this idea has been long since exploded, as 
the reader will have seen who has glanced through the 
preceding pages. The peasantry have the same notion : 
but their error arose from confounding the Dedannans or 
Dananns with the Danes, through similarity of sound.* 

The Anglo-Normans built stone castles in Ireland 
according to their fashion : and not unfrequently they 
selected the very site, or the very vicinity, of the old Irish 
fortresses : for an Anglo-Norman had at least as keen an 
eye for a good military position as an old Irish warrior. 
Accordingly the circumvallations of the ancient native 
forts still remain round the ruins of many of the Anglo- 
Norman castles ; as at Rahinnane in Kerry, Knockgraffon 
near Cahir in Tipperary, and Dundrum in County Down.f 
Several of those Anglo-Norman or Anglo-Irish castles are 
figured throughout this chapter. It is to be observed 
that the Irish began to abandon their earthen forts and 
build stone castles — many of them round like the older 
earthen forts and cahirs — shortly before the arrival of 
the Anglo-Normans in 1169 : but this was probably in 
imitation of their warlike neighbours.]: 

Crannoges. — For greater security, dwellings were often 
constructed on artificial islands made with stakes, trees, 
and bushes, covered with earth and stones in shallow lakes, 
or on small flat natural islands if they answered. These 
were called by the name crannog [crannoge], a word de- 
rived from crann, ' a tree,' as they were constructed almost 
entirely of wood. Communication with the shore was 
carried on by means of a small boat, commonly dug out 
of one tree-trunk. At night, and at other times when 
precaution was necessary, it was kept in a boat-house on 
the island. But in ordinary times, for the convenience 

* On this see Stokes's Petrie, 218 : and Wilde, Boyne, 70. 
f See Kilk. Arch. Journ., 1854-5, pp. 394-7. 
% Stokes's Petrie 212 et seq. 





of the residents and visitors coming and going, it was 
usually left floating in the lake-channel, with a cable 
from boat to island and another from boat to mainland, 
so that whether arriving or departing the person could 
pull it towards him* Usually one family only, with 
their attendants, lived on a crannoge island ; but some- 
times several families, each having a separate wooden 
house. Where a lake was well suited for it — pretty large 
and shallow — several crannoge islands were formed, each 
with one or more families, so as to form a kind of little 
crannoge village. 

Fie. 191. 

Section of Crannoge in Ardakillen Lough, near Strokestown, Co. Roscommon. Gives a good 
idea of the mode of constructing these little islands. The three horizontal lines at top show the 
level of the water according to season. (From Wilde's Catalogue, p. 226.) 

Crannoge dwellings were in use from the most remote 
prehistoric times, as is clearly shown by the remains found 
in them, implements of various kinds, which belonged to 
primitive ages. They are very often noticed, both by 
native Irish and by English writers, and they continued in 
use down to the time of Elizabeth. They are referred to 
in the Tripartite Life by the name insola in gronna, ' an 
island in a bog.' Cambrensis describes them as he saw 
them in the twelfth century, though he does not use the 
name crannoge : — " These lakes [of Ireland] encompass 
" some slightly elevated spots, most delightfully situated, 
" which, for the sake of security, and because they are 
" inaccessible except by boats, the lords of the soil appro- 
" priate as their places of refuge and seats of residence."! 

*Bec Fola, 179. fTop. Hib., I. vii. 




Great numbers of crannoges have of late years been 
explored, and the articles found in them show that they 
were occupied by many generations of residents. In most 
of them rude " dug-out " boats have been found, many 
specimens of which are preserved in the National Museum, 
Dublin, and elsewhere. In some cases the original crannoge 
dwelling was, in later ages, replaced by a stone-and-mortar 
castle, of which the finest existing example is the Hag's 
Castle in Lough Mask in Mayo. This is circular like 

Fig. 192. 

A Crannoge Village in Lough Eyes, near Lisbellaw, County Fermanagh. The little artificial 
islands are there still, but the wooden dwelling-houses are all gone : and this is an attempted resto- 
ration, by Mr. Wakeman, of the appearance of the whole group when a house stood on each. 
(From Colonel Wood-Martin's Traces of the Elder Faiths, I. 223.) 

the original structure, occupying almost the whole of the 
little island : and it is perhaps the earliest stone-and= 
mortar castle erected in Ireland before the Anglo-Norman 
invasion* Lake-dwellings similar to the Irish crannoges 
were in use in early times all over Europe, and explorers 
have examined many of them, especially in Switzerland.! 

* See Wilde, Lough Corrib, 260: and Kilk. Arch. Journ., 1872-3, p. II. 

t Numerous descriptions of individual crannoges and of their exploration 
will be found in the. Proc. Roy. Ir. Acad., and in the Kilk. Arch. Journ. 
and its continuation, the Journ. of the Roy. Soc. Antiqq., Irel. Easily 
found out by glancing through the Indexes. 


The word ' crann6ge ' was also used by the ancient Irish 
to designate a small wooden vessel of some sort. In the 
" Colloquy of the Ancients," Feredach, king of Ossory, is 
spoken of as using gold and silver for decorating such 
things as drinking-horns, crannoges, swords, chessboards, 
and chessmen* In later times the ' crann6ge ' was familiar 
among the Anglo- Irish — a sort of basket of a certain 
size used as a measure for corn.f In the Senchus Mor 
* crannog ' is used simply to denote a rod.J 

4. Domestic Vessels. 

The material in most general use for vessels was wood ; 
but there were vessels of gold, silver, bronze, and brass, all 
of which however were expensive. Occasionally we read 

of iron being used : 
among the treasures 
possessed by Ailill 
and Maive, as we find 
stated in the Tain, 

Fig. 193. Fig. 194- • 7 , 

^.,. ,, , ., were larn-lestair, 

Figure 193, Stone Dnnking-cup, tM inches wide across 
thebowj. Found, buried deep, in the bed of the Shannon.' 'irotl VCSSels '§ There 

Figure 194, Stone Cup. (Wilde's Catal.. p. 114) ' " 

were also vessels of 
stone : but these were not much in use. A stone bottle, 
of the kind hitherto known only in Egypt, has been found 
in the big rath near Lucan|| : and two stone drinking-cups 
are figured here. Drinking-goblets of glass have been 
already noticed ; and leather vessels for holding liquids 
will be described in chap, xxvi., sect. 5. 

* Silva Gad., 416. t Ware, Antiqq. 223. 

% Brehon Laws, I. 152, 29 ; 153, 9 from bot. 

§ O'Curry, Man. & Oust., II. 89 : LL, 54, a, 33. 

|| Stokes's Petrie, 420: see Wilde, Catal., 114. Petrie and his biographer, 
Dr. William Stokes, seem to have regarded this as an Irish imitation of an 
Egyptian bottle. Many years later Miss Margaret Stokes calls it an 
" Egyptian alabaster bottle," and looks upon it as brought hither by some 
of those Egyptian monks mentioned in vol. I., p. 413, supra. See Miss 
Stokes, Inscr., II. 137. 



6 9 

For making wooden vessels beech was oftenest em- 
ployed : but the best were made of yew. In one of 
O'Curry's Lectures,* he gives, from an old ms., a curious 
list of vessels, all made from the trunk of one immense 
yew-tree. A large proportion of the timber vessels used 
were made of staves bound by hoops, like those in use 
at present, indicating skill and accuracy in planing 
and jointing. This is proved by many passages. St. 
Finnchua's mother during her pregnancy — according to 

FIG? 195. 

Carrickfergus Castle in 1840. On a rock over the sea. Built originally by John 

de Courcy in the end of the twelfth century, on the site of an older Irish fortress, but 

greatly enlarged and altered after his time. One of these Anglo-Norman castles 

referred to at p. 65, supra. (From Wright's Ireland Illustrated.) 

the legendf — once longed for a drink of ale, and asked 
the brewers for a little : but the churlish fellows refused. 
She went away : but scarcely had she turned her back 
when the hoops (circatta : sing, circall, a ' circle ' or ' hoop ' : 
Lat. circulus) slipped off the vats, and the ale all ran 
about. In the list of yew-tree vessels noticed above, 
several are mentioned as having grown so old that the 
hoops at last fell off. There was also a native term for 
a hoop — fonnsa : the Brehon Law (v. 483) enumerates 
the material for fonnsa or hoops — i.e. plantations of the 

* Man. & Cust., 11. 61. 

f Stokes, Lives of SS., 85, 232. 


proper timber, such as sallow-trees — as forming part of 
the " Commons " property of a territory. 

A large open hooped tub or vat was called by several 
names. One was dabach or dabhach [dauvagh], of which 
the derivation in Cormac's Glossary (p. 52) shows that the 
vessel was a two-handled tub like that of the present 
day : — " Dabach, derived from de-oach, two o's or ' ears,' 
" meaning two handles upon it : for at first there used to 
" be no handles on vessels." It may be remarked that the 
side-handle of a vessel was often called an or ' ear,' of 
which the word oach is a modification. In O'Clery's 
Glossary it is stated that a coimde was the same as a 
dabach. Another name for this sort of vessel was lotar 
or lothar [loher] : " a trough wherein are kept braichles 
or grains left after brewing," says Cormac's Glossary 
(p. 105). A moderately-sized tub with two handles, called 
a drolmach, was used by women for bringing water. This 
word is still in use and pronounced drowlagh. 

There was a special drinking-vessel, originally made of 
yew (ibar), and thence called ibrach, or in modern spelling 
iubhrach [yooragh]. This was until lately in use in Mayo, 
and called by its old name : it was deep, and grew narrow 
from bottom to top. 

The people used a sort of pitcher or hand-vessel called 
a cilorn [keelorn], having a stuag or circular handle in its 
side, from which it was also called stuagach, i.e. ' circle- 
handled ' : sometimes called milan-duirn and metair-duirn, 
i.e. ' hand- vessel ' ; for milan and metair (or medar) both 
mean ' a small vessel ' : and dom, gen. duirn is a ' hand ' 
or ' fist.' In Zeuss (p. 41, 26) cilornn glosses urceus, ' a 
pitcher.' In the Coir Anmann we read that a certain 
Lugaid went with a cilorn in his hand to bring water for 
drinking at dinner.* Milan glosses urna in the Tract on 
Lat. Decl. (No. 138) ; and it must have been intermediate 
in size between the cilorn and the medar : for in the list of 

* Ir. Texte, 111. 319. 

CHAP. Xt] 

The HotfsE 


yew vessels mentioned above (p. 69) it is stated that when 
the original cilom became worn out from age, the owner 
made a milan out of what was left : and when this milan 
again became decayed and worn, there was only as much 
sound wood left as made a medar. In the Voyage of 
Maildune it is related how a certain man gathered up a 
great many valuable articles, and among them a number 
of brazen cilorns* This ancient term is still preserved in 
the south of Ireland, where it passes quite current as an 
English word, in the form of keeler, though the people 
apply it now to a vessel of a different shape and with a 
different use. A broad, shallow tub about 18 inches across 
and 6 or 8 inches deep, and having two handles formed by 

Fig. 196. 

Fig. 197. 

Two bronze Drinking-vessels in National Museum. Figure 196 is 1% inches wide : hammered 
out and shaped with great skill from one single thin piece of metal : found in a crannoge in 
County Roscommon. Figure 197, oval, sH inches in the long diameter. Handle decorated, 
and terminating at top in an animal's head. Found near Keshcarrigan, County Leitrim. (Both 
from Wilde's Catalogue, pp. 533, 534.) 

the projection upwards of two of the staves, is universally 
called a ' keeler.' Milk is ' set ' in a keeler for a night to 
throw the cream to the top for churning. 

A com [curn] or horn was a drinking-vessel, usually 
made from a bullock's horn, hollowed out and often highly 
ornamented with metal-work and gems. A com mounted 
with silver was sometimes called a fethal (Corm. 80). 
This word com seems to be borrowed from Latin comua : 
but there was a native name also, viz. buabaill [boovill], 
from bo or bu, ' a cow ' : and another, adarc [ey-ark], which 
is now the common Irish word for ' horn ' : but buabaill may 
be a borrowed . word. Conn the Hundred Fighter, on a 

* Rev. Celt., X. 83. 





certain occasion during the feis of Tara, stood up from 
where he sat on his throne, and with a polished buabaill in 
his hand, spoke to the assembled nobles.* That a corn is 
the same thing as a buabaill is proved by this : — that while 
in one version of the Boroma it is stated that certain 
messengers, arriving at the palace of Ailech, found king 
Aed mac Ainmirech drinking mead from a com, in another 
version the self-same drinking-horn is 
called a corn- buabaill.^ Drinking-^my 
were made at home from cows' or bul- 
locks' horns ; but very large ones were 
imported and much valued. Among the 
" foreign " valuables mentioned in the 
Brehon Law, the glossator enumerates 
cuirn-buabaill {cuirn, pi. of com) : 
these no doubt were genuine 
buffalo-horns, as is perhaps indi- 
cated by the Irish word 
buabaill. % 

These corns were 
sometimes , given as a 
Fio, 198. part of the stipend due 

The " Kavanagh Horn,'' drawn from an exact model f 1" O IT1 Olie kin°" tO 
in National Museum, Dublin : 22 inches along the convex 

or under side. On a brass plate round the top is this another aS We filld bv 

inscription : — "TlGERN ANUS O'LAUAN ME FECIT ' 

DEO Gracias. I. H.S.": which gives the name of the manV entries in tile 
artist, Tiernan O'Lavan. This is not a very old speci- * 

men. (From Wildes Catalogue, p. 266.) Book of Rights, where 

they are often called 
curved corns from their shape. Sometimes they were 
coloured : part of the stipend or taurcrec due to the king 
of Offaly from the king of Ireland was four corns " of 
various colours " (Bk. of R., 253). According to the 
bardic history, ornamented drinking-horns covered with 
gold and silver were first introduced into Ireland by king 
Tigernmas, many centuries before the Christian era, a 

Silva Gad., 143 : Ir. Texte, 131, i . t Rev. Celt., XIII. 63. 

\ Brehon Laws, V. 220, 23 ; 221, 20 . 




record which, though, legendary, indicates the general 
costliness of the workmanship. Some of these corns are 
preserved in our museums, of which one is figured on 
page 72. 

The escra was a drinking-goblet : Cormac's Glossary 
(p. 69, twice) says it was a copper vessel for distributing 

Fig. 199. 

Ancient Irish vessel, 15 inches high, and 15 inches in width at the mouth : made 
out of a single piece of oak, except the bottom, which is of alder. The carving on 
the side is the Opus Hibernicwn or interlaced work. It had a lid when found, 
similarly carved ; but this has been lost. The whole outer surface was originally 
painted in a kind of dark enamel, portions of which still remain. This very ancient 
vessel was found five feet deep in a bog in County Fermanagh. (From Kilk. 
Archaeol. Journ. for 1879-82, p. 98.) 

water ; but it was sometimes made of silver. The sons 
of O'Corra, in the course of their voyage, landed on an 
island, where a lady came towards them having in one 
hand a copper cilorn full of food like cheese, and in the 
other a silver escra. And she gave them the food to eat: 
and she brought them water in the escra from a well on 
the strand : " and there was no delicious flavour that was 



fPARt lit 

"ever tasted by man that they did not find in this food 
" and drink."* In the Life of St. Darerca it is stated 
that the cscra was a silver drinking-cup from which great 
people were wont to drink.f 

The word lestar was applied to vessels of various kinds, 
among others to drinking-vessels : it was often used as a 
generic term for vessels of all kinds, including ships. In 
the Life of St. Brigit it is related that on one occasion the 
king of Teffia was drinking out of a lestar covered with 
gems, when a careless man took it 
from his hand and let it drop so 
that it was broken into bits.* The 
beautiful lestar represented in fig. 
199 was found some years ago, as 
stated in the descriptive note, but 
what special name was applied to 
it we cannot tell. There was a 
drinking-cup of some kind called 

h/ T 1 fy* indtile which Cormac (p. 58) defines 
KX^M I J " a little lestar or vessel in which 

drink fits," i.e. for holding drink : 
which agrees with O'Reilly's "inntille, 
a drinking-cup." 
The simple word cua, and its derivatives cuad and 
cuach, all mean ' a cup.' In the gloss on the Senchus 
Mor we are told that a folderb (see below) is a cua in 
the shape of a bell {cua cluic)% : and the Crith Gabhlach 
speaks of a cuad 12 inches high for drinking milk out of.|| 
Cuach, which is the common word for ' cup,' is retained in 
Scotland to this day and used as an English word, in the 
forms of quaigh and cogue, for a drinking-cup. It was 
prophesied for Finn mac Cumail by his wife that on 

*Rev. Celt., XIV. 47: Joyce, Old Celtic Romances, 415. This food 
like cheese, containing every delicious flavour, is a stock incident in ancient 
Welsh, as well as in Irish, tales. 

t Three Fragments, 9, i. § Brehon Law, I. 134, 5. 

t Stokes, Three Irish Homilies, 73. || Ibid., IV. 306, 9 . 

Fig. 200. 
Grotesque figure of a man 
drinking : from the Book of 
Kells (seventh century 1. (From 
Wilde's Catalogue, p. 299.) 

CHAP. Xx] 



whatsoever day he should drink from a horn [adarc\ he 
would die. Accordingly he took good care always to 
drink from a cuach. But one day in his old age, being 
overcome with thirst, and not having his cuach, he drank 
from an adarc : and on that same day he was killed on 
the Boyne* Ian, gen. ena, means ' a vessel ' : it is often 
applied to a small drinking-mug. The Crith Gabhlach 
mentions a vessel, ian-oil [drinking-zVm], three hands high, 
used for drinking milk.f Cormac's Glossary (p. 34) de- 
scribes a sort of drinking-goblet called a cingit, in such a 
way that we may infer it was slender in the middle and 
opened out at the top 
and bottom, so that 
the two halves were 
alike, or nearly so, in 
shape and size, and 
were probably intended 
to be reversible. 

The usual drinking- 
vessel among the com- 
mon people, especially 

at meals and drinking-bouts, was a medar or mether (so 
called from the drink called mead), made of wood, with two 
or four handles : it circulated from hand to hand, each 
passing it to his neighbour after taking a drink. Many of 
these methers are preserved in museums, of which two are 
figured here.J People drank from the corners. A sort of 
hamper or vessel called a ritsc [roosk], made of bark-strips 
on a wicker-work frame, was much used in farmhouses.§ 

A churn was known by several names — among others 
cuinneog, which is the present name. In the Senchus 

Fig. 201. Fig. 202. 

Wooden Methers. From Wilde's Catalogue, p. 214. 

* Silva Gad., 98. t Brehon Laws, IV. 302, 23. 

% See Dub. Pen. Journ., 1. 300: Kilk. Arch. Journ., 1860-61, p. 54: 
and Wilde, Catal., 21*4. 

§ Stokes, Lives of SS., line 1277, and p. 320 bot. Also Three Irish 
Homilies, 62,11: and "Ruse" in Windisch's Worterbuch. 


Mor it is called comm ; and in the gloss, this again is 
interpreted by three names: — ctiairt (' round -vessel '), 
belcumaug ('narrow-mouth'), and muide* showing that 
it was something Jike the hand-churn still in use. Derb is 
another name for a churn, according to O'Clery's Glossary, 
which explains this word as meaning either a churn {cuin- 
nedg) or a cup {balidn). A saying quoted in the same 
glossary, " my ear to the ear of the derb" shows that the 
ancient churn had two ears or handles, which the modern 
hand-churn has not. The compound foilderb was also used ; 
but from the words of the gloss in the Senchus Mor, it 
would seem to denote here an ordinary drinking-cup : — 

E" Folderb which has a ring or handle 
{foil or fair) out of its side, and 
it is a bell -shaped cua."f Another 
compound of derb is given in 
Cormac's Glossary (p. 58) to denote 
FiG.203. a churn — derb-loma, i.e. the derb of 

Pan or bucket, made o.,t' 3 f o..e 16 im or milk. From all the preced- 

piece of red deal : i foot long, 6% 

inches deep, and iotf inches broad, jng we may infer that derb and its 

Cover made of yew, pressed into 

shape when softened (see •• can- compounds were used to denote 

tair," ' a press,' in Index). Now in 

National Museum. (From wildes both a drinking-cup and a churn. 

Catalogue, p. 213 ; A 

The form of churn used among the 
ancient Irish was that in which the cream or milk is 
agitated by a dash worked with the hand. The churn- 
dash was — and is still — called lotmd, and sometimes 
loimdha, from loi?n, ' milk.+ 

For bringing home milk from the milking-place, Adam- 
nan (pp. 126, 445) mentions a wooden vessel of such a 
make that it could be strapped on the back. The lid was 
kept in its place by a wooden cross-bar {gercenn) which 
ran through two holes at opposite sides near the rim. 
Adamnan tells a story of a young member of the Iona 
community bringing home on his back a vessel full of milk 

* Brehon Laws, I. 124, 4; and 134, 2, 3 . t /bid., 124,5. 

X O' Curry, Man. & Cust., 1. 133. 




into which a demon had entered : and when St. Columba, 
blessing the milk according to his custom, made the sign 
of the cross, it became agitated, and the bar which fastened 
the lid was driven through the two holes and shot away to 
a distance. Adamnan, writing in Latin, uses the original 
Irish word for this bar in the form ger genua. The word is 
elsewhere explained as a slender bar which passes through 
two openings and fastens the lid. In " Mac Conglinne's 
Vision" (p. 87, 1.9) it is used for an ordinary door-bolt — 
which seems its primary meaning. It appears that this 
term is not found in any other Latin document besides 
Adamnan's : and it is probably an original Gaulish or 
Celtic word.* 

Fig. 204. 

Natural Boulder-Stone : height 3 feet, with three artificial balldns. (From Col. Wood-Martin's 
Pagan Ireland, p. 411. Drawn by Wakcman.) 

It will be seen from what precedes that there was in 
old times in Ireland quite as great a variety of vessels of 
all kinds, with distinct names, as there is among the 
people of the present day ; and there are, besides, other 
names not yet noticed. The cup that St. Patrick was 
drinking out of at Tara, when the druid attempted to 
poison him, is called ardig in the Tripartite Life (p. 54, 7) : 
ardig or airdig being a common old word for a drinking- 
goblet. A balldn seems to have been a simple, cheap, 
wooden drinking-cup in very general use : in one place, 

* Sec Kuno Meyer in Rev. Celt., xill. 506. 




Cormac's Glossary (p. 25) defines it as <: a poor man's 
vessel " : and elsewhere (p. 27) as a vessel used by lepers. 
Keating applies it to a drink ing-cup, and it was sometimes 
also applied to a milk-pail.* In Connaught it is used to 
designate round holes in rocks usually filled with water : 
which use modern antiquarians have borrowed, and they 
now apply " ballaun " to those small cup-like hollows, 
generally artificial, often found in rocks, and almost 
always containing water.t 

Esconn, escand, or escann is 
described in Cormac's Glossary 
as a vessel for distributing water, 
derived from esc, 'water,' and 
"cann, the name of a vessel." 
This last phrase is interesting 
as showing the existence in 
ancient Gaelic of a term for a 
drinking -vessel identical with 
the English word can. There 
was a pail or vessel of some 
kind called sitheal, which was 
sometimes made of silver. 

The word cernhi [kerneen] is 
given in Cormac's Glossary (p. 37) 
as meaning mi'ass, i.e. a dish on 
which food is placed at table ; in which sense it is also found 
in an ancient satire given in the Book of the Dun Cow, 
said to have been the first satire ever composed in Ireland. 
Cernin is a diminutive of the simple word cern — modern 
form cearn — which is used to denote a dish of any kind, 
for measuring commodities, such as grain: Peter O'Connell 
explains it " a certain dry measure " : and Keating has the 
expression cearn-arbhair ; a ' cern of corn.' Bleid or bleide 
was the name of a goblet or vessel of some kind, mentioned 
both in the Brehon Laws and in the Tales. The word 

*Corm. Gloss., 54 ("Del "). t Ibid., 25. 

Fig. 205. 

Earthenware glazed Pitcher. 13 
inches high. Found in a crannoge 
in County Down. (From Wilde's 
Catalogue, p. 158.) 


miass or mias, given above from Cormac's Glossary, is 
very commonly used for a platter or dinner dish.* Coire, 
' a caldron ' ; cusal ; criol ; and some other terms, as well 
as the vessels they denote, will be dealt with elsewhere in 
this book. Most of those named in this section will be 
found mentioned in vol. v. of the Brehon Laws, p. 407, and 
the following pages. 

Earthen vessels of various shapes and sizes were in 
constant use. They were made either on a potter's wheel, 
or on a mould, or on both. This appears from a curious 
commentary on the Latin text of a passage in the Psalms 
(11. 9), written in the Irish language by an Irishman, in the 
eighth or ninth century, contained in a manuscript now in 
Milan. This old writer, evidently taking his illustration 
from his native country, explains " a potter's wheel " as " a 
" round wheel on which the potters [Irish na cerda, ' the 
" cairds or artisans '] make the vessels, or a round piece of 
" wood about which they [the vessels] are while being 
" made."f The " round piece of wood " was the block or 
mould on which they were first formed roughly, to be 
afterwards perfected on the wheel. 

5. Royal Residences. 

Almost all the ancient residences of the over-kings of 
Ireland, as well as those of the provincial and minor kings, 
are known at the present day ; and in most of them the 
circular ramparts and mounds are still to be seen, more 
or less dilapidated after the long lapse of time. As 
there were many kings of the several grades, and as each 
was obliged to have three suitable houses (vol. I., p. 58), 
the royal residences were numerous ; of which the most 
important will be noticed here.J In addition to these, 

* See the story of B6thar-na Mias in Joyce, Ir. Names of Places, n. 191. 
f Stokes and Strachan, Thesaurus, 1., p. 23. 

% The present appearance of several of the royal residences will be found 
described in Mr. T. O. Russell's " Beauties and Antiquities of Ireland." 


several of the great strongholds described in vol. I., pp. 84 
to 90, were royal residences. 

Tara. — The remains of Tara* stand on the summit 
and down the sides of a gently-sloping, round, grassy hill, 
rising 500 feet over the sea, or about 200 over the sur- 
rounding plain, situated six miles south-east of Navan, in 
Meath, and two miles from the Midland Railway Station 
of Kilmessan. It was in ancient times universally regarded 
as the capital of all Ireland, or, as Muirchu, in his seventh- 
century Life of St. Patrick calls it, caput Scotorum, the 
' capital of the Scots ' ; so that in building palaces elsewhere 
it was usual to construct their principal houses and halls 
in imitation of those of Tara. It was the residence of the 
supreme kings of Ireland from prehistoric times down to 
the sixth century, when — as already mentioned — it was 
deserted in the time of King Dermot the son of Fergus 
Cervall on account of St. Ruadan's curse. Although it 
has been abandoned to decay and ruin for thirteen 
centuries, it still presents striking vestiges of its ancient 

Preserved in the Book of Leinster and other ancient 
manuscripts there are two detailed Irish descriptions of 
Tara, one written in the tenth century by Kineth O'Harti- 
gan, and the other in the eleventh by Cuan O'Lochain (for 
whom see vol. 1., p. 462, supra). Both these distinguished 
men examined the remains personally, and described them 
as they saw them, after four or five centuries of ruin, giving 
the names, positions, and bearings of the several features 
with great exactness. These two interesting documents 
are published with translations and learned annotations in 
Petrie's essay on Tara. More than sixty years ago Dr. 
Petrie and Dr. O'Donovan made a most careful detailed 

* Old Irish name Temair (modern Teamhair), gen. Temrach, dat. 
Temraig, which is represented by the present name " Tara." For more 
about this name, and for other places of the same name, see Joyce, Ir. 
Names of Places, 1., p. 294. 

fig. 200. 

Flan ol Tara, as it exists at the present day. (From the two plans given by Petrie in his Essay on Tan. 


examination of the hill and its monuments ; and with 
the aid of those two old topographical treatises they 
were able, without much difficulty, to identify most of the 
chief forts and other remains, and to restore their ancient 
names. The following are the most important features 
still existing, and they are all perfectly easy to recognise 
by any one who walks over the hill with the plan given here 
in his hand. A much more detailed description of them, 
with their full history and all extracts relating to them 
from Irish manuscripts, is given in Petrie's essay on Tara, 
from which most of the account given here has been con- 
densed. It is to be borne in mind that the forts now to be 
seen were the ramparts or defences surrounding and pro- 
tecting the houses. The houses themselves, as has been 
already explained (p. 55), were of wood, and have of 
course all disappeared. 

The principal fortification is Rath Righ [Rath-Ree] or 
Rath-na-Righ, the ' fort of the kings,' also called Caher 
Crofitw, an oval occupying the summit and southern slope 
of the hill, measuring 853 feet in its long diameter. The 
circumvallation can still be traced all round ; and judging 
from the existing remains, it consisted originally of two 
walls or parapets with a ditch between. Moreover it is 
pretty certain that one at least of these two ramparts was 
of stone, as the " caher " (Irish cathair) in the name 
" Caher Crofinn " would indicate ; and as a matter of fact 
the stones still remain for about a fifth of the whole circuit. 
This seems to have been the original fort erected by the 
first occupiers of the hill and the most ancient of all the 
monuments of Tara. 

Within the enclosure of Rath Righ are two large 
mounds, the Forrad or Forradh [Forra] and Tech Cormaic, 
beside each other, and having portions of their ramparts in 
common. The Forradh has two outer rings or ramparts 
and two ditches : its extreme outer diameter is nearly 300 
feet. The name " Forradh " signifies ' a place of public 




meeting,' and also a judgment-seat, cognate with Lat. 
forum ; so that it seems obvious that this is the structure 
referred to by the writer of the ancient Norse work called 
" Kongs Skuggsjo " or ' mirror for kings,' already referred 
to (vol. 1., pp. 226, 300). This old writer, speaking of 
Tara says : — " And in what was considered the highest 
" point of the city the king had a fair and well-built castle, 
" and in that castle he had a hall fair and spacious, and in 
" that hall he was wont to sit in judgment." 

On the top of the Forradh there now stands a remark- 
able pillar-stone six feet high (with six feet more in the 
earth), which Petrie believed was the Lia Fail, the inaugu- 
ration-stone of the Irish over-kings, the stone that roared 

Fig. 207. 
The Mound called the Forradh, at Tara. (From Mrs. Hall's Ireland. Drawn by Wakeman.! 

when a king of the true Milesian race stood on it (see 
vol. 1., p. 45) ; but recent inquiries have thrown grave 
doubts on the accuracy of this opinion. 

Tech Cormaic (' Cormac's house ') was so called from 
the illustrious King Cormac mac Art, who reigned a.d. 
254 to 277. It is a circular rath consisting of a well-marked 
outer ring or circumvallation, with a ditch between it and 
the inner space : the extreme external diameter being 
244 feet. We may probably assign its erection to King 
Cormac, which fixes its age. 

Duma nan Giall or the ' mound of the hostages,' 
situated just ins'ide the ring of Rath Righ, is a circular 
earthen mound, 13 feet high, 66 feet in diameter at the 


base, with a flat top, 25 feet in diameter. The timber 
house in which the hostages lived, as already mentioned 
(vol. 1., p. 54), stood on this. 

A little to the west of the Mound of the Hostages stands 
another mound called Duma na Bo (the ' mound of the 
cow '), about 40 feet in diameter and 6 feet high. It was 
also called Glas Temrach (the ' Glas of Tara ') which would 
seem to indicate that the celebrated legendary cow called 
Glas Gavlin, which belonged to the Dedanann smith 
Goibniu,* was believed to have been buried under this 

About 100 paces from Rath Righ on the north-east is 
the well called Nemnach (' bright ' or ' sparkling ') so cele- 
brated in the legend of Cormac's mill — the first mill erected 
in Ireland, for which see chap. xxv. (page 330), below. A 
little stream called Nith (' shining ') formerly ran from it, 
which at some distance from the source turned the mill. 
The well is now nearly dried up ; but it could be easily 

Rath na Seanaid (the ' rath of the synods '), now popu- 
larly called " the King's Chair," has been partly encroached 
upon by the wall of the modern church : the two ramparts 
that surrounded it are still well-marked features. Within 
the large enclosure are two mounds, 106 and 33 feet in 
d iameter respectively. Three Christian synods are recorded 
as having been held here, from which it had its name : — 
one by St. Patrick on the occasion when he preached to 
King Laegaire and his nobles at Easter, a.d. 433 ; one by 
St. Ruadan or Rodanus when he pronounced the curse 
that caused Tara to be abandoned (for which see page 
437, below) ; and the last by Adamnan, probably in 
the year 697, in which he procured acceptance for the law 
exempting women from taking part in battles (see vol. I., 
p. 96, supra). 

* For Goibniu, see vol. 1. 261, supra : and for the wonderful cow, Glas 
Gavlin, see Joyce, Ir. Names of Places, 1. 163 : and FM, 1., p. 18, note s. 


Near the Rath of the Synods, and within the enclosure 
of the modern church, stood Adamnan s Cross, of which the 
shaft still remains, with a human figure rudely sculptured 
in relief on its side. A little to the south-east of this cross 
was situated the house which — as already related, vol. I. 
p. 307 — was burned round young Benen and the druid 
Lucet Mail, when Benen escaped and the druid was 
reduced to ashes. 

On the northern slope of the hill are the remains of the 
Banqueting-Hall, the only structure in Tara not round or 
oval. It consists of two parallel mounds, the remnants of 
the side walls of the old Hall, which, as it now stands, is 
759 feet long by 46 feet wide ; but it was originally both 
longer and broader. It is described in the old documents 
as having twelve (or fourteen) doors : and this description 
is fully corroborated by the present appearance of the ruin, 
in which six door-openings are clearly marked in each side 
wall. Probably there was also a door at each end : but all 
traces of these are gone. 

The whole site of the Hall was occupied by a great 
timber building, 45 feet high or more, ornamented, carved, 
and painted in colours. Within this the Feis or Convention 
of Tara held its meetings, which will be found described in 
chap. xxix. (p. 436), farther on. Here also were held the 
banquets from which the Hall was named Tech Midchuarta, 
the ' mead-circling house ' ; and there was an elaborate 
subdivision of the inner space, with the compartments 
railed or partitioned off, to accommodate the guests 
according to rank and dignity. For, as will be seen in 
chap. xxi. (p. 105), they were very particular in seating the 
great company in the exact order of dignity and priority. 
From this Hall moreover, the banqueting-halls of other 
great houses commonly received the name of Tech 

Rath Caelchon was so called from a Munster chief 
named Caelchu (gen. Caelchon), who was contemporary 


with Cormac mac Art, third century. He died in Tara, 
and was interred in a leacht or earn, beside which was 
raised the rath in commemoration of him. The rath is 
220 feet in diameter ; and the very earn of stones heaped 
over the grave still remains on the north-east margin of 
the rath. 

Rath Grdinne is a high well-marked rath, 258 feet in 
diameter. It received its name from the lady Grainne 
[Graunya : 2-syll.], daughter of King Cormac mac Art, and 
betrothed wife of Finn mac Cumail. She eloped with 
Dermot O'Dyna, and the whole episode is told in detail in 
the historic romance called ' 'The Pursuit of Dermot and 
Grainne."* This mound, and also the smaller mound 
beside it on the south called the Fothad of Rath Grdinne, 
are now much hidden by trees. 

A little north-west of the north end of the Banqueting- 
Hall, and occupying the space north of Rath Grainne and 
Rath Caelchon, was the sheskin or marsh of Tara, which 
was drained and dried up only a few years before Petrie's 
time : but the well which supplied it, Tober Finn (Finn's 
well), still remains. 

Rath Laegaire [Rath Laery], situated south of Rath 
Righ, was so called from Laegaire, king of Ireland in 
St. Patrick's time, by whom, no doubt, it was erected. 
It is about 300 feet in diameter, and was surrounded by 
two great rings or ramparts, of which one is still very well 
marked, and the other can be partially traced. Laegaire 
was buried in the south-east rampart of this rath, fully 
armed and standing up in the grave, with his face, towards 
the south as if fighting against his enemies, the Leinster 
men. The whole account of his interment will be found 
at page 551, farther on. 

West of Rath Righ was the well called Laegh [Lay], a 
name signifying ' calf ' : it is now dried up, though the 
ground still remains moist. In this well, according to 

* This fine story will be found in Joyce's Old Celtic Romances, 


the seventh-century Annotations of Tirechan, St. Patrick 
baptised his first convert at Tara, Ere the son of Dego 
who afterwards became bishop of Slane, and who is com- 
memorated in the little hermitage still to be seen beside 
the Boyne (vol. I., p. 320). This well is called Laeg by 
Cuan O'Lochain ; and Tirechan calls it Loig-les, which he 
translates vitultis civitaium, the ' calf of the lisses or cities.' 
Probably there was some legend accounting for this very 
old name of Laeg or ' calf,' but it is not now known. 

The five main sliges [slees], or roads, leading from 
Tara in five different directions through Ireland, will be 
found described at pp. 393 and 395, below. Of these, 
portions of three are still traceable on the hill. The 
modern road traverses and covers for some distance the 
sites of two of them, Slige Dala and Slige Midluachra, as 
seen on the plan : Slige A sail still remains, and is sometimes 
turned to use. 

In one of the ancient poetical accounts quoted by 
Petrie (Tara, 147, top), it is stated that the houses of the 
general body of people who lived near Tara were scattered 
on the slope and over the plain east of the hill. 

In connexion with Tara, two other great circular forts 
ought to be mentioned. A mile south of Rath Righ lies 
Rath Maive, which is very large — 673 feet in diameter ; it 
forms a striking object as seen from the hill, and is well 
worth examining. It was erected, according to one 
account, by Queen Maive, wife of Art the solitary, the 
father of King Cormac mac Art, which would fix the period 
of its erection as the beginning of the third century. This 
lady, observe, was different from Queen Maive of Croghan. 

The other fort is Rathmiles, 300 feet in diameter, lying 
one mile north of the Banqueting-Hall : but nothing is 
known of its histoiy. 

After the abandonment of Tara the kings of Ireland 
took up their abode where they pleased, each commonly 
in one of his other residences, within his own province or 


immediate territory. One of these seats was Dun-na-Sciath 
(the ' Fort of the Shields ' : pron. Doon-na-Skee), of which 
the circular fort still remains on the western shore of Lough 
Ennell in County Westmeath (MS. Mat., 24). Another 
was at Rath near the western shore of Loch Leibhinn (now 
Lough Lene in Westmeath), two miles from the present 
town of Castlepollard. This residence was occupied for a 
time by the Danish tyrant Turgesius, so that the fort, which 
is one of the finest in the country, is now known as Dun- 
Torgeis or Turgesius's fort ; while the Old Irish name has 
been lost (Petrie, Tara, 128). In the time of St. Fechin, 
Dermot, one of the joint kings of Ireland (a.d. 656-664), 
had a residence in an island in Lough Lene, which, accord- 
ing to a local tradition, was also occupied for a time by 
Turgesius.* The tradition is probably correct, for the 
island is now known by the name of Turgesius's Island. 

Cenannus. — In the second century Conn the Hundred 
Fighter, while yet roydamna, before he became king of 
Ireland, resided in his stronghold at Cenannus, now Kells, 
County Meath : and four centuries after his time Dermot 
(son of Fergus), king of Ireland (544-565), had a palace 
here, probably the very stronghold occupied by Conn. 

Fremainn. — The kings of Tara had two royal residences 
at two different places called Fremainn (LU, 129, b, 25). 
One of these is still well known, and retains its old name : 
now locally designated Frewen Hill, rising over the western 
side of Lough Owel in Westmeath, on the top of which the 
old fort still stands (see MS. Mat., 285). 

Raeriu. — There were two very ancient palaces at two 
places called by the same name Raeriu, both of which still 
retain the name slightly altered. One was at the place 
now called Reary-more near the village of Clonaslee in the 
north of Queen's County. The other at Mullagh-Reelan, 
five miles south-east of Athy in Kildare, where the old 

* See FM, 1., p. 501, note r : Three Fragments, 169, note c : and Rev, 
Celt., xii., 343. 


mound still remains near Kilkea Castle. In this anglicised 
name, the termination " Reelan " represents Raerenn, the 
genitive of the old name, by the usual change of r to /.* 

Maistiu (gen. form Maistenn). — A better-known royal 
homestead stood five miles nearly east from Ath-I (now 
called Athy), where on the summit of a low hill the large 
circular fort remains, and is now well known by the name 
of the " Rath of Mullamast," in which the last syllable 
represents the old name. (This is mentioned farther on, 
at pages 366, 442, 552.) 

It has been already stated (vol. 1., p. 38) that Tuathal 
the Legitimate, king of Ireland in the second century, built 
four palaces at Tara, Tailltenn, Ushnagh, and Tlachtga. The 
fort of Tlachtga still remains on the summit of the Hill of 
Ward near the village of Athboy in Meath. There were 
royal residences also at Dunseverick in Antrim, the ancient 
Dun-Sobairce ; at the Old Head of Kinsale in Cork, called 
in Irish Dun-mic-Patraic ; at Derry ; at Rathbeagh on the 
Nore, where the rath is still to be seen ; at Dun-Aenguis 
on Great Aran Island (for which see p. 57, supra) ; and 
on the site of the present Baily Lighthouse at Howth, 
where several of the defensive fosses of the old palace-fort 
of Dun-Criffan can still be traced. 

Emain. — Next to Tara in celebrity was the palace of 
Emain or Emain-Macha, or, as its name is Latinised, 
Emania. It was for 600 years the residence of the kings 
of Ulster, and attained its greatest glory in the first century 
of the Christian era, during the reign of Concobar (or 
Conor) MacNessa, king of Ulster. It was the centre 
round which clustered the romantic tales of the Red 
Branch Knights, f The most ancient-written Irish tradi- 
tions assign the foundation of this palace to Macha of the 

* As to these two mansions, see FM, 1., p. 38, note r ; and Book of 
Rights, pp. 210, 211. 

f For the Red Branch Knights, and the literature connected with 
them, see vol. 1., pp. 83 and 536, supra. 


Golden Hair, wife of Cimbaeth (Kimbay], king of Ireland 
three or four centuries before the Christian era. From that 
period it continued to be the residence of the Ulster kings 
till a.d. 335, when it was burned and destroyed by three 
brothers, cousins of Muredach, king of Ireland — Colla Uais, 
Colla Menn, and Colla Dachrich, commonly known as the 
Three Collas — after which it was abandoned to ruin. The 
imposing remains of this palace, consisting of a great 
mound surrounded by an immense circular rampart and 
fosse half obliterated, the whole structure covering about 
eleven English acres, lie two miles west of Armagh. Nay, 
the ruin retains to this day the old name " Emain " 
slightly disguised ; for it is familiarly called " The Navan 
Fort or Ring," in which " Navan " correctly represents the 
sound of 'n-Emain, i.e. the original name with the Irish 
article 'n prefixed. 

When the Red Branch Knights came to the palace 
each summer to be exercised in feats of arms, they were 
lodged in a great house near Emain, called the Craobh- 
Ruadh,. commonly Englished the ' Red Branch,' from 
which the whole body took their name. But according to 
an old glossary, ruadh here means not ' red,' which is its 
usual sense, but ' royal ' : so that Craobh-Ruadh should be 
translated ' royal branch ' : but the designation " Red 
Branch Knights " is now too well established to be dis- 
placed. The name of this house is also preserved : for 
" Creeveroe," which very well represents the sound of 
Craobh-Ruadh, is still the name of a townland near the 
Navan fort. So far as we can judge from old tales, the 
Craobh-Ruadh seems to have been altogether built of 
wood, with no earthen rampart round it, which explains 
why the present townland of Creeveroe contains no large 
fort like that of Emain.* (See also Allen, vol. I., p. 88, supra). 

* According to LL, as quoted by O'Curry, Man. & Cust., I. 332, there 
were, in the time of Concobar, three chief houses in Emain : — The craeb 
ritaid, or ' royal branch,' where the kings and chiefs feasted : the craeb 


Ailech or the Grianan of Ailech. — Another Ulster palace, 
quite as important as Emain, was Ailech, the ruins of which 
are situated in County Donegal, on the summit of a hill 
800 feet high, five miles north-west from Derry, command- 
ing a magnificent view of Lough Foyle and Lough Swilly 
with the surrounding country. It is a circular stone cashel 
of dry masonry, 77 feet in internal diameter, the wall about 
13 feet thick at the base, and on the outside sloping 
gradually inwards. This central citadel was surrounded 
at wide intervals by five concentric ramparts, three of 
which may still be traced, the whole area originally 
including many acres. According to the old tradition it 
was founded by the Dedannans, and continued to be a 
royal residence to the time of its destruction, sometimes of 
the king of Ulster, and sometimes of the king of Ireland. 
After the fourth century it was the recognised residence of 
the northern Hy Neill kings, down to the year 1101, when 
it was destroyed by the Munster King Murkertagh, in 
retaliation for the destruction of Kincora by the Ulstermen 
thirteen years before. After this it was abandoned ; and 
the kings of Ailech transferred their residence to Inis- 
Eanaigh — now called Inchenny — in the County Tyrone, 
near Strabane, where they probably resided till the arrival 
of the Anglo-Normans. For nearly eight centuries Ailech 
continued in a state of ruin, the wall being reduced to a 
height of about 6 feet : but during the years 1874-8, it was 
rebuilt — in the face of great difficulties — by Dr. Bernard of 
Derry, a man of culture, with antiquarian tastes, who, as 
far as he could, restored it to its original shape. The wall 

derg, or ' crimson branch,' where were kept their spoils and trophies and 
the skulls of their enemies : and the teite brecc, or ' speckled house,' 
wherein were deposited the heroes' arms, so as to have them safely out of 
reach in case the owners should quarrel over their cups. There was also, 
according to Keating (p. 271), a hospital for the sick and wounded, called 
Brdn-Bherg, or the " Warrior's Sorrow." See the Paper on the Plan of 
Navan Fort, by M. D'Arbois de Jubainville in Rev. Celt., xvi., p. 1 : Joyce. 
Irjsh Names of Places, 1., p. 90 : and Mr. T. O. Russell, p. 58, 


is now about 17 feet high. It still retains — has all along 
retained — its ancient name, in the form of Greenan-Ely, 
where Ely correctly represents the sound of Ailigh, the 
genitive of Ailech* 

The Dalaradian princes had their residence in the sixth 
and seventh centuries, in a place of great repute called 
Rathmore-Moylinny. The fort still remains in the town- 
land of Rathmore two miles from the town of Antrim. 
Adjoining Rathmore townland is another called Rathbeg, 
which takes its name from the rath of another palace of 
much celebrity, where Dermot, the son of Fergus Cervall 
king of Ireland, lived for some time after he had left Tara, 
and where he was slain in the year 565, by Aed Dubh, 
king of Dalaradia.f 

Cruachan. — The chief palace of the kings of Connaught 
was Cruachan (or as it is now called, Croghan) from times 
beyond the reach of history down to the death of King 
Raghallach, who, as already related (vol. I., p. 409), was 
assassinated a.d. 648. It figures in various parts of this 
book, and is chiefly celebrated as being the residence of 
Ailill and Maive, king and queen of the province, in the 
first century of the Christian era. Here they held their 
court, which is described in the Tales of the Red Branch 
Knights in a strain of exaggerated magnificence : and from 
this the warlike queen set forth with her army to ravage 
Ulster and bring away the great brown bull which was the 
main object of the expedition, as described in the epic 
story of the " Tain bo Chuailnge " [Quelna]. 

The remains, which are situated three miles north-west 
from the village of Tulsk in Roscommon, are not imposing : 
for the main features have been effaced by cultivation. 
The principal rath, on which stood the timber palace and 

* See Ordnance Survey of Londonderry, p. 217. In this work, Dr. 
Petrie, with the assistance of O'Donovan and O'Curry, has given an 
elaborate historical, literary, and topographical description of Ailech. 

f See Reeves, Eccl. Antiqq., pp. 69 and 278-281 : Adamnan, p. 68, 
note : FM, a.d. 558 : and Voyage of Bran, 1. 58. 


the subordinate houses, is merely a flat, green, circular moat 
about an English acre in extent, elevated considerably 
above the surrounding land, with hardly a trace of the 
enclosing circumvallation. There are many other forts 
all around, so that, in the words of O'Donovan, who has 
described the place in some detail in the Four Masters 
under a.d. 1223 — the whole site may be said to be " the 
" ruins of a town of raths, having the large rath called 
" Rathcroghan placed in the centre ' : but they are scattered 
much more widely and at greater distances than those at 
Tara. Besides the homestead forts there are also, in the 
surrounding plain, numerous other antiquarian remains, 
indicating a once busy centre of royalty and active life — 
cromlechs, caves, pillar-stones, and mounds, including the 
cemetery of Relig-na-ree (about half a mile south of the 
main rath), which will be described at p. 556, below. 

Durlus Guaire or Dungory. — The royal house known by 
this name was the abode of Guaire [Goorie], the hospitable 
king of Connaught in the seventh century. It was built 
on a little island beside the seashore, half a mile north-east 
from the present village of Kinvarra on Galway Bay. On 
the site of the old dun a stone castle was subsequently 
erected, the ruin of which is now called " Dungory Castle," 
a name that commemorates the fortress of the hospitable 

Ailenn or Ailend, now Knockaulin. The most important 
residences of the kings of Leinster were Ailenn, Dinnrigh, 
Naas, Liamhain, and Belach-Chonglais or Baltinglass, in all 
of which the raths still remain. Ailenn is a round hill, now 
commonly called Knockaulin (Aulin representing ' Ailenn ') 
near Kilcullen in Kildare, rising 600 feet over sea-level, 
and 200 or 300 feet over the Curragh of Kildare which 
lies adjacent, and over all the plain around. The whole 
summit of the hill is enclosed by a huge oval embankment, 
514 by 440 yards, enclosing an area of 37 statute acres, 
* See Tromdamh, p. 120, note 2. 


one of the largest forts, if not the very largest, in Ireland. 
Within this great enclosure stood the spacious ornamental 
wooden houses in which, as we learn from our records, the 
Leinster kings often resided : for each king had at least 
three palaces (vol. I., p. 58, supra) which he occupied in 
turn, changing from one to another as it suited his pleasure 
or convenience. On the present 6-inch Ordnance map the 
fort is called the " Hill of Allen," instead of the proper 
modern popular name Knockaulin, which tends to confound 
it with the equally celebrated hill, now properly and 
universally called the Hill of Allen near Newbridge in 
Kildare, Finn mac Cumail's residence already described 
(vol. 1., p. 88) : a mistake evidently committed without 
O'Donovan's knowledge, when he and others were 
employed sixty or seventy years ago to settle the local 
names of Ireland.* 

Dinnrigh. — One of the most noted, and probably the 
oldest, of the Leinster palaces was Dinnrigh [Dinnree : the 
' dinn or fortress of kings '], also called by two other names, 

* I am informed on the best authority, that this mistake will be rectified 
in future editions of the 6-inch map. 

The fact that there are no remains of a fort on the Hill of Allen — i.e. the 
hill properly so-called, near Newbridge — a place of such celebrity as having 
been the residence of the renowned hero Finn mac Comail — seems so unac- 
countable as to lead some to conjecture that the ancient Irish writers 
confounded the names of the two hills — names which are somewhat like 
each other ; and that the fort of Ailenn above described may have been 
really Finn's residence. But this is pure conjecture without a shadow of 
evidence to support it. The absence of remains on the Hill of Allen has 
been already satisfactorily accounted for (vol. i., p. 90). As to the two 
names : they have never been confounded by any old writer, and could not 
possibly be, except by an amount of stupidity never exhibited by the 
writers of the Book of Leinster or the Book of the Dun Cow. The oldest 
form of the name of Finn's residence was A Imu, gen. Alman, dat. Almain ; 
which dative — in accordance with a well-known linguistic law — is often 
used as a nominative, on which a second genitive A Imaine has been formed. 
The oldest form of the name of Knockaulin is Ailend or Ailenn. The 
names of both vary somewhat in. form ; but there is one obvious and 
never-failing distinction : — that however Almu is varied, it always has the 
m : however A ilenn is varied, it never has an m The evidence that Finn 
lived at Almu, or what is now properly called the Hill of Allen, is quite 
as clear as that by which we know that Brian Boru lived at Kincora. 




Tuaim-Tenba and Duma-Slainge, ' Slainge's burial-mound,' 
because the Firbolg king Slainge died and was buried 
there (FM, a.m. 3267). Besides being very often men- 
tioned in the records, it was the scene of a tragedy which 
is related in detail in the historical story called " The 
Destruction of Dinnree," contained in the Book of Leinster, 
which has been edited and translated by Dr. Whitley 
Stokes.* Some two centuries and a half before the Chris- 
tian era, Cobthach the Slender of Breg murdered the king 

FIG. 208. 
Dinnree, the most ancient residence of the kings of Leinster. Now Ballyknockan 
Fort, on the west bank of the Barrow, half a mile below Leighlin-bridee, Carlow. 
(From Mrs. Hall's Ireland.) 

of Ii eland — his own brother — and also the king's son 
Ailill, and usurped the throne. But Ailill's son Labra 
Loingsech, or Lavra the Mariner, who fled to the Continent, 
returned after some years with a party of Gauls, and 
landed at Wexford, where he was joined by large contin- 
gents of the men of Leinster and Munster, who hated the 
usurper. Marching quickly and silently by night to Dinn- 
ree, where the king then happened to be holding court, he 
surrounded the palace, and setting fire to the houses while 

* In Zeitschr. Celt Phil., vol. in. 


the company were engaged in feasting, he burned all — 
palace, king, and courtiers — to ashes.* 

Dinnree continued to be used as a royal residence far 
into Christian times. From a passage in the Life of 
St. Finnchua,f we know that it was occupied early in 
the seventh century by " Old Nuada the Sage," king of 
Leinster : but when it was abandoned is not known. The 
old documents define very clearly the position of this 
palace : and the fine old fort still exists in good preserva- 
tion. It is situated on a high bank over the River Barrow 
on the west side, half a mile south of Leighlinbridge, and 
is now commonly known by the name of " Ballyknockan 
Moat." The moat or mound — figured in the illustration, 
p. 95 — is 237 feet in diameter at the base ; the circular 
plateau on the top is 135 feet in diameter, and 69 feet over 
the River Barrow (FM, vol. I., p. 15, note /). 

Fig. 209. 

North Moat, Naas : remains of ancient palace. House on top modern. 
(From a drawing by the author, 1857.) 

Naas. — In old times Naas was a place of great celebrity, 
where the Leinster tribes held some of their periodical 
aenachs or fair-meetings, from which it got the name 
of Nds-Laigen [Naas-Lyen], i.e. the ' assembly-place of 
Leinster,' corresponding exactly with the name of Nenagh 
in Tipperary. There were here two royal houses, the forts 
of which still remain. One is an ordinary circular, flat 

* See O'Curry, MS. Mat., 252 to 257 : Keating, 253 : and Joyce, Irish 
Names, 1. 93. -f Stokes, Lives of SS., pp. 237, 238. 


rath, now called the south moat, situated near the southern 
end of the town. The other, called the north moat, is a 
high, flat-topped mound on which the citadel once stood, 
but which is now occupied by an ugly modern house. 
Naas continued to be a residence of the Leinster kings till 
the death of King Cerball (already referred to, p. 49), who 
was slain by the Danes in 908.* 

Belach Chonglais. — Another of the Leinster palaces was 
at Baltinglass in the county Wicklow, whose old name was 
Belach-Chonglais (Cuglas's road) : but a still older name was 
Belach Dubthaire [Duff era]. f Here resided in the sixth 
century Branduff, the powerful king who defeated and 
slew Aed mac Ainmirech, king of Ireland, in the Battle 
of Dunbolg, a.d. 598 (vol. 1., p. 141, supra). On the hill 
rising over the town are two great raths or forts, the 
remains of the old residences. One, now called Rathcoran, 
is on the very summit, 1256 feet over sea-level. It is an 
oval, about a quarter of a mile in its longer diameter, 
having two ramparts, and containing about twenty-five 
statute acres. The other and smaller fort, now called 
Rathnagree, is on the northern slope of the hill : it has 
also two ramparts and covers about seven acres. 

Liamhain. — The name of Liamhain or Dun-Liamhna 
[Dun-lavna] is still preserved in that of Dunlavin, a 
small village in the county Wicklow. The mound of this 
residence is still to be seen a mile south of the village : 
but it has lost its old name and is now called " Tornant 
Moat." (Tornant, ' nettle-mound ' : ominous of ruin.) 

Side-Nechtain. — The Hill of Carbery in Kildare has a dim 
legendary history as a royal residence. It was anciently 
called Side-Nechtain [Shee-Nechtan] , i.e. ' Nechtan's Shee 
or fairy-hill ' : showing that it was the site of one of those 
elf-mounds described in vol. I., p. 254, supra. This Nechtan, 
according to the old documents, was king of Leinster, and 

* FM, vol. ir., p. 573, note o : see Tromdamh, p. 166, for a further 
account of Naas. f ^ ev - Celt., xin. 57 : Silva Gad., 411 


9 8 



also a poet. But the place contained a residence of a less 
shadowy kind ; for on the north-west slope there are still 
two remarkable and very perfect military raths or forts. 
Near the base of the hill is Trinity Well, the source of 
the Boyne, the enchanted well that in old times burst up 
and overwhelmed Boand, Nechtan's queen, as described 
in vol. I., p. 284, supra. But in subsequent times the 
Christian missionaries — as in case of many another well 
(vol. 1., p. 366) — removed its heathenish character and 

Fig. 210. 
Carbury Castle, County Kildare. (From a photograph.] 

associations, and dedicated it to the Holy Trinity. The 
Anglo-Norman De Berminghams, who took possession of 
the district, having an eye to something more substantial 
than Dedannan fairy palaces, took advantage of the selec- 
tion of their immediate Milesian predecessors and built 
a splendid castle not far from the old Irish fortresses, near 
the summit, the ruins of which are now conspicuous for 
leagues round the hill.* 

Cashel was one of the most renowned seats of the 
North Munster kings, though not the oldest as a royal 

* See Wilde's Boyne, pp. 24 to 32. 




residence. Its chief feature is the well-known lofty 
isolated rock overlooking the surrounding plain — the 
magnificent Golden Vale, as it is called, from its fertility. 
The most ancient name of the rock was Sidh-Dhruim [Shee- 
drum or Sheerim], ' fairy-ridge ' ; but it was also called 
Lec-na-gced [Lack-na-gade] , the ' rock of the hundreds,' 
and Druim-Fiodhbhuidhe [Drum-Feevee] , ' woody-ridge ' ; 
and in Christian times Lec-Phatraic, ' St. Patrick's Rock.' 
An ancient legend still preserved in old Irish mss., and 


Eft'fo 1 rk 

I gfiS m r^» 

'^^**Ajjr ^ifciiiiij 

\r S5S8 

Fig. an. 

Rock of Cashel (top of Round Tower appears to the right). (From Brewer's 

Beauties of Ireland. Drawn by Petrie.) 

given by Keating, relates that two swineherds, while 
feeding their flocks in the woods round the hill, in the 
beginning of the fifth century, saw an angel as bright as 
the sun standing on the rock, blessing the place with voice 
more melodious than any music, and prophesying the 
coming of St. Patrick. Core mac Luighdheach, king of 
Munster, coming to Ijear of this, immediately took posses- 
sion of the whole place, and on the summit of the rock 
built a stronghold, which was known as Lis-na-Laochraidhe 
[Laikree], the ' fort of the heroes,' and which then became 
the chief residence of the Munster kings, and continued so 


till the beginning of the twelfth century. In iioi King 
Murkertagh O'Brien dedicated the whole place to the 
church, and handed it over to the ecclesiastical authorities, 
since which time it figures chiefly in ecclesiastical history. 
Then began to be erected those splendid buildings which 
remain to this day ; so that the " Rock of Cashel " is now 
well known as containing the most, imposing group of 
ecclesiastical ruins in the united kingdom.* 

Grianan Lachtna. — One of the ancestral residences of 
the Dalcassian kings of Thomond or North Munster was 
Grianan-Lachtna or Greenan-Lachna, the fine old fort of 
which is still to be seen occupying a noble site on the 
south slope of Craig-Liath or Craglea in Clare, over the 
western shore of Lough Derg, two miles north of Killaloe. 

Kincora. — But when Brian Boru ascended the throne, 
he came to live at Kincora, where the remains of the 
palace have all disappeared, inasmuch as the site is now 
occupied by the town of Killaloe. The O'Briens, as 
kings of Thomond, continued to reside at Kincora for two 
centuries after the Battle of Clontarf : but about 12 14 
they removed their residence to Clonroad near Ennis. 
One of the outlying forts — a very fine one — still remains, 
however, beside the Shannon, a mile north of Killaloe, and 
is now known by the name of Beal Boru. 

Dungrud. — In East Munster there were, from remote 
times, two royal residences. One was Dun-gcrot, now 
called Dungrud or Dungrott, in the Glen of Aherlow, at 
the foot of the Galtys, on the site of which the English 
of Galbally erected a strong castle. 

Caher. — The other East Munster palace was on a little 
rocky island in the river Suir at the town of Caher, in 
Tipperary. It was oringially called Dun-iasgach, the ' fish- 
abounding dun,' from the dun that constituted the original 

* On all this, see O'Curry, MS. Mat., 485 and 623 : Comyn's Keating, 
1. 123 : and for a description of the buildings on it, Petrie's article in the 
Irish Penny Journal, p. 17. 




fortress-palace. This was succeeded by a circular stone 
caher, which gave the place its present name. The castle 
was- built by the Anglo-Normans on the site of the caher.* 
Another of these Munster palaces was Dun-gclaire 
[Doonglara], the fort of which is still in good preserva- 
tion, standing at the northern base of the mountain of 
Slievereagh near Kilfinnane, two miles nearly north-west 
from Ballylanders, on the left of the road as you go from 

Caher Castle in 1845 : on site of the old palace. (From Mrs. Hall's Ireland.) 

this village to Knocklong. It covers about four statute 
acres, and is now called Doonglara, or more often Lis- 

Brugh-righ. — Bruree in the county Limerick, situated 
beside the river Maigue, was from remote times one of 
the seats of the kings, of South Munster, as its Irish name 
Brugh-righ indicates, signifying the ' House of Kings.' 
It was also called Dun Eochair Mhaighe, the ' fort on the 

* See Petrie's article on Caher Castle in Irish Penny Journal, p. 27. 
For all these Munster palaces, see Comyn's Keating, pp. 121 to 129. 


brink (eochair) of the River Maigue.'* The illustrious 
King Ailill Olom, ancestor of many of the chief Munster 
families, lived there in the second centuryf : and it 
continued to be occupied by the Munster kings till long 
after the Anglo-Norman Invasion. The Anglo-Norman 
chiefs also adopted it as a place of residence, as they did 
many others of the old Irish kingly seats : and the ruins of 
two of their fine castles remain. There are still to be seen, 
along the river, several of the old circular forts, the most 
interesting of which is the one now universally known in 
the neighbourhood by the name of Lissoleem, inasmuch as 
it preserves the very name of King Ailill Olom, whose 
timber house was situated within its enclosure. It is 
situated on the western bank of the river, a mile below the 
village, in the townland of Lower Lotteragh, in the angle 
formed by the Maigue and a little stream joining it from 
the west. It is a circular fort with three ramparts, having 
the reputation — like most other raths — of being haunted 
by fairies : and as it is very lonely and much overgrown 
with bushes, it is as fit a home for fairies as could well be 

This king's name, Ailill Olom, signifies ' Ailill Bare-Ear,' 
so called because — as already mentioned (vol. I., p. 263) — 
one of his ears was cut off in a struggle with the fairy lady 
Aine of Knockainy. Olom is accented on the second 
syllable, and is compounded of 0, ' an ear,' and lorn, 
' bare,' : in the name " Ailill Olom " it is in the nomina- 
tive case : " Ailill Bare-Ear " (not " of the Bare-Ear ") : 
like the English names William Longsword, John Lackland, 
Richard Strongbow. But when placed after " Lis," it takes 
— as it should take — the genitive form, " Oluim " : and 
" Lis-Oluim," which is exactly represented in sound by 
" Lissoleem," signifies ' Olum's lis or residence.' Many 
examples of the preservation of very old personal and 
other names in our existing topographical nomenclature 

* Comyn's Keating, p. 123 t Silva Gad > P- 34 8 - 


are given in my " Irish Names of Places " ; and this case 
of Lissoleem — which has not been noticed before — is fully 
as interesting as any of them. 

Temair-Luachra. — In the time of the Red Branch 
Knights and of the Munster Degads (vol. I., p. 86), and 
from immemorial ages previously, the chief royal residence 
of South Munster was Teamair or Tara-Luachra, the fort 
of which in all probability still exists, though it has not 
been identified. Mr. W. M. Hennessy, in his Introduction 
to the Mesca Ulad, has brought together the several 
notices bearing on its position : and the Rev. Dr. Hogan 
has a remark on the subject in Rossnaree (p. 23, note 7 ). 
It was well known in the time of Elizabeth ; and anyone 
acquainted with the country, who would take the trouble 
to walk over the exact locality indicated, and make 
inquiry among the old people, would be able, as I believe, 
to light on and identify the very fort. 

Knockgraffon. — Another noted Munster palace was 
Cnoc-Rafonn, now called Knockgraffon, three miles north 
of Caher in Ppperary, where the great mound, 60 or 70 
feet high, still remains, with the ruins of an English castle 
beside it. Here resided, in the third century, Fiacha 
Muillethan [Feeha-Mullehan] , king of Munster, who, when 
the great King Cormac mac Art invaded Munster in an 
attempt to levy tribute, defeated him at Knocklong and 
routed his army : an event which forms the subject of the 
historical tale called Forbais Droma Damhghaire, or the 
" Siege of Knocklong." The fort is now as noted for 
fairies as it was in old times for royalty (see Crofton 
Croker's story " The Legend of Knockgraffon.")* 

* A full list of the royal seats of Munster, annotated by O'Donovan, is 
given in the Book of Rights, pp. 89 to 95. 

Ornament : composed from the Hook of Ivells 



Section i. Meals in General. 

inner, the principal meal of the day, was called 
in Irish prainn or praind, probably a loan- 
word from the Latin prandium, which is 
explained by the Irish proind in Zeuss (67, »). 
Hence the refectory of a monastery was called 
praintech, literally ' dinner-house.' Dinner was 
taken late in the evening both among the laity and in 
monasteries. "At the end of the day his [Patrick's] 
" charioteer said to him : ' Thou hast left a cross to-day 
" in thy path without visiting it.' Thereupon Patrick left 
" the guest-house and his dinner (a tech-noiged ocus a 
" praind), and went back to the cross."* In the notes 
to the Feilire of Aengus (p. 62), it is stated that Ciaran's 
dinner (praind), every night, consisted of a little bit of 
barley-bread, two roots of a vegetable, and a drink of 

It was usual to have a light meal between breakfast 
and dinner corresponding with the modern luncheon. It 
was called etsruth or etrud, which Cormac (p. 68) explains 

* Trip. Life, 125 : see also Silva Gad., 113, 34> 


as " eter-shod [' middle-meal '], the middle-meal of the 
day." The time is given more definitely in an entry 
(quoted by Stokes under this explanation) in an ancient 
ms. : — " Etrud, i.e. etar-suth (' middle-fruit '), i.e. between 
" morning and evening ; or rith-etir, or ' middle-running,' 
" i.e. [running] at midday." It was a custom among the 
laity, as well as in the monastic communities, to have 
better food on Sundays and church festivals than on other 
days, as appears from many passages in the Laws, and in 
ecclesiastical and general literature. 

Among the higher classes great care was taken to seat 
family and guests at table in the order of rank ; any 
departure from the established usage was sure to be 
resented by the person who was put lower than he 
should be ; and sometimes resulted in serious quarrels or 
wars. The placing of Prince Congal by Domnall king 
of Ireland below his proper place at the banquet of 
Dun-nan-gedh was one main cause of the great Battle of 
Moyrath (fought a.d. 637).* It was especially necessary 
to observe the proper formalities at banquets and on all 
state occasions, where the arrangements were under the 
direction of the rechtaire or ' house-steward ' (for whom see 
vol. 1., p. 64). At the banquet given by King Concobar 
mac Nessa at Dun-da-Benn, as narrated in the Mesca Ulad 
(p. 13) of which the original is in the Book of Leinster, the 
banqueting-hall was " arranged by Concobar according to 
" deeds and parts and families ; according to grades and 
" arts and customs, with a view to the fair holding of the 
" banquet." (Any great banquet or feast was called fled, 
and sometimes imdell.) 

The account given by Keating (pp. 302-3), which he 
took from ancient documents now lost, of the seating of the 
guests at the state banquets of Tara, is very interesting. 
The persons entertained were of three main classes : — 

♦Moyrath, 29, 31. For the battle, see Joyce, Short History of 
Ireland, p 153. 


Lords of territories ; the commanders of the bands of 
warriors who were kept permanently and maintained at 
free quarters by the king at Tara ; and the ollaves or 
learned men of the several professions. The territorial 
lords were regarded as of higher rank than the military 
commanders ; and each chief of both classes was attended 
by his " shield-bearer " or squire. It was the duty of the 
ollave shanachie to have the names of all written in two 
separate rolls, in exact order of precedence : and in this 
order they sat at table. 

The banquet-hall was a long narrow building, with 
tables arranged along both side-walls. Immediately over 
the tables were a number of hooks in the wall at regular 
intervals to hang the shields on. One side of the hall was 
more dignified than the other, and the tables here were for 
the lords of territories : those at the other side were for 
the military captains. The upper end was reserved for 
the professional ollaves : the dependents — always a large 
company — sat at the lower end. 

Just before the beginning of the feast all persons left the 
hall except three : — A Shanachie or historian : a Bollscari 
or marshal to regulate the order : and a trumpeter (fear- 
stuic) whose duty it was to sound his trumpet just three 
times. At the first blast the shield-bearers or squires of the 
lords of territories came round the door and gave their 
masters' shields to the marshal, who, under the direction of 
the Shanachie, hung them on the hooks according to ranks, 
from the highest to the lowest : and at the second blast 
the shields of the military commanders were disposed of 
in like manner. At the third blast the guests all walked 
in leisurely, each taking his seat under his own shield 
(which was marked with his special cognisance : see vol. I. 
p. 125). In this manner all unseemly disputes or jostling 
for places were avoided. No man sat opposite another, 
as only one side of each row of tables was occupied, namely, 
the side next the wall. 


Keating does not, in this passage, give the arrangement 
when the king was present : but this is described in other 
authorities ; as well as elsewhere in Keating (415). The 
king was always attended at banquets by his subordinate 
kings, and by other lords and chiefs : and great formality 
was observed in seating all. In the " Wooing of Emer " 
(p. 69), it is stated that when the company sat drinking in 
the banquet-hall of Emain, " no man of them would touch 
the other." Those especially on the immediate right and 
left of the king had to sit at a respectful distance. At 
the feasts of Tara, Tailltenn, and Ushnagh, it was the 
privilege of the king of Oriell to sit next the king of 
Ireland, but he sat at such a distance that his sword just 
reached the high king's hand : and to him also belonged 
the honour of presenting every third drinking-horn brought 
to the king.* According to Kineth O'Hartigan, while 
King Cormac mac Art sat at dinner, fifty military guards, 
or " heroes," remained standing beside him.f The arrange- 
ments for seating subordinate kings, at banquets given by 
the Hy Neill Monarchs, may be seen in the Battle of 
Moyrath, pp. 29, 31 ; and a much more detailed account 
of those for king and guests at Brian Boru's banquets 
at Kincora is given, from old authorities, in O'Curry's 
Lectures. I At Tara it often happened that the women 
did not sit at banquets with the men : they had a banquet- 
hall for themselves. But in the feasts at other places men 
and women always, or nearly aways, banqueted in the 
same hall : the women, however, generally sitting apart : 
and they often wore a mask — sometimes called fethal — 
which hid or partly hid the face.§ 

This rigid adherence to order of priority at table con- 
tinued in Ireland and Scotland down to a recent period : 

* Book of Rights, 137. t Petrie, Tara, 191, 192. 

% Man. & Cust., 1. 121. See also Petrie's Tara, p. 199 et seq., for the 
detailed arrangements in Tara. 

§ See Law Tract quoted by O'Curry, Man. & Cust., n. 114. 


and it continues still in a modified and less strict form 
everywhere. Readers of Scott will call to mind the scene 
in " The Lord of the Isles," when the seneschal — corre- 
sponding with the Irish rechtaire — seated the unknown 
strangers next the prince : — 

" Then lords and ladies spake aside, 
And angry looks the error chide 
That gave to guests, unnamed, unknown, 
A place so near their prince's throne." 

An odd instance of the Irish " pride of place " in the 
eighteenth century is related by Hardiman* concerning 
Arthur O'Neill, the celebrated Irish harper. He was 
universally respected, partly on account of his musical 
abilities, but more because he belonged to the illustrious 
family of O'Neill : and he always sat at table among the 
highest people. Once at a public dinner in Belfast, which 
was attended by all the local nobility and gentry, the 
noble lord who presided apologised to him for being 
accidentally placed so far down from the head of the 
table. " O my lord," replied he, " apology is unnecessary : 
wherever an O'Neill sits, that is the head of the table." 

The host stood up before the meal and formally 
welcomed his guests. f At all state banquets particular 
joints were reserved for certain chiefs, officials, and pro- 
fessional men, according to rank. These are set forth in 
several authorities, though with some differences : they 
may be seen in detail in Petrie's Tara (pp. 199 et seq.), 
taken from the Book of Leinster. The following shorter 
statement is given in the treatise on Irish Ordeals trans- 
lated by Stokes, which is almost identical with that given 
by the commentator on the Senchus M6r — "A thigh 
' [laarg] for a king and a poet : a chine [crotchet] for a 

* Ir. Minstr., n. 412. 

t See Moyrath, 25 : Ir. Texte, 1. 99. paragraph, 6, with a translation 
in Hib. Minora, 59. 


" literary sage [sai litri : vol. I., p. 434] : a leg [colptha] 
" for a young lord [ogtigern] : heads [cuind] for charioteers : 
" a haunch [les] for queens."* A similar custom existed 
among the ancient Gauls and also among the Greeks. f 
A remnant of this old custom lingered on in the Western 
Islands of Scotland to the time of Martin (p. 109), 200 
years ago. When the chief of an island killed an animal, 
he gave head, feet, entrails, and such like, to his depen- 
dents, the head being due to the smith, the udder of a cow 
to the piper, &c. At a still later time — 1773 — Dr. Johnson, 
in his account of his visit to the Hebrides, records the 
prevalence of the custom there in the following words :— 
' When a beef was killed for the house, particular parts 
' were claimed as fees by the several officers or workmen. 
' . . . The head belonged to the smith, and the udder of a 
' cow to the piper : the weaver had likewise his particular 
' part : and so many pieces followed these prescriptive 
' claims that the laird's was at last but little." Even so 
late as 1839, when Petrie wrote his Essay on Tara, the 
custom was partially kept up in some parts of Ireland, 
where the farmers, when they killed a beef or a pig, always 
sent the head to the smith, whose kitchen was often 
garnished with from fifty to a hundred heads, obtained in 
this manner. J Sometimes the marrow-bones were assigned 
to a particular member of the household, to whom everyone 
passed his bone after picking it : and woe betide anyone 
else who broke a bone for marrow. § 

In the time of the Red Branch Knights, it was the 
custom to assign the choicest joint or animal of the whole 
banquet to the hero who was acknowledged by general 
consent to have performed the bravest and greatest exploit. 
This piece was called curath-mir, i.e. ' the hero's morsel or 

* Ir. Texte, in. 206 : Br. Laws, 1. 49 : see also " Milgitan " in Corm. 
Gloss., 107 : and for a further detailed account see Ulster Journ. Archaeol., 
in. 119. 

f Iliad, xxn., and vn. 320; Odyss., iv. 66. 

I Petrie's Tara, p. 212. § For instance, see Moyrath, 71. 


share ' (inir). There were often keen contests among the 
Red Branch heroes, and sometimes fights with bloodshed, 
for this coveted joint or piece : and some of the best 
stories of the Tain hinge on contests of this kind.* This 
usage, which, according to Diodorus Siculus, prevailed 
among the continental Celts in general, and which also 
existed among the Greeks,f seems to have continued in 
Ireland to comparatively late times : for the Senchus Mor 
mentions among the offences for which penalty was due 
" Carrying away the hero's morsel from the person to 
whom it belongs." The word used here is danttnir, which 
the gloss explains by curath-mir. O'Donovan, the trans- 
lator of the Senchus M6r, considers the marrowbones 
mentioned above — assigned to one particular individual— 
as a sort of curath-mir. X 

Fig. 213. 

Small Table : 28 inches long, :6 inches broad, and 5 inches high : made 
of willow: found in a bog in Tyrone, five feet under the surface. (From 
Wilde's Catalogue, p. all.) 

Tables were, as we have seen, used at the great feasts. 
But at ordinary meals, high tables, such as we have now, 
do not seem to have been in general use. There were 
small low tables, such as that in the illustration, each used 
no doubt for two or more persons, who sat or reclined 
on low couches or seats of some kind at meals. Often 
there was a little table laid beside each person, on which his 
food was placed — the meat on a platter.§ In late times — 
the sixteenth century — Derrick, in his " Image of Ireland," 

* See Fled Bricrend, p. 15. Nearly the whole of this tale is occupied 
with the contests of the three Red Branch heroes, Cuculainn, Loegaire the 
Victorious, and Conall Cernach for the curath-mir. 

t De Jubainville, Cours de Litt. Celt., VI., pp. 3, 4. 

J Br. Laws, 1. 177, 181, bottom, and note 1. 

\ Joyce's Old Celtic Romances, 125. 


represents the Irish at their meals in this fashion : but 
Derrick's words and pictures must be received with 
caution, for they are all more or less caricatures. 
According to Giraldus,* his countrymen, the Welsh, had 
no tables at all at their meals : and very probably this 
was the case in the general run of the houses of the Irish 

Forks are a late invention : of old the fingers were used 
at eating. In Ireland, as in England and other countries 
in those times, each person held his knife in the right hand, 
and used the fingers of the left instead of a fork : just as 
we see described in the " Vision of Mac Conglinne " (p. 64). 
Sometimes — as at banquets, and among very high-class 
people — the carvers cut off great pieces from the joint, 
which they brought round and put on the platters. The 
attendants who supplied food and drink in this manner at 
great dinners were called ddilemain, ' carvers, distributors, 
or spencers ' (from ddil, ' to cut or divide ') ; and deogbhaire, 
or deochbhaire, ' cupbearers ' (from deoch ' a drink '). But 
more commonly each person went to the joint, and using 
his left-hand fingers to catch hold, cut off a piece for 
himself and brought it to his own platter. Even so late 
as the sixteenth century this was the custom in England, 
according to Roberts (p. 342), who says that dinner was 
served without knives or forks, but each had his own clasp- 
knife, and going to the dish, cut off a piece for himself : and 
he gives this illustrative verse from " The Mirror of Good 
Manners," by Alexander Barclay (sixteenth century) : — 

" If the dish be pleasant, either flesche or fische, 
Ten hands at once swarm in the dishe." 

Even towards the end of the seventeenth century " they 
" scarce ever make use of forks or ewers, for they wash 
" their hands by dipping them into a basin of water " 
(Social Engl., iv. 490). That this was the manner of dining 

* Description of Wales, x. 


in very early times in Ireland appears from a passage in 
the " Second Battle of Moytnra," where it is said of the 
inhospitable King Bres : — " Their knives [i.e. the knives of 
" his subjects the Dedannans] were not greased [nibtar 
" beoluide a sceanai] by him, and their breaths did not smell 
" of ale at the banquets."* The Greeks and Romans had 
no knives or forks at meals : they used the fingers only, 
and were supplied with water to wash their hands after 
eating : yet the meat must have been cut in some way 
either by the guests or by the attendants. The Irish 
people picked the bones as many do now, partly with the 
knife and partly with the teeth, f In the story told in 
vol. I., p. 414, sufita, the bishop, being suddenly called on 
deck, came up from his dinner holding in his hand a big 
bone, which no doubt he was picking in the good old 

As early as the eighth or ninth century the higher 
classes used napkins at table, for which they had a native 
word, lambrat, i.e. ' hand-napkin ' (lam, ' hand ' : ' brat, a 
cloth '). In a manuscript quoted by Zeuss (653, 45 ) the 
Latin words mappa and mantile are explained by the 
eighth-century Irish Glossator, lambrat bis tar glune, ' a 
napkin that is usually placed over the knees ' : and in 
another part of the same manuscript the Latin gausape is 
explained by the single Irish word lambrat (Z., 854, 22). 
In the Latin version of the Voyage of St. Brendan there 
is a more direct reference to the use of napkins. The 
voyagers went into a mansion on an island, in which they 
found a large hall with couches and seats and water to 
wash their feet, and plenty of food. " St. Brendan ordered 
" the serving-brother to bring forward the meal which God 
" had sent them : and without delay the table was laid with 
" napkins (linteamina) and with white loaves and fish for 
" each brother." % But perhaps linteamina here means 

* Rev. Celt xn. 69. f See for example Moyrath p. 71. 

I Brendaniana, 121 : Card. Moran, 93, , 3 . 


' tablecloths.' I suppose the chief use they made of the 
napkins was to wipe the left-hand fingers ; which was 
badly needed. They sometimes used dried hides as 
tablecloths. Cathal the king-glutton (eighth century : 
for whom see Index) was once eating apples as part of his 
dinner, and " he began supplying his mouth from both 
" hands with the apples that were on hides round about 
" him." Mac Conglinne (pp. 46, 50) was there and began 
importuning him for some of the apples, so that the king 
threw him one after one ; till at last he " flung him hide, 
apples, and all." 

It was the custom, both in monastic communities and 
in secular life, to take off the shoes or sandals when sitting 
down to dinner ; which was generally done by an attendant. 
The Romans we know had the same custom : " the cover- 
ing of the feet was removed before reclining at meals."* 
It is related in the " Vision of Mac Conglinne " (p. 46) that 
the gluttonous king of Munster mentioned above, who had 
a lon-craos, i.e. a 'demon of, gluttony' in his stomach, 
sitting down one day to dinner, was in such a hurry that 
he fell to before the attendant had loosed the thong of 
one shoe (broc). And Adamnan tells us in his " Life of 
St. Columba" (pp.121, 122) that St. Canice, in his monastery 
of Aghaboe (sixth century), while in the act of breaking 
the bread in the refectory, having a revelation that 
St. Columba and his companions were in great danger 
on sea, hastily left the table and went to the church to 
pray " with one shoe on his foot, the other being left 
behind in his great haste." -Another similar incident is 
related in Oengus's Feilire (p. 9, i 5 ). We may infer 
from the existence of this custom that the Irish, like the 
Romans, reclined during meals on couches on which the 
feet also rested. 

The Irish, like the people of all other countries, had 
their meals commonly served hot, immediately after 

* Smith's Diet. Greek & Rom. Antiqq., " Calceus." 



cooking. But at great banquets the food must have been 
generally taken cold ; for in such cases, with the appliances 
then available, it would have been impossible to serve hot. 
Accordingly, we constantly read that before a banquet the 
whole of the food was cooked and laid out on tables in the 
first instance. Just before the guests came into the hall to 
the Feast of Dun-nan-gedh, Prince Congal is brought in 
to view the viands, all laid out and ready, and eats part 
of a goose-egg. When Lomman, king of Hy Fidgente, 
prepared a feast for St. Patrick, a youth named Nessan, as 
soon as he heard of it, came with his mother, bringing «. 
cooked ram as a contribution.* 

2. Drink. 

In old times people were quite as fond of intoxicating 
drinks at dinners and banquets as they are now : and we 
are constantly told in the tales that when the cups went 
round, the company became mesca medarchaini, ' ex- 
hilarated and right merry.' They sometimes drank more 
than was good for them too ; and on one occasion of this 
kind the Ulstermen marched southwards in a drunken raid 
on Munster, which is the subject of the old tale in the Book 
of Leinster called Mesca Ulad, the " Intoxication of the 
Ultonians," edited by Mr. Hennessy in the MS. series of the 
Proc. Roy. Ir. Acad. Yet drunkenness was looked upon 
as reprehensible. In Cormac's Glossary (p. 116) is a deri- 
vation for the word mesci, ' drunkenness,' implying that 
radically it meant " more of reproach than sense or 
sobriety " : and in the Tripartite Life (p. 137) it is related 
that when St. Patrick was in Connaught, a certain king, 
while in a state of intoxication, came to visit him : at which 
the saint was sorely displeased, and prophesied that that 
king's descendants would be ale-tipplers and would go 
to the bad in the end. At their feasts they often 

* Trip. Life, 205. 


accompanied their carousing with music and singing. 
Maildune and his men visiting a certain island, saw the 
people feasting and drinking and " heard their ale-music " 
(corm-cheol) .* 

In very early times ladies often sat with the men at 
the banquets, and joined in the festivities. It appears, too, 
that the Irish ladies of those times could take a moderate 
part in a drinking-bout with their male friends, like those 
of Wales and Scandinavia, as we read of them in the tales 
of those countries ; like the Egyptian ladies of four or five 
thousand years ago ; and like English ladies of much later 
times (Soc. Engl., II. 422 ; and iv. 161). At Bricriu's Feast, 
where the ladies were present, after the revelling had gone 
on for some time, Fedelma, wife of Loegaire the Victorious, 
went forth from the banquet-hall with her fifty hand- 
maidens, " after heaviness of drinking " (iar trommi oil), 
as much as to say, they went out to shake off the effects 
by a walk in the fresh air.f In the ancient tale called the 
" Vision of Cahirmore," king of Ireland a.d. 174-177, we 
read that on one occasion, while the king was celebrating 
the Feis of Tara, the whole company, after dinner, got so 
drunk that they fell asleep, and a thief slipped in and stole 
the queen's diadem % : which seems to imply that the queen 
herself was present taking a comfortable nap like the rest. 

Besides plain water and milk, the chief drinks were ale. 
mead or metheglin, and wine. Giraldus Cambrensis (Top. 
Hib., 1. v.) remarks that Ireland never had vineyards : but 
that there was plenty of wine supplied by foreign com- 
merce ; and he mentions Pditou especially as supplying 
vast quantities in exchange for hides. This account is 
corroborated by the native records, from which we learn 
that wine (Irish fin, pron. feen : a loan-word from Latin) 
was imported in very early ages ; and it is frequently 

* Rev. Celt., x. 81. 

f Fled Brier., 17, 154 : see also Bee Fola, 179 and note 23. 

J Kilk. Archaeol. Journ., 1872-3, p. 29. 


mentioned as an accompaniment at banquets. Muirchu, 
writing in the seventh century, tells us that when St. 
Patrick came to Tara on Easter Sunday a.d. 433, the 
kings, princes, and druids were feasting and drinking wine 
in the palace with King Laegaire.* In the year a.d. 533 
the palace of Cletty was set on fire by a revengeful woman 
while Murkertagh mac Erca, king of Ireland, was feasting 
in it with his nobles ; and the king, to avoid the fire, 
plunged into a butt of wine, in which he was drowned, f 
Domnall, king of Ireland (a.d. 627 to 641), in preparation 
for the banquet of Dun-nan-gedh, provided three kinds of 
drink — wine, mead, and ale. Wine is also often mentioned 
in Cormac's Glossary and in other Irish authorities as a 
well-known drink. 

Of all the intoxicating drinks ale was the most general, 
not only in Ireland, but among all the peoples of northern 
Europe : and the more intoxicating it was the more 
esteemed. One of the attractions of Midir's wonderful 
fairyland was " ale which is strongly intoxicating."]: Irish 
ale was well known from the earliest period, even on the 
Continent, as we see from the statement of Dioscorides in 
the first century : — " The Britons and the Hiberi or Irish, 
" instead of wine, use a liquor called courmi or curmi, made 
" of barley. "§ This author caught up correctly the ancient 
Irish name for ale, which was cuirm or coirm (gen. corma) : 
and hence coirmthech, ' ale-house,' i.e. a house in which ale 
was made. The present word for ale is linn or leann: 
and although this, too, was one of the words for ale in old 
times, it was often used to denote drink in general. Ale 
was a native product, and was reddish in colour as now. 
Its manufacture was understood everywhere ; and the 
whole process is given in detail in the Senchus Mor, and 
in the commentaries and glosses on it.|| The grain chiefly 

* Hogan, Docum., 37. % O'Curry, Man. & Cust., XX. 191. 

t Petrie, Tara, 120. § Ware, Antiqq., 183. 

|| Br. Laws, XI., pp. 241 to 245, from which the following details of the 
processes are chiefly taken. 




used was barley ; and what grew on rich land was most 
valued for the purpose : but it was also often made from 
rye, as well as from wheat and oats. 

The corn, of whatever kind, was first converted into 
malt : Irish brae or braich : gen. bracha. For this purpose 
it was steeped in water for 
a certain time, after which 
the water was let off slowly, 
and the wet grain was 
spread out on a level floor 
to dry. During this time 
persons turned it over and 
over and raked it into 
ridges to bring all parts 
in turn to the surface. It 
was next dried in a kiln 
{aith, pron. ah) till the 
grain became hard. This 
dried grain was malt. If 
not intended to be kept in 
grains, it was ground with 
a quern or in a mill, and 
was then either put into 
sacks as it came from the 
mill, or made into cakes 
and dried. Malt cakes 
were often so hard that 
before using they had to 
be broken in pieces with a 
mallet and ground again 
in a mill to reduce them back to meal.* Whether as 
unground kiln-dried grains, or as meal in bags, or as dried 
cakes, this brae or malt kept for any length of time ; and 
it was often given in payment of rent or tribute, as 
repeatedly mentioned in the Book of Rights. 

* Fled Bricrend, 67 ; O'Curry, Man. & Cust., 1. 309. 

FIG. 214. 
Bronze Strainer, found in the crannoge of 
Moylarjr, County Antrim. Cup-shaped, 4^ inches 
wide and ij£ inch deep. Observe, the holes are 
not at random ; they form curve-patterns. (From 
the Journ. Roy. Soc. Antiqq. for 1894, p. 319.) 


When the ale was to be prepared, the ground malt 
was made into a mash with water, which was fermented, 
boiled,* strained, &c., till the process was finished. Conall 
Derg O'Corra had in his house strainers (men) with their 
cries always at work ag sgagadh leanna, ' a-straining ale,' in 
hospitable preparation for guests, f Malt, and of course the 
ale, might be spoiled by mismanagement at any stage of the 
process ; and the Senchus Mor mentions three successive 
tests : one after kiln-drying and before being ground, by 
putting a grain under the tooth to try whether it was 
sound and free from bitterness : another after grinding, 
before it was made into a cake, to ascertain if it was free 
from mawkishness ; and a third when it was in mash, 
before it was put to ferment. 

Ale was often made in private houses for family use : 
for everywhere among the people there were amateur 
experts who understood the process. But there were 
houses also set apart for this purpose, where a professional 
brewer carried on the business. Some ale-making houses 
were what were called " lawful " (dligtech, ' lawful,' ' legal- 
ised,' or ' licensed '), that is, the law took cognisance of them 
and received their certificates : others were unlawful — un- 
legalised or unlicensed, which meant, not contrary to law, 
but merely that the law took no cognisance of them — did 
not accept their certificates. This made an important 
difference in cases of dispute ; for whenever a tenant paid 
part of his rent or tribute in ale which had been made in a 
lawful ale-house, if he proved that the three tests had been 
applied with satisfactory results while the malt was in the 
house, he was free from responsibility, even though the ale 
turned out bad. J But if ale which had been made in an 
unlawful house proved to be bad after being sent in pay- 
ment to the chief, it was forfeited, and the tenant had to 

* Boiled : see Br. Laws, iv. 311, „. f ^ ev - Celt., xiv., 27. 

I On this see Br. Laws, V. 167, in addition to the passages in Br. Laws, 
vol. u. referred to in note ||, p. 116, abov$. 


make good the loss, even though the three tests had been 
applied. : for the certificate from the unlawful house counted 
for nothing. Probably the proprietor of a licensed ale- 
making house took advantage of his privilege to make 
higher charges, like legally recognised experts of all kinds 
at the present day. 

Among the members of St. Patrick's household was a 
brewer — a priest named Mescan. A professional brewer 
was called cerbsire or cirbsire [kirvshirre], a loanword from 
Latin cervicia or cerviciarius (which is itself a borrowed 
Gaulish word). But there was also a native term for a 
brewer, scoaire [3-syll.], which is given in Cormac's Glossary 
(p. 31) to explain cerbsire. It is probable that a " lawful " 
alehouse had always one of these men, and not a mere 
amateur, in charge of it. 

When people felt indisposed or out of sorts, it was usual 
to give them a draught of ale to refresh or revive them, as 
we now give a cup of tea or a glass of wine. At Easter 
time, and after the restraint of Lent was over, the people 
sometimes indulged in a good drink of ale : and a supply 
was commonly kept in the churches, so that members of 
the congregation might take a drink when it was lawful to 
do so. At a certain Easter time, as we are told in the Life 
of St. Brigit, she brewed ale to supply the churches all 
round her : and this she did as a kindly and charitable act. 
St. Domangart or Donard, a disciple of St. Patrick, always 
kept a pitcher of ale and a larac or leg of beef with its 
accompaniments every Easter at his church of Maghera 
near Slieve Donard : " and -he gives them to Mass-folk 
" [i.e. those that have been at Mass] on Easter Tuesday 
" always."* 

Yeast or leaven — called in Irish descad and serba — 
made from malt, was used in brewing, and also in baking. 
The house of the hospitable brewy Conall Derg O'Corra 

Trip. Life, 131. 


was never without a " sack of malt for preparing yeast " 
(miach bracha re jrithealamh ndeasgadh)* 

There was a kind of ale called in Cormac's Glossary 
and elsewhere brocoit, bracaut, or braccat, which, although 
made and named from brae or malt, was somewhat different 
from the ordinary cuirm or ale. Cormac (p. 19). says that 
brocoit is a Welsh word. It has descended to our day, 
and is found in English dictionaries as bragget, used to 
designate a sort of ale sweetened with honey and seasoned 
with spices. The Glossary states that " braccat is a goodly 
ale [or goodly drink : sain-linn], made from malt " [and 
honey] : and we know that honey was also used in making 
the Welsh bragget. This kind of ale is often mentioned in 
later Irish writings under the name of brogoit. 

Mead or metheglin (Irish mid, pron. mee) was made 
chiefly from honey : it was a drink in much request, and 
was considered a delicacy, which is indicated in the designa- 
tion applied to it in the " Vision of Mac Conglinne " (p. 98) — 
the " Dainty drink of Nobles " (sercoll sochenelach) . It was 
intoxicating, though not so much so as ale : the O'Caith- 
niadhs are spoken of in " Hy Fiachrach " as " the host who 
are most excited by mead." Where mead abounded the 
people of the district were the more thought of for it ; so 
that we often meet with such laudatory expressions as 
" the mead-abounding murrisk," " the O'Gillens who have 
encouraged mead-drinking," " the mead-drinking men of 
Meath." A visitor on arrival was often treated to mead : 
when the king of Leinster came to visit St. Brigit at her 
nunnery, she gave him a cup of mead to drink. Mead is 
mentioned, in our most ancient authorities. The rule of 
St. Ailbe, a contemporary of St. Patrick, directs that when 
the monks sit down to dinner, they " shall get on clean 
" dishes herbs or roots washed in water, likewise apples, 
" and mead from the hive to the depth of a thumb." f In 

* Rev. Celt., xiv. 27. f Lynch, Cambr. Ev., 11. 137 


the story of the Children of Lir, Finola, speaking of their 
former happy life, says that they often drank " hazel- 
mead " (miodh cuill)* ; from which we may infer that 
hazel-nuts were sometimes used as an ingredient in making 
mead, probably to give it a flavour. Mead continued to 
be made in the south of Ireland till about the year 1824.! 

Sullivan, in his Introduction to O'Curry's Lectures 
(p. 378), states that the Irish made a kind of cider called 
nenadmim, from the wild- or crab-apple ; and that the 
people kept wild-apple trees planted in hedgerows to 
supply the fruit. He also says that they made another 
drink bearing the same name from " wood-berries," which 
are probably the berries still well known as fraechoga or 
fraechdin, anglice " froghans," the Vaccinum myrtillus, 
better known in Munster as " whorts " or " hurts." No 
doubt he had good authority for these statements, but I 
have found none. 

In old times it would appear that people consumed 
their drink in large quantities : but then it was only mildly 
intoxicating. The law-tracts assign the quantity of ale 
allowed at dinner to laymen and to clerics respectively. 
There are some contradictions and obscurities in these 
statements]: : but it seems probable that the following are 
the allowances meant : — Six pints to a layman, and three 
to a cleric. It is added that the latter were so restricted 
in order that the clerics " may not be drunk and that their 
canonical hours may not be set astray on them " : in which 
there is an implication that a man was liable to get drunk 
on six pints but not on three. 

The word beoir is used in Irish for ' beer ' — obviously 
the same as the English word. There is a late tradition 
that a kind of beer was made from heath, or from the 

* Three Sorrowful Stories, 141 ; and old Celt. Rom., 24. See also 
" a cup with hazel-nut mead " in K. Meyer's King and Hermit, p. 17. 
f For mead see also Ware, Antiqq., 183 : and Sullivan, Introd., 377. 
I Br. Laws, in., 337 and note. 


red heather-berries called tnonadan (" bog-berries " : not 
" hurts " or whortleberries), which was designated in 
English " bog-berry wine," and in Irish beoir Lochlannach, 
i.e. " Lochlann or Norse wine " : but I have not met with 
any reference to it in old Irish literature.* 

Whiskey is a comparatively modern innovation. The 
first notice of it in the Irish annals appears to be at a.d. 
1405, where there is the ominous record that Richard 
Mac Rannall, chief of Muinter Eolais, died from an over- 
dose of uisge beatha [ishke-baha] or whiskey. 

3. Cooking. 

In great houses there were professional cooks, who, 
while engaged in their work, wore a linen apron round 
them from the hips down, and a flat linen cap on the 
head.f Among ordinary families the women did the 
cooking : and in monasteries a few of the monks, specially 
skilled, were always assigned for this part of the work of 
the community. Among St. Patrick's household his cook 
was Athcen, who is still remembered as the patron saint of 
Bodoney in Tyrone. J The Irish for ' a cook ' is coic, which 
is a loan word from Latin coquus : and the Irish cucenn, ' a 
kitchen,' is from the Latin coquina : both Irish words are 
found in Cormac's Glossary (p. 31). 

Meat and fish were cooked by roasting, boiling, or 
broiling. The word inneonadh [innona] was commonly 
applied to the process of broiling or roasting, as dis- 
tinguished from fulachta, ' seething, stewing, or boiling.' 
A spit (bir) for roasting — made of iron — was an article in 
general use, and was regarded as an important household 
implement. But the spits commonly used in roasting, as 
well as the skewers for trussing up the joint, were pointed 
hazel-rods, peeled and made smooth and white. § Meat, 

* See Sullivan, Introd., p. 378. J Trip. Life, 265 : FM, a.d. 448. 

f Mac Conglinne, 62. § For spits, see Petrie, Tara, 213, 214. 


and even fish, while roasting, were often basted with honey 
or with a mixture of honey and salt* ; and it is to be 
presumed that the joint or animal was kept turning round 
simply by hand. Meat and fish were often broiled on a 
gridiron, or something in the nature of a gridiron. In a 
very ancient story in Cormac's Glossary (p. 130), we read 
that on a certain occasion Finn mac Cumail found one of 
his men, Coirpre, in an empty house, cooking fish upon an 
indeoin [innone]. There is some doubt about the exact 
meaning of indeoin in this passage. Stokes translates it 
' stone,' while O'Donovan renders it ' spit ' : but in another 
place he makes it ' gridiron.' An indeoin was no doubt a 
gridiron or griddle of some kind : probably of stone in 
primitive times, but subsequently of metal. This word 
indeoin has, however, several meanings : as may be seen 
by reference to the Index at the end. 

When bodies of men marched through the country, 
either during war or on hunting excursions, they cooked 
their meat in a large way. Keating and other writers give 
the following description of how the Fena of Erin cooked, 
a plan which is often referred to in the ancient tales, and 
which was no doubt generally followed, not only by the 
Fena but by all large parties camping out. The attendants 
roasted one part on hazel spits before immense fires of 
wood, and baked the rest on hot stones in a pit dug in the 
earth. The stones were heated in the fires. At the bottom 
of the pit the men placed a layer of these hot stones : then 
a layer of meat- joints wrapped in sedge or in hay or straw 
ropes to keep them from being burned : next another 
layer of hot stones : down on that more meat : and so on 
till the whole was disposed of, when it was covered up ; 
and in this manner it was effectively cooked. The remains 
of many of these cooking-pits are still to be seen in 
various parts of the country, and are easily recognised by 
the charred wood and blackened stones ; and sometimes 

* Mac Conglinne, 62. 


the very pits are to be seen.* To this day they are called 
fulachta-na-bhfiann [fullaghta-na-veen], the fulachta or 
' cooking-places of the Fena ' : for in popular legend they 
are still attributed to the Fena of Erin. 'These cooking- 
places are referred to in Cormac's Glossary (p. 69, " Esnad") 
where they are called fulacht-fiansae ; and it is stated that 
while the cooking was in progress the hunters chanted a 
kind of music called esnad (see vol. 1., p. 592, supra). A 
pit in which meat was cooked in this manner was called 

In the house of every chief and of every brewy there 
was at least one bronze caldron for boiling meat. Its 
usual name was coire or caire [2-syll.] : but it was some- 
times called aighean, or more correctly adhan [ey-an], 
which is now its usual name in Scotland. It was highly 
valued, as a most important article in the household ; and 
it was looked upon as the special property of the chief or 
head of the house — much in the same way as his sword 
and shield. Everywhere we meet with passages reminding 
us of the great value set on these caldrons. One of them 
was regarded as a fit present for a king. St. Patrick when 
a boy in slavery in Ireland was sold to some mariners at 
the mouth of the Boyne for two caldrons of bronze, f The 
caldron of a chief or of a brewy was supposed to be kept 
in continual use, so that food might be always ready for 
guests whenever they happened to arrive. A common 
appellation for one of these was coire ainsec, or caire ainsic, 
or caire ainsecan, the ' un-dry caldron.' It is laid down in 
the Senchus MorJ that a brewy of the highest class should 
have a caire ainsec, which is denned as " a caldron which 
•' should be always kept on the fire for every party that 
" should arrive " : and the old book goes on to give several 
derivations for ainsec, of which one — from an, a negative, 
and sic, ' dry ' — ' not dry,' ' always wet ' — is probably 

* KUk. Archaeol. Journ., 1885-6, p. 390. t Trip. Life, 417. 

+ Br. Laws, 1. 41, 47, 49. 


correct ; reminding one of the modern Irish invitation to 
" a dry bed and a wet bottle." 

Some caldrons were believed to possess magical pro- 
perties, one of which was that, whatever quantity of food 
was put into the vessel to boil, it cooked just as much 
as was sufficient for the company and no more : and when 
the attendant {luchtaire) thrust in the fleshfork to serve 
any particular individual, he always — by the same magic 
virtue — brought forth the very joint specially allotted to 
him (p. 108, supra). This virtue is alluded to in Cormac's 

Fig. 215. 

Ancient Bronze Caldron : 12 inches deep : now in National Museum : 
formed of separate pieces, beautifully riveted, the head of each rivet 
forming a conical stud or button, like the rivets of the gold gorgets and 
of some of the bronze trumpets. (From Wilde's Catalogue.) 

Glossary (p. 45, " Caire atnsic") in a derivation of the 
name : — it was called caire ainsic " because it returns 
(aisces) his right to everyone." The Welsh, too, had their 
magical caldrons, to which they attributed magical virtues 
something like those in Ireland.* 

If we are to believe the tales and other old writings, 
some caldrons were large enough to hold two or three 
sheep or hogs together, cut up into joints ; the Brehon 
Law (iv. 327) tells us that in an aire-tuisi chief's house 
there must be a caldron in which a cow and a hog will fit ; 
and in other parts of Irish literature these very large 

* Ulst. Arch. Journ., V. 85 : Mabinogion, p. 31. 

126 Social and domestic life [part hi 

caldrons are continually referred to. Many bronze cal- 
drons have been found from time to time, and are now 
preserved in the National Museum, Dublin — several of 
beautiful workmanship — but none are large enough to hold 
all the chief joints of a cow and a hog. 

All those caldrons that have been preserved have a 
pair of ears or rings at the sides by which they were hung 
over the fire on hooks ; and this use is alluded to in the 
expression sadail ar cairi da drol, " cozy our caldron on 
the hook," in the story of the cave of Ben Etair or Howth.* 
Caldrons appear to have been always made of brass or 
bronze — most often the latter. Those hitherto found are 
all of that material ; and the Brehon Law says that in 
every brewy's house there should be a cairi humai, ' a 
bronze (or brazen) caldron.' Caldrons were manufactured 
at home : but that some at least, and those among the 
most valuable, were imported, is shown by Muirchu's 
record, written in the seventh century, that Daire gave 
Patrick an aeneum mirabilem transmarinum, ' a wonderful 
brazen caldron from over sea.'f 

Accompanying every caldron was an del or fleshfork, 
for lifting out pieces of meat. On one occasion, soon 
before the Battle of Dunbolg, a.d. 598 St. Maidoc of 
Ferns, as we are told in the " Boroma," brought to Branduff, 
king of Leinster, a present of a three-pronged fleshfork 
(del-trebend) , a caldron, a shield, and a sword! : an °dd 
combination, quite characteristic of the times. But in 
early ages kitchen utensils were everywhere regarded as 
important. The inventory of the jewels of the English 
King Edward III. gives a list of this king's frying-pans, 
gridirons, spits, &c.§ A fleshfork was also called gdbal 
[goul], which was, and is now, the ordinary word for a fork 
of any kind. There is a curious provision in the Brehon 

* Rev. Celt., XI. 133. f Trip. Life, 291. 

J Rev. Celt., xm. 57 : Silva Gad., 408, 409 : Wilde, Catalogue, 529 : 

Man. & Cust., 1. 338. § Roberts' Soc. Hist., p. 318. 


Law that if any accident occurred to a bystander by the 
lifting of the joint out of the boiling caldron, the attendant 
was liable for damages unless he gave the warning : " Take 
care : here goes the ael into the caldron ! "* 

4. Flesh-meat and its accompaniments. 

The flesh of wild and domestic animals, boiled or roast 
or broiled, much as at the present day, formed one of the 
staple food-materials in old times in Ireland as in other 

Pork (muicc-fheoil, i.e. ' pig-flesh/ pron. muckole) was a 
favourite among all classes, as it was among the Greeks 
and Romans. When the fairy-king Midir tried to entice 
Befinn to Fairyland, one of the allurements he held out 
was that — among other choice viands — it had plenty of 
fresh pork (vol. I., p. 295). This preference is noticed in 
later ages by Stanyhurst. — " No meat they fansie so much 
" as porke and the fatter the better. One of John O'Nel's 
" household demanded of his fellow whether beefe were 
" better than porke : ' that ' (quoth the other) ' is as intricat 
" a question as to ask whether thou art better than O'Nele." 
And the partiality for this meat continues to the present 
day among the peasantry, but they generally eat it in the 
form of bacon. Pork was made into bacon as at present 
by being salted and hung up on the wall over the fire. 
Old bacon was considered good for chest-disease, f 

Beef, or as it was called in Irish mairt-fheoil (i.e. ' ox- 
flesh ' : pron. morthole), was much in use. The animal 
seems to have been generally killed with a spear. % The 
flesh of fattened calves, either boiled or roast, was con- 
sidered a dainty food. Mutton — in Irish caer-fhedil or 
muilt-fheoil (' sheep-flesh,' ' wether-flesh ' : pron. kairole 
and multhole) — was perhaps in more request than beef. 
Boiled mutton (muilt-bruithi, ' boiled wether ')is mentioned 
in the " Vision of Mac Conglinne " (98) as a savoury viand. 

* Br. Laws, 111. 267. t Ma c Conglinne, 98. % Br. Laws, iv. 311, l8 . 


Venison was in great favour : everywhere in the tales 
we read of hunters chasing deer artd feasting on the flesh. 
It was sometimes called fiadh-fheoil, ' deer-flesh ' [pron. 
fee-ole] : and there were other names. It was food fit for 
kings : one of the seven prerogatives of the king of Ireland 
was to receive a tribute of the milradh [milra] or venison 
of Naas.* On a certain occasion an envoy promised Finn 
mac Cumail among other choice viands the feoil daimh 
(' flesh of deer ') of Knockclare. Salted venison, which is 
sometimes called serccol-tarsain (' dainty-condiment ') is 
mentioned in the Brehon Law (iv. 309, i 7 ) as one of the 
refections due to an og-aire or ' junior chief.' Goats were 
quite as common in old times as now, and their flesh was 
as much used, as well as their milk. 

Some of the animals mentioned in the records as 
supplying food are no longer used for this purpose. That 
badgers were eaten we have certain proof. Cormac Gaileng, 
preparing a grand feast for his father Teige, held it as a 
point of honour to have the flesh of all eatable animals, 
and put himself to much trouble to procure badgers from a 
neighbouring warren, f Deirdre, when recalling the life 
she had led in Scotland, says that the sons of Usna 
brought her for food fish and venison and the flesh of 
badgers : and badger-flesh from Beare in Cork was one of 
the dainties promised on a certain occasion to Finn mac 
Cumail by Moiling the Swift. J Badgers were eaten in 
Ireland until lately. In a comic description of a wedding 
in an Irish poem by a Connaught poet named Mac Sweeny, 
of about a hundred years ago, " the badger of the glen " is 
enumerated as among the animals to be procured for the 
feast.§ A small animal named a togmall is mentioned in 
some of the oldest of the Irish tales (see chap, xxx., 
page 519, infra). O'Curry makes it a squirrel : but it 

* Book of Rights, 3, 9. 

t Joyce, Irish Names of Places, 11. 244 : see also vol. 1. 287, supra. 

X Rev. Celt., xm. 47. § Hardiman, Iar C, 286, note ; 290, , . 


appears that the squirrel is not a native Irish animal, and 
that it was introduced only in late times. About the 
togmal we only know that it was sometimes tamed as 
a pet, and as such was often carried on the shoulder — as 
was then the custom — so that it must have been small : 
and that its flesh was used as food ; f or " speckled togmalls 
from Berramain " (in Kerry) are included in the Colloquy 
as part of the food-supplies of the Fena of Erin.* 

Seals were valued chiefly for their skins, but partly also 
for their flesh as food. Adamnan mentions the seal and calls 
it by two Latin names, fhoca, and marinus vitulus (' sea- 
calf ') : the Irish name was, and is, rdn [roan]. There was 
a little rocky island near Mull where seals congregated and 
bred, and which — according to Adamnan (p. j8) — was 
regarded as a preserve belonging to Iona : but he does not 
state what use the monks made of the animals : probably 
for the skins. We have direct evidence however that seals 
were anciently used as food in Ireland. In the Book of 
Lismore it is related that the seven bishops of Tulla in 
the east of Leinster (near Killiney) came on a visit to 
St. Brigit, on which she sent one of her people to sea to 
fish. This man succeeded in spearing a seal, which, with 
some difficulty, he brought home for the use of the visitors, f 
The flesh of seals is now seldom used as food, for which 
— at least in parts of the west — there is a very good reason. 
There is a legend that at some former time several members 
of a certain family were metamorphosed into seals, so that 
a latter-day member who sat down to a dinner of seal could 
never be quite sure that he was not feasting on his own 
great-great-great-grandfather. :{: Martin, in his description 
of the Hebrides in 1703 (p. 64), says that seals were eaten 
by the meaner people, who salted the flesh with " burnt 
sea- ware." The higher classes however ate only the hams. 

* Silva Gad., 119 (Irish text, no,,,), 

■f Stokes, Lives of SS., 196 : see also Hardiman in Iar C., 95. 

t Iar C., 27, t ; 95, *■ 



Corned meat was everywhere in use. A dead pig 
salted was usually called Untie [tin'ne] : but this word 
was also often applied to a salted joint of any animal. 
A number of whole -pig-tinnes commonly formed part of 
the tribute paid to a superior king or chief.* The word 
saill or saille [sal, sal-le] from sal, ' salt,' was applied to 
any sort of salted meat : and it is still in use in this sense. 

Besides the main joints boiled or roast, we find mention 
of various preparations of the flesh of animals, mixed up 
with many ingredients. A pottage or hash formed of 
meat chopped up small, mixed with vegetables, was called 
craibechan [craiv'ahan]. We find it stated in an Irish 
document that Esau sold his birthright to Jacob for a 
craibechan] : and elsewhere the term is defined " fine or 
small meat." In the " Vision of Mac Conglinne " (p. 34) is 
mentioned as a dainty food " sprouty craibechan with 
purple-berries " : " sprouty," i.e. mixed with vegetable 
sprouts. The " purple-berries " were probably the quicken- 
berries or rowan-berries added to give a flavour. There 
are several other terms used to designate meat-prepara- 
tions of this kind : such as brothchdn (a dim. of broth), 
follach, and scaiblin. No doubt each of these pointed to 
some special mode of preparation : but the distinction — if 
it ever existed — is now lost. 

Simple broth or meat- juice without any mixture of 
minced-meat was called by several names : — beochail, 
bruth, broth, and enbruithe. This last, which is still a 
living word for broth, is given in Cormac's Glossary ; and 
is there said to be derived from en, ' water,' and bruith, 
' flesh,' i.e. ' water of flesh,' a natural interpretation. The 
Irish broth or bruth, which is also a living word, is the same 
as the English broth. In later times broth was a favourite 
with the Irish, and also among the Scottish Highlanders, 
as is noticed in " The Fair Maid of Perth " : — " The hooped 
" cogues [Gaelic cuach : see p. 74, supra] or cups, out of 

* Br. Laws, n. 201, bottom. f Trip. Life, Introd., xviii. 12. 


" which the guests quaffed their liquor, as also the broth or 
" juice of the meat, which was held a delicacy." 

Sausages or puddings were a favourite dish, made 
much the same as at the present day, by filling the intestines 
of a pig, cow, or sheep with minced-meat and blood. 
They were known by the terms indrechtan and mar 6c, the 
latter of which (spelled marog) is still in use among the 
Highland Scotch.* In O'Clery's Glossary innreachtan is 
given as equivalent to putog, ' a pudding.' In the " Vision 
of Mac Conglinne " (p. 88) we find mention, as a delicacy, 
of indrechtana finda bo-bdn-methi , ' white-coloured puddings 
of white fat cows.' Puddings and sausages got a boil after 
making, so as to half cook them, and were then put aside 
till wanted : when about to be brought to table they were 
fried and served hot as at the present day. Accordingly 
in the same piece (p. 66) we find mentioned maroca arna 
cetberbad, ' puddings first-boiled,' i.e. having got a boil 
after making. The belly of a pig, called tarr, when 
properly cleaned and boiled, was much in use, but was 
regarded as rather an inferior meat food.f 

In the " Vision of Mac Conglinne " is mentioned, as good 
food, the dressan of an old wether : and Kuno Meyer, the 
editor, on the authority of an ancient Glossary, translates 
dressan, ' the spleen.' The word is a diminutive of dress 
or driss, which is familiarly applied to things of a branchy 
nature, such as a bramble or the smaller intestines : and as 
applied to an article of food is still in use in Cork in the 
form of drisheen, which has the Irish diminutive in instead 
of the an of Mac Conglinne. The name drisheen is now 
used in Cork as an English word, to denote a sort of 
pudding made of the narrow intestine of a sheep, filled 
with blood that has been cleared of the red colouring 
matter, and mixed with meal and some other ingredients. 
So far as I know, this viand and its name are peculiar to 
Cork, where drisheen is considered suitable for persons 

* Mac Conglinne, 32, 66, 86. t Rev. Celt., v. 252. 


of weak or delicate digestion. The fact that the word 
drisheen (old form drisin) is now used in this sense makes 
it probable that the other diminutive dressan was used in 
a wider sense than that of ' spleen,' as given in the above- 
mentioned Glossary. Perhaps the drisheen of Cork is the 
same as the dressan of Mac Conglinne. 

Sometimes the gullet (lonlongin) of an ox was filled 
with minced-meat and cooked like a sausage : and this 
appears to have been regarded as a delicacy, for it is 
designated in " Mac Conglinne's Vision" (98, i 5 ) " the choice 
" easily discussed thing for which the hosts contend — the 
" gullet of salted beef " (Idnlongin bdshaille). The contents 
only were eaten, not the enveloping gullet, which was 
not fit for food. Tripe, whether of pig or sheep, was 
designated by the word caeldn* (kailaun], which means 
something slender, ' a slender gut ' (from cael, ' slender '). 
In Cormac's Glossary (p. 44), it is stated that the coelan 
or small gut is " the slenderest thing in the body " : and 
farther on in the Glossary the word innbi is found as 
another name for a caeldn. Spenser says that in his time 
the northern Irish used to draw the blood of living beasts 
" to make meat [i.e. puddings] thereof " : but I find no 
mention of this custom in old Irish literature. 

Lard was known by three names, geir, usca, and blonog, 
which last is the word now in use. It was much used as an 
annlann or condiment, and entered into cooking in various 
forms. It is very often mentioned in " Mac Conglinne's 
Vision," showing how much it was in request. The Culdee 
monks were allowed lard on festival-days, f We also find 
mention of olar, ' rich gravy ' ; and of inmar, ' dripping,' 
both used as a condiment or relish. J 

Most of the birds used for food at the present day were 
eaten in old times : and frequent allusions to birds as food 
are found in ancient Irish writings. Among the food of 

* Mac Conglinne, 38, 88, 98. f Reeves, Culdees, 85. 

{ Mac Conglinne, 32. 


the Fena are enumerated " birds out of the trackless oak 
woods " : " wood-cocks (cailig fheda) out of [the wood of] 
Fidrinne " : and " speckled nests from the mountain 
peaks."* This last entry shows that they made some 
use of birds' nests in cooking : but how is not known : for 
we have in Ireland no edible nests. Stokes (Acall. 279) 
throws out the suggestion that the nests may have been 
used to make fires as a charm : just as in India milk- 
porridge boiled on a fire of birds' nests was used as a 
charm against certain evil spirits. Giraldus Cambrensis 
says that the Irish loathed the flesh of the heron ; but that 
Henry II. induced those kings and chiefs he entertained in 
Dublin at Christmas, 1171, to taste it. They do not seem 
to have much relished it : for ever since that time the Irish 
people have let the herons alone. 

Eggs were extensively used : they seem to have been 
often boiled hard and eaten cold. One of the relaxations 
allowed to Culdee monks on festivals was " a dry [i.e. a 
hard] egg."f Goose-eggs, if we are to judge from their 
frequent mention, were a favourite. In a legendary 
account of Bishop Ere of Slane given in the " Feast of 
Dun-nan-gedh," we are told that he kept a flock of geese 
to lay eggs for him, and that his dinner every evening was 
" an egg and a half and three sprigs of the cresses of the 
Boyne."| At the great banquet itself, some of these eggs 
were on the table, cold, and Congal, going in to view the 
feast, ate a part of one. And when the company sat down, 
a goose-egg [cold] on a silver dish was placed before each 
chief. § From all this we may infer that the eggs were 
boiled hard. 

All the fish used for food at the present day were eaten 
in Ireland in old times, so that there is no need to go into 
details. Only it may be remarked that salmon was then 
the favourite ; and we meet with constant reference to it as 

* Silva Gad., 119. J Moyrath, 19. 

f Reeves, Culdees, 85. § Ibid., 25, 29. 


superior to all other fish. The salmon of the " salmon-full 
Boyne," of Linnmhuine or Lough Neagh, and of the 
Barrow, were much prized. The subject of fishing, will 
be treated of in chapter xxix., p. 472, below. 

Any viand eaten with the principal part of the meal as 
an accompaniment or condiment, or kitchen as it is called 
in Ireland and Scotland — anything taken as a relish with 
more solid food — was designated by one of the words 
annlann, tarsunn, ionmar, all equivalent to the Latin 
obsonium. The Brehon Laws, when setting forth the 
refections legally due to various classes of persons, specify 
the tarsunns with much particularity : — butter, salt, bacon, 
lard, salt meat of any kind (when used in small quantities 
and not the principal part of the meal), honey, kale, onions, 
and other vegetables, &c. Thus in one place (iv. 119) we 
find mentioned " three cakes with their annlann of butter 
or bacon," as the fine for a hen's trespass in a garden. 
According to the Rule of the Culdees, while they could 
not increase the quantity of bread on festival-days, they 
were allowed the use of various annlanns such as kale, 
apples, &c* 

Salt — Irish sal, salann — was used for domestic purposes 
much the same as at the present day — for corning various 
kinds of provisions, especially butter, pork, and beef, and 
at meals with all viands requiring it. It was not so easily 
made or procured then as now, so that the supply was 
limited, and people kept it carefully, avoiding waste. In 
rich people's houses it was kept in small sacks. In the Life 
of St. Senan it is related that on a certain occasion the 
saint sent, as a present to St. Brigit of Cluain Infide, a 
basket containing certain articles, among which were two 
masses of salt (da cloich t-salainn, ' two stones of salt '), one 
for herself and the other for St. Diarmait of Inis Cleraun 
in Lough Ree.f The Senchus Mor mentions salt as one 
of the important articles in the house of a brewy, on which 

* Reeves, Culdees, 84. f Stokes, Lives of SS., line 2408. 

CHAP. XXl] fOOt), FtJEL, Attt> LIGHT f 35 

the glossator remarks that it is " an article of necessity at 
all times, a thing which everyone desires "* : and in confir- 
mation of this we find, in the story of the " Voyage of the 
O'Corras," that the house of the rich brewy Conall Derg 
O'Corra was never without certain plentiful supplies, 
among them a sack of salt (miach salainn) " to make each 
food taste well."f It was kept in lumps or in coarse 
grains ; and at dinner each person was served with as 
much as he needed. In the sixteenth century in England 
— as we are told by Roberts — each guest at dinner was 
given a little lump of salt, which he ground into powder 
with the bottom of his glass or drinking-goblet : and 
something of the same plan may have been followed in 
Ireland. English salt was largely imported, and was 
considered the best. Mac Conglinne (p. 60), when calling 
for a number of viands specially delicious, has among 
them " English salt (salann saxanach) on a beautiful 
polished dish of white silver." In this last point the Irish 
accounts are corroborated by an English authority of a 
later time, Higden's " Polychronicon," which mentions the 
export of salt from England to Ireland : — " Also Flanders 
" loveth the wolle of this lond [England], Ireland the oor 
" [ore] and the salt."! But there were at home professional 
salt-makers, as we find by a passage in the story of the 
Tromdamh (119), where it is related how a ship's crew 
from Ireland meets on the coast of the Isle of Man (then 
occupied by the Irish) a person who was every alternate 
year a maker of salt. At a much later time, a.d. 1300, 
salt was exported from Ireland, as we know from the fact 
that it was one of the commodities sent to Scotland to 
supply the army of Edward I. (for which see p. 433, infra). 
The salt must have been manufactured either from sea- 
water, or from rock-salt taken from the earth, or more 

* Br. Laws, 1. 127, 143. 

f Stokes, Rev. Celt., xiv, 27 ; Joyce, Old Celtic Romances, 401. 

X Quoted by Kuno Meyer in Mac Conglinne, 142. 


likely from both. For, according to Kinahan,* there are 
plenty of salt deposits in Ulster : and we have seen above 
how St. Senan sent to his friend a present of " two stones 
(or rocks) of salt." But of salt mines, or of the mode of 
preparing the salt, the ancient literature — so far as I know 
— contains no details. The word salanndn (a dim. of 
salann) is still used in the Irish language to signify a 

5. Milk and Us products. 

There are several ancient Irish words for milk, three of 
which are ass, loim, and melg or melc : this last evidently 
cognate with Latin mulgeo and with English milk ; and 
from it is derived the old verb omalgg ("mulxi;" Zeuss,6i, 15). 
The most general word in modern use is bainne [bon-ya], 
which is also an ancient word. Another old word for 
milk, according to Cormac's Glossary (149), is siig, whence 
comes sigamlae, ' milkiness,' "as in the saying of the 
Bretha Nemed : — ' a cow is estimated by her segamlae.' " 

The milk chiefly used in Ireland was that of cows ; 
but goats' and sheep's milk was also in much request. Deer's 
milk was sometimes made use of, and the milking of the doe 
is often mentioned in the records, but always in a manner 
implying that it was exceptional. In the Tripartite Life (73) 
we read that St. Patrick left at Ath-da-laarg (the ancient 
abbey of Boyle in Roscommon) three brothers with their 
sister Cathnea : " She it is that used to milk the hinds " 
(eillti : sing, eillit, ' a hind '). Nia Segamain, who was king 
of Ireland, a.m. 4887, was so called because " during his 
time cows and does were milked alike : and it is for him 
that does were cows." This it seems was effected through 
the incantations of his mother, who was an enchantress, f 

Milk was used in a variety of ways, as at the present 
day. For drinking, the choice condition was as new milk 

* Geology of Ireland, 358, 359. 

f C6ir Anmann in Ir. Texte, m. 295 : FM, a.m. 4887 : Keating, 260. 
For other instances of milking does, see Trip. Life, 233 and Dr. Healy, 211. 


(lernnacht or lemlacht) : and cream was sometimes added 
as a luxury. But skimmed milk, i.e. milk slightly sour, 
and commonly thick, from which the cream had been 
skimmed off, was considered a good drink. This was 
called draumce and also bldthach [draumke, blawhagh], 
which last word is the name used at the present day. 
Thick milk was improved by mixing new milk with it* 
as I have often seen done in our own day. 

The people made butter (Irish im or imtri) in the 
usual way, in a small churn : the churn has been already 
described (p. 75, supra). The process of churning was 
called maistred. In the description of an 
imaginary house, all made of choice viands, 
in "Mac Conglinne's Vision" (p. 92), is men- 
tioned " a pure-white bed-tick of butter," 
from which it may be inferred that a 
whitish colour was a mark of good fresh 

Butter of any kind was considered a 

J Fig. 216. 

superior sort of condiment. Salt butter Ancient Butter . prtnt 
was called gruiten and sometimes grusden. °[ *; «f *£%£* 
In Cormac's Glossary (86) gruiten is %££&£?"" 
derived from groit, ' bitter,' and sen, ' old.' 
Its inferiority to fresh butter is brought out clearly in 
the Brehon Law provision (II. 149) that, in fosterage, 
the sons of farmers are to have gruiten with their 
stirabout, the sons of chiefs fresh butter, and the sons 
of kings honey. A lump of butter shaped according 
to fancy was called a mescan, a word given in Cormac's 
Glossary (116), where it is stated that a mescan was 
so called because it was produced by the mescad or 
agitation of the milk. This word is still in very general 
use even among the English-speaking people, who pro- 
nounce it miscaun or miscan. A earn on a mountain top 
is sometimes called a m-iscaun from its shape, as, for 
* Br. Laws, iv. 303, 4 from bottom : and 306, ]0 . 

1 3 8 



example, Miscaun Maive on the top of Knocknarea hill 
near Sligo. Another name for a roll or miscaun of butter 
was brechtan : a brechtan tir-imme, a ' roll of fresh butter,' 
was portion of the viands procured for the May-day feast 
according to an old Irish poem*: and in O'Clery's Glossary 
brechtan is explained 'wheat,' and also im tir, 'fresh butter.' 
But brechtan was also applied to a viand like what we call 
a custard, made of flour, milk, and perhaps eggs, sweetened 
with honey : a brechtan cruthnechta (of 
wheat) is mentioned in " Mac Conglinne's 
Vision" (123, 21). 

In later times it was customary to sink 
\ J\ 1 butter deep down in bogs, closed up in 

y\ casks or wrapped up in cloths, to give it 

a flavour, or, as some think, as a mode of 
preserving it.f Among the food of the 
Irish, Dineley (A.D. 1675) mentions butter 
" mixed with store of ... a kind of garlick, 
''and buried for some time in a bog to 
" make a provision of an high taste for 
" Lent." Sir William Petty also mentions 
butter made rancid by keeping in bogs ; 
and other authorities to the same effect 
might be quoted. Whether this custom 
existed in ancient times I am unable to 
say ; but at any rate, its prevalence, even 
at this late period, is a sufficient explanation of the fact 
that butter is now very often found in vessels of various 
shapes and sizes, deeply embedded in bogs ; sometimes in 
firkins not very different from those now in use.+ Several 
specimens of this " bog butter," as it is commonly called, 
are to be seen in the National Museum. In all cases the 

*Sick Bed, Atlantis, I. 271. 

t Kev. James O'Laverty, in Kilk. Arch. Journ., i8q2, p. 356, thinks so, 
and advances good reasons for bis opinion. See Sullivan, Introd., 367 ; and 
the authorities referred to by Wood-Martin, Pagan Ireland, 601. 

J Wilde, Catalogue, p. 212. 

Fig. 217. 

A firkin of Bog-butter 
26 inches high : made 
from a single piece of 
sallow. Top and bottom 
and part of side of firkin, 
with butter inside, still 
remain. In the National 
Museum. (From Wilde's 
Catalogue, p. 212.) 


butter is found to be changed, by the action of the bog 
water, into a greyish cheese-like substance, partially 
hardened, not much like butter, and quite free from 

Curds — called in Irish gruth [gruh] — formed one im- 
portant article of diet. Milk was converted into curds and 
whey by calves' rennet, Irish Unit, so called, according to 
the fanciful derivation in Cormac's Glossary (p. 20), because 
" it strikes (benait) in milk till it [the milk] is thick and 
coagulated."* A light collation of curds, butter, and milk, 
flavoured with crem or wild garlic, was known by the names 
samit or sam-ith and crimmes. It was well recognised as 
a summer dainty, for the wild garlic grows only in summer. 
The second name crimmes means ' wild-garlic harvest,' from 
crim, another form of crem, and mes, ' harvest or produce." 

Curd was converted into cheese of various sorts, which 
was greatly valued as an article of food. Cheese was 
denoted by several different words, of which the most 
common were cdisse or cdise [cawsha], and maethail [maihil]: 
but this last word was often applied to dried curd. Cheese 
was made from curd as now, by pressing in a mould, from 
which it was turned out in firm shapes. Curds were much 
used in an intermediate stage, not quite turned into cheese, 
but sufficiently pressed to squeeze out all the whey, so as 
to form a mass moderately firm and capable of keeping for 
a long time. In this state curd was a well-recognised food : 
in the " Circuit of Muirchertach of the Leather Cloaks " 
(p. 55) we are told that on a certain occasion he contributed 
to a banquet, among many other supplies, three score vats 
of curds (gruidh). This soft material, half curd, half 
cheese, was often called milsen, which O'Clery in his 
Glossary explains by gruth-caisse, or ' curd-cheese.' It 
was also called maethail, and mulchan, words understood 
to mean soft unpressed cheese, f This sort of food was 

* For Curds, see Sullivan, Introd., 368 ; and Reeves, Culdees, 85. middle. 
I Conn. Gloss., p. 117: Mac Conglinne, 80, 189. 


often given as rent or tribute*: thus we are told in the 
Tripartite Life (p. 15) that the steward of a certain king 
came to St. Patrick's foster-mother for tribute of curd 
(gruth) and butter. 

Cheese pressed tightly in a mould, and turned out very 
hard, was called tanag. Its hardness is illustrated in one 
version of the death of Queen Maive of Connaught in her 
old age : — That her nephew Furbaide, who had a grudge 
against her, catching sight of her one day in some distance, 
put into his sling a piece of tanag that he happened to be 
eating, as he had no stone, and flung it with so true an 
aim that he struck her on the forehead and killed her on 
the spot.* Masses of cheese have been found in bogs, 
of which some specimens may be seen in the National 

Whey— Irish midg [maig] — was made use of ; but it 
was considered a poor drink, so much so that it was in 
constant use among monks as a fasting beverage. Mac 
Conglinne, grumbling at the beggarly reception he got in 
Cork monastery, complains that they gave him nothing 
but the whey-water (medg-usci) of the church to drink. 
New milk from a cow that had just calved, now called 
beestings, was in Old Irish called nus, a word still in use, 
which in Cormac's Glossary (126) is derived — probably 
erroneously — from Latin novus, ' new.' This milk was 
not fit for drinking ; but it was turned into curds and 
whey by merely heating, and in this form it was used as 
food. But more often the curd was made into thin pan- 
cakes. It was evidently valued — as it is at the present 
day — for one of the blessings brought on the country by 
Cormac Mac Art's benign reign, was that the cows after 
calving had their udders full of mis or beestings, f 

Milk and all food-preparations from it, such as curd, 
cheese, butter — as distinguished from flesh-meat — were 
called ban-bid [bawn-bee], ' white-meat.' They were con- 

* LL, 125, a, ,,. f Silva Gad., 97, 7 (Ir. Text, 90, a ). 


sidered inferior in nutritive qualities to flesh-meat : and 
they were often permitted — and are permitted still — by 
the Roman Catholic ecclesiastical authorities, on fast 
days, when flesh-meat is forbidden. Mac Conglinne 
mentions the constant consumption of white-meats as one 
of the causes of his low condition of health. 

6. Com and its preparations. 

It will be seen in chapter xxiii., sect. 2 (pp. 271, 272, 
below), that all the various kinds of grain cultivated at the 
present day were in use in ancient Ireland. Corn was ground 
and sifted into coarse and fine, i.e. into meal and flour, 
which were commonly kept in chests.* The staple food of 
the great mass of the people was porridge, or as it is now 
called in Ireland, stirabout, made of meal (Irish miri), 
generally oatmeal. It was eaten with honey, butter, or 
milk, as an annlann or condiment. So well was it under- 
stood, even in foreign countries, that stirabout was almost 
the universal food in Ireland — a sort of characteristic of 
the country and its people — that St. Jerome takes occasion 
to refer to the custom in a letter directed against an Irish 
adversary, generally believed to be the celebrated heresi- 
arch Celestius, the disciple of Pelagius. Jerome could 
use tongue and pen in hearty abuse like any ordinary poor 
sinner : and he speaks revilingly of Celestius, who was a 
corpulent man, as " a great fool of a fellow swelled out 
with Irish stirabout. "f 

The common word for stirabout was, and still is, littiu, 
modern leite, gen. leitenn [letthe, letthen] ; but in the 
Brehon Laws and elsewhere it is often called gruss. Gruel 
was called menadach : it is mentioned as part of the fasting- 
fare of the Culdees.J The Senchus Mor annotator, laying 
down the regulations for the food of children in fosterage,§ 

* Todd, Book of Fermoy, 17. 

f Todd, St. Patrick, 190, 191 : Lanigan, 1. 17 : FM, vol. 1., Pref. li. 

J Reeves, Culdees, 86. § Br. Laws, 11. 149 and note ; 151 and note. 


mentions three kinds of leite" or stirabout : — of oatmeal, 
wheatmeal, and barleymeal : that made from oatmeal 
being the most general. Wheatmeal stirabout was con- 
sidered the best : that of barleymeal was inferior to the 
others. For the rich classes, stirabout was often made on 
new milk : if sheep's milk, so much the better, as this was 
looked upon as a delicacy.* Finn-leite, ' white-stirabout,' 
i.e. made on new milk, is designated by an epicure, f in an 
exaggerated strain — " the treasure that is smoothest and 
sweetest of all food " : it was eaten with honey, fresh 
butter, or new milk. For the poorer classes stirabout was 
made on water or buttermilk, and eaten with sour milk or 
salt butter : but butter of any kind was more or less of a 
luxury. All young persons in fosterage were to be fed, 
up to a certain age, on stirabout, the quality and condi- 
ment (as distinguished above) being regulated according 
to the rank of the parents. J 

All the various kinds of meal and flour were baked 
into cakes or loaves of different shapes. The usual word 
for a cake was bairgen, now pronounced borreen : hence 
borreen-brack, ' speckled cake ' (speckled with currants and 
raisins), eaten on November eve, now often written barn- 
brack, sometimes corrupted to barm-brack. Flour was 
usually mixed with water to make dough : but bread made 
of flour and milk was also much in use. Honey was often 
kneaded up with cakes as a delicacy : and occasionally the 
roe of a salmon was similarly used.§ The word tort was 
applied to a cake, or to a loaf of bread of any shape ; 
whence the diminutive tortine [torteena], ' a little cake '|| : 
connected with Lat. torta : Span, tortilla. 

By a curious custom, often referred to in the Brehon 
Laws, what was called a " cake of man-baking " (bairgen 

* Mac Conglinne, 32, ^ Introduction, 365, note : and 

f Ibid., 98, 29 Tromdamh, 73. 

I Br. Laws, 11. 151, top ; 177. || Corm. Gloss., 156. 

§ Fled Brie, 9 bottom : Sullivan. 


fer-fuine) was twice the size of a " cake of woman-baking." 
" Three cakes of man-baking," says the Brehon Law, " are 
the equivalent of six cakes of woman-baking " ; and this 
proportion is given in many parts of the law.* Accord- 
ingly the Crith Gabhlach, when setting forth the legal 
allowances of an og-aire chief, includes among them 
either one cake of man-baking or two of woman-baking, f 
Perhaps the meaning of these terms is that the larger cake 
was considered as a meal for a man, and the smaller for a 
woman. There is something like a confirmation of this 
conjecture in the " Small Primer," where it is stated that 
under certain circumstances each man of a company is 
allowed two cakes of men's baking per night as his refec- 
tion : i.e. presumably supper (or dinner) and breakfast. J 
If this is so, the allowance was liberal : for the Senchus 
Mor states in one place that a cake of woman-baking was 
two " fists " or ten inches in width, and one fist or five 
inches thick.§ Wheaten bread was considered the best, 
as at present : barley-bread was poor. St. Finntan, the 
son of Gaibrene, never ate anything but " woody bread of 
barley," and a drink of muddy water. || 

We have seen that yeast, or barm, or leaven, was used 
in brewing. That it was used also in baking appears 
from the fact that in this application there was a native 
word (descaid) for it, as well as from an eighth-century 
commentary on 1 Corinth, v. 7, 8, written in the Irish 
language by some Irish writer, in which the use of descaid 
or leaven in souring dough is spoken of in such a manner 
as to show that the writer was quite familiar with the 
process. ^ 

The several utensils used in making and baking bread 
are set forth in the Senchus Mor ; and baking and the 

* Br. Laws, n. 177, 24 ; iv. 119, 9 . || Feilire, 52. 

f Ibid., iv. 307, IS: see also v. 31 If Zeuss, 777, 28 to 32 . Stokes and 

and 47. Strachan, Thesaurus, 1 552. Here 

% Ibid., v. 47, in. the word for sour is serb ; modern 

§ Ibid., 11. 255, 8 . Irish, searbh fsharrav]. 


implements employed therein are always spoken of as 
specially pertaining to women.* The woman had a 
criathar [criher] or sieve for separating the fine part of the 
flour from the coarse, which was done on each particular 
occasion just before baking. f Having made the flour into 
dough (Irish toes, now written taos), she worked it into 
cakes on a losat [losset] or kneading-trough — sometimes 
also called lethech% — a shallow wooden trough, such as we 
see used for making cakes at the present day. The cake 
was baked on a griddle of some kind, which was called 
lee or lec-juine.% Lee signifies a flat flagstone : lec-fuine, 
' flag of baking ' : which shows that whatever may have 
been the griddle or baker in later times, it was originally 
a lee or flagstone, heated to the proper degree. In some 
Irish Glosses, lapisfulta is explained by lec-an-ardin, ' the 
flag or griddle of the bread. '|| And O'Davoren, as quoted 
by Stokes, has " cert-fuine, i.e. the stone {lee) on which 
cooking is done."^f In the " Courtship of Emer " in LU 
(123, b, 9 from bot.), it is called lee or flagstone, without 
any qualifying epithet : Cormac's Glossary (103) calls it 
lecc, which is the same word. We know that in Ireland, 
down to the time of Elizabeth, cakes were sometimes 
baked on a hot stone.** A common metal griddle was 
usually called gretel or greidel ; and sometimes lann, which 
however means any thin plate of metal. 

7. Honey. 

Before entering on the consideration of honey as food, 
it will be proper to make a few observations on the 
management of bees by the ancient Irish. From the 
earliest times Ireland was noted for its abundance of 
honey. A foreign writer, Solinus, who lived in the third 

* Br. Laws, 1. 123, 149. || Stokes in Conn. Glossary, 103 

t Miss Hull, Cuch. Saga, 97. (" Lecc "). 

X Corm. Gloss., 102. If Corm. Gloss., xn. 5. 

§ Br. Laws, in. 275 ; iv. 11, 9 from ** Tribes of Ireland, 51 and note 2. 


century, says that there were no bees in Ireland : but in 
this he was undoubtedly misinformed, as he was in many 
other particulars regarding this country.* Giraldus ex- 
presses the curious opinion that honey would be still more 
abundant in Ireland if the bee-swarms were not checked 
by the bitter and poisonous yews with which the woods 

The management of bees was universally understood, 
and every comfortable householder kept hives in his 
garden. Wild bees, too, swarmed everywhere — much 
more plentifully than at present, on account of the extent 
of woodland. Before cane-sugar came into general use 
— sixteenth century — the bee industry was considered 
very important, so that a special section of the Brehon 
Laws is devoted to it. The Irish name for a bee was bech 
or beach : a swarm was called saithe [saeha]. The hive 
was known by various names, such as cliabh [cleeve], 
which means ' basket ' ; bechdin, ' little bee-house ' ; and 
cesach-bech, ' bee-basket ' : but the name now universally 
in use is corcog. A honeycomb was called criathar 
[criher], literally a ' sieve.' Hives stocked with bees were 
sometimes given as part of a tribute to a king.f 

The Brehon Law tract on " Bee-judgments," of which 
the printed Irish text occupies twenty pages (of vol. iv.), 
enters into much detail concerning the rights of the various 
parties concerned, to swarms, hives, nests, and honey : of 
which a few examples are given here. If a man found a 
swarm in the faithche or green surrounding and belonging 
to a house : one-fourth of the produce to the end of a year 
was due to the finder, the remaining three-fourths to the 
owner of the house. If he found them in a tree growing 
in a faithche or green : one-half produce for a year to the 
finder : the rest to the owner. If they were found in land 
which was not a green : one-third to the finder and two- 

* See Keating, Pref., xxiv : and Gir. Cambr., Top. Hib., 1. v. 
f Book of Rights, 245, 3rd verse. 



thirds to the owner of -the land. If found in waste land 
not belonging to an individual, but the common property 
of the tribe, bees and honey belonged to the finder, except 
one-ninth to the chief of the tribe. As the bees owned by 
an individual gathered their honey from the surrounding 
district, the owners of the four adjacent farms were entitled 
to a certain small proportion of the honey : and after the 
third year each was entitled to a swarm. If bees belonging 
to one man swarmed in the land of another, the produce 
was divided in certain proportions between the two. It is 
mentioned in " Bee- judgments " that a sheet was sometimes 
spread out that a swarm might alight and rest on it : as is 
often done now. At the time of gathering the honey the 
bees were smothered. The Senchus M6r prescribes a 
penalty for stealing bee-hives with their bees. 

The Book of Aicill has a long enumeration of injuries 
done by bee-stings to men and animals, with the corre- 
sponding fines, and also the fines for killing bees. Great 
care is exhibited on the one hand to protect bees from 
wanton or unnecessary destruction, and on the other to 
provide compensation for men and animals injured by 
their stings* : but some of the provisions are so minute 
and trifling that we may doubt if they were ever seriously 
intended to be carried into practice. The whole article 
however shows that the subject of bees and bee-culture 
much occupied the attention of the public. 

One of the circumstances indicating the great plenty of 
honey in historic times in Ireland is the large size of the 
vessels sometimes used in measuring it, as instanced in 
chap, xxvii. (p. 376, below). It was used with most kinds 
of food, sometimes mixed and sometimes separately as a 
condiment. In the Book of Aicill the penalty for a 
certain class of offences is laid down as " a full meal of 
honey "f : taken of course with other food. 

* Br. Laws, in. 433-441. 

•f Ibid., in. 433, I2 from bottom. 


In the tale of the Feast of Bricriu (p. 9) we are told 

that in a certain house, among other choice viands, were 

one hundred wheaten cakes kneaded up with honey. A 

mixture of milk and honey was sometimes drunk : the 

Culdee monks were allowed to drink thick milk mixed 

with honey on the eves of Christmas and Easter* : a drink 

which would be hardly relished nowadays by either monk 

or layman. A mixture of lard and honey was sometimes 

used as a condiment, f When the gluttonous Munster King 

Cathal — eighth century — was cured of the craes Ion or 

wolf in his stomach, he was ordered to get one more good 

meal before toning down to his natural appetite. So they 

boiled, according to directions, a mixture of new milk, 

fresh butter, and honey, in a great caldron, of which he 

drank a prodigious quantity ; and " that was the last great 

" bellyful that Cathal took under the influence of the 

" glutton-demon."! After this he fell asleep, and woke 

up well. In another part of the same story (p. 78) is 

mentioned brechtan- jo-mil, some sort of custard mixed 

with honey, with probably flour and eggs. 

Honey was sometimes brought to table pure, and 
sometimes in the comb.§ Often at meals each person had 
placed before him on the table a little dish, sometimes of 
silver, filled with honey ; and each morsel whether of meat, 
fish, or bread was dipped into it before being conveyed to 
the mouth. || Stirabout was very generally eaten in the 
same way with honey as a delicacy. Honey was used to 
baste meat while roasting, as well as salmon while broiling. 
In the " Tain bo Fraich " (p. 153) we read that Ailill and 
Maive, king and queen of Connaught, had a salmon broiled 
for the young chief, Fraech, which was basted with honey 
that had been " well made by their daughter, the Princess 
Findabair " : from which again we learn that the highest 

* Reeves, Culdees, 84, top % Mac Conglinne, 106. 

f Mac Conglinne, 90, 7 from § Ibid., 60. 

bottom. || Ibid., 64, 8 . 


persons sometimes employed themselves in preparing 
honey. It has been already stated that honey was the 
chief ingredient in mead ; and it is probable that it was 
used in greater quantity in this way than in any other. 

8. Vegetables and Fruit. 

Table vegetables of various kinds were cultivated in 
an enclosure called lubgort [loo-ort], i.e. ' herb-garden ' or 
kitchen-garden : from lub, ' an herb,' and gort, a fenced-in 
cultivated plot. The manner in which the kitchen-garden 
is mentioned in literature of all kinds — lay, ecclesiastical, 
and legal — shows that it was a common appanage to a 
homestead. We find it often noticed in the Book of 
Armagh (eighth century) : and in the eighth-century 
glosses of Zeuss (37, 2 s) lubgartdir — which is still the 
common word for a gardener, and pronounced looartore — 
explains the Latin word olitor, a ' kitchen-gardener.' The 
word lub, now spelled luibh [luv or liv], glosses frutex in 
Zeuss, and is cognate with the English leaf. Another and 
more usual Irish word for herb is lus : but this term was 
often used specifically to designate the leek. 

Cabbage of some kind was an important food-herb 
among the early Irish, so that it is often mentioned in old 
authorities. Its Irish name was braisech [brasshagh], 
borrowed probably from the Latin brassica : but this word 
braisech was sometimes applied to a pottage made of herbs 
of any kind.* Among the articles of food noticed in the 
" Vision of Mac Conglinne " is " boiled, leafy, brown- white 
kale or cabbage." In the Culdee rule the monks were 
permitted on festival days to add kale as a condiment to 
their customary scanty allowance of bread. 

Among the vegetables cultivated in kitchen-gardens and 
used at table were leeks and onions. " Mac Conglinne's 
Vision " mentions the leek by one of its Irish names lus, 

* As in Stokes, Lives of SS., 362, a, I0 . 


and the onion by the name cainnenn. Lus is now the 
general word for a leek, and was often used in this special 
sense in old writings : in the Rule of the Culdees " three 
or four sprigs of luss " are mentioned as part of the refection 
of the monks : but lus primarily means an herb in general. 
A leek had a more specific name, folt-chep {jolt, ' hair ' : 
" hair-onion " : chep or cep, corresponding with Lat. cepa, 
' an onion '). A pregnant woman, as we are told in the 
Tripartite Life (201), once longed for leeks, so that she 
was like to die ; whereupon St. Patrick miraculously 
changed a rush into a folt-chep, which she ate and was 
cured : and Patrick declared that " all women who shall 
eat of this herb (the leek) shall be cured of their [longing-] 
illness." That the word cainnenn or cainne signifies an 
onion there can be no doubt, for an old Irish authority 
remarks that a cainnenn will draw tears from the eyes.* 
Under this name onions are mentioned as part of the 
refection due to a chief from his tenant, f But cainne was 
also occasionally applied to a leek : as in the expression 
in the Crith Gabhlach, " a handful of cainne with their 
heads." % Garlic appears to have been a pretty common 
condiment, and the same word cainnenn was often applied 
to it. O'Donovan sometimes translates cainnen and coinne 
by garlic in the Brehon Law :§ from the Law also we see 
that garlic was cultivated in gardens. But in individual 
passages it is often doubtful whether cainnen or cainne 
means onion or garlic. 

Wild garlic, called in Irish crem or creamh [crav or 
craff] was often used as a pot-herb, gathered no doubt from 
the fields, for I find no evidence that it was cultivated. 
The facts that it is often mentioned in Irish literature, 
and that it has given names to many places, || show that it 

* Mac Conglinne, 163, last line. f Br. Laws, iv 339. 

X Br. Laws, iv. 303, 6 from bottom 

§ Ibid., 11. 255, „. iv. 117, l6: 119, 27 . 

|| For which see loyce, Irish Names of Places, 11. 347. 


was a well-recognised plant and pretty generally used. 
The Chronicon Scotorum records the winter of a.d. 1006 
as being so mild that creamh (which is a summer plant) 
grew in the fields. Sprigs of creamh are mentioned as 
portion of the food of the Fena of Erin,* no doubt as an 
annlann ; and in the Brehon Law (11. 327, note 2), a certain 
food allowance is mentioned as given to a chief in the 
time of the " creamh harvest " (see also crim-mes, p. 139, 
supra) . 

Tap-rooted plants were designated by the general term 
mecon [mackan], with qualifying terms to denote the 
different kinds : but mecon used by itself means a parsnip 
or a carrot. Both these vegetables were cultivated in 
kitchen-gardens, and are often mentioned in old writings. 
St. Ciaran of Saigir had for his dinner every evening a 
small bit of barley-cake and two mecons of murathach, 
with a drink of water, f 

Good watercress (birir) was prized and eaten raw as a 
salad or annlann, as at present. It is often spoken of in 
connexion with brooklime, which is called fochlocon in 
Cormac's Glossary (p. 72), but more commonly fothlacht 
[fullaght], and which was also eaten. Constant references 
to both are found in the ancient tales. In the Story of 
the Colloquy a certain well is praised for producing large 
birir and fothlacht% : and St. Caillin, when the people of 
Magh-Ae in Connaught had received him well, left them, 
among other blessings, the palm of pure water and brook- 
lime^ Among the prerogatives of the king of Ireland 
mentioned in the Book of Rights were the cresses of the 
river Brosna in Westmeath. The three drink-bearers of 
Conaire, king of Ireland in the first century, are repre- 
sented in the story of Bruden Da Derga, as having before 

* Silva Gad., 119 {creamh here translated " gentian " by O'Grady). 
f Feilire, 62, b, 3. Stokes does not translate murathach : Sullivan 
makes it equivalent to gort, ' an enclosed garden ' ; Introd., 366 : but ? 
I Silva Gad., 103, 104. § Book of Fenagh, 179, and note 10. 



each of them, ready for the use of the guests, a cup of 
water in which was a bunch of birir.* From all this it is 
plain that both watercress and brooklime were in constant 
use as salads. In Cormac's Glossary (p. 19) the first 
syllable of birir, ' watercress,' is derived from bir, ' a well 
or stream,' which is certainly correct. 

Poor people sometimes ate a pottage made of the 
tender tops of nettles, as I have seen them do in my 
own day in time of scarcity : but they mixed a little 
oatmeal with it when they could get it. Once when 
St. Columba was walking near the monastery of Iona, he 
saw a poor old woman cutting nettles ; and he asked 
what she wanted them for. She replied : — " I have but 
" one cow that I am expecting to calve soon : and until 
" that happens I live on nettle-pottage, which I have eaten 
"for a long time back." He was much impressed with 
this, and said : — " This poor woman eats nettles, and 
" endures hunger, waiting for an uncertain event — the 
" calving of her cow : why should I not live on that same 
" pottage too, since the thing I look for is very certain — 
" namely, heaven ? " Whereupon he ordered his cook to 
give him for supper thenceforward nettle-pottage without 
milk or butter. But as time went on, the brethren, who 
had heard with dismay of the change for the worse in 
his diet — which was poor enough before — were rather 
surprised to observe that he still continued in excellent 
condition. Their talk among themselves coming to his 
ear, he began to suspect some kindly pious fraud on the 
part of the cook. So he sent for him and asked him : — 
" What do you put into my pottage every day ? " The 
cook, looking as innocent as a lamb, replied : " I know 
" nothing that goes into the pottage unless it could come 
" out of the iron of the pot or out of the potstick." The 
saint, who was not so easily hoodwinked, examined the 
potstick, and found that the cook had ingeniously made 

* O'Curry, Man. & Cust., II. 150. 


it hollow like a pipe, and thus contrived to pour in some 
milk or meat juice unknown to Columba, to keep his 
master from starving. The saint at once put a stop to 
the thing : but with characteristic kindness of heart left 
a blessing on the cook for his affectionate solicitude.* 

We find it stated by several Anglo-Irish writers that 
in former times the Irish occasionally ate the shamrock. 
Spenser, for instance, mentions that in time of famine the 
poor people who were reduced to the last stage of starvation 
were glad to eat water-cresses and shamrocks ; Fynes 
Morrison has a passage of much the same import ; while 
Thomas Dinely, who made a tour through Ireland in 1675, 
tells us that the people ate shamroges to cause a sweet 
breath. In the time of Elizabeth, Aengus O'Daly, the 
notorious satirist (for whom see vol. I., p. 455), reviling one 
of the Irish clans, represents them as at certain seasons 
making an onslaught on the shamrocks, f All this has led 
some persons to believe that the true shamrock is the Oxalis 
acetocella, or wood-sorrel. I see no reason, however, why 
these passages should not refer to the white trefoil, which 
is quite as fit to be used as a food-herb as wood-sorrel ; for 
I think we may assume that neither cress nor shamrocks 
were eaten in any quantity except under pressure of 
extreme hunger, but only as an annlann with other food, 
as watercress is eaten now. 

Moreover seamar and seamrog are given in Irish 
dictionaries as meaning Trijolium repens, which is the 
name of the true shamrock, while wood-sorrel is desig- 
nated by samhadh-coitte and seamsog. And as corroborating 
the dictionary explanations, we find the compound scoith- 
sheamrach (translated by O'Donovan " abounding with 
flowers and shamrocks " : scoth, ' a flower ') a favourite 
term among Irish writers to designate a green, open 
plain. The old records, for instance, tell us that Fiacha 

* Stokes, Lives of SS., 302 : Feilire, 100. 
| Tribes of Ireland, 51, .note 8 ; 53, note 3. 


Finnscothach (Fiacha of the white flowers), king of Ireland 
before the Christian era, was so called because " every 
plain in Ireland was scoith-sheamrach in his time " : and 
the same term is used by the Irish poet, Ferfeasa O'Cointe, 
about the year 1617 (Misc. Celt. Soc, 1849, p. 355), and by 
the writer of the Life of St. Scuithin (O'Cl. Cal., p. 5). In 
these passages seamar cannot mean ' wood-sorrel,' which is 
not produced in sufficient abundance, and moreover does 
not grow in open plains, but in shady places under trees 
and hedges.* 

The sea-plant called in Irish duilesc, and in English 
dillesk, dulse, dulsk, or dilse (Rhodymenia palmata) , growing 
on sea-rocks, was formerly much used as an article of food, 
that is, as an accompaniment. It was eaten with butter, a 
practice that Martin (p. 68) found in the Western Isles of 
Scotland in 1703. According to the Brehon Law, seaside 
arable land was enhanced in value by having rocks on its 
sea-border producing this plant, and there was a penalty 
for consuming the duilesc belonging to another without 
leave, f Among the various choice articles used by the 
Fena as food was duilesc from the coves of Cape Clear. J 
In the Crith Gabhlach duilesc is included with other food 
as part of the refection due to an og-aire or junior chief :§ 
and it is mentioned in the " Vision of Mac Conglinne " 
(p. 88) as a desirable viand. On one occasion St. Senan, 
while at Scattery, sent one of his disciples to cut some for 
him on the island-rocks. j| Dillesk is still used ; and you 
may see it in Dublin hawked about in baskets by women : 
it is dry and, people eat it in small quantities raw, like 

The marine-plant called Porphyra vulgaris, a species 
of laver, found growing on rocks round the coast, was 

* See this whole question well discussed by Mr. Nathaniel Colgan in 
Journ. Soc. Antiqq., liel., 1896, pp. 211 and 349. 

f Br. Laws, 1. 171, middle. § Br. Laws, iv. 309. 

J Silva Gad., 119. || Stokes, Lives of SS., line 2331. 


esteemed a table luxury : it is now often sold in flsh-shops 
and eaten with pepper, vinegar, &c. It is called in Gaelic 
sleabhacdn [slavacan, sloakan], which in the anglicised 
forms sloakan, sloke, and sluke, is applied to it all over the 
Three Kingdoms.* 

Though there is not much direct mention in old Irish 
literature of the management of fruit-trees, various detached 
passages show that they were much valued and carefully 
cultivated. It would appear from a remark in the Irish 
language on St. Paul's Epistle to the Romans, xi. 24, 
written in the eighth century by some Irish commentator, 
that the art of grafting was probably understood in Ire- 
land : at least this old writer shows himself familiar with 
the process : but whether he saw it practised at home is 
not made clear. The remark in question is : — " For they 
" [the Romans] have a custom to cut a tree and to insert 
" another tree therein."f One of the Brehon Law tracts 
(iv. 149) has a curious provision showing much thought- 
fulness and knowledge in the management of trees — though 
the case instanced in not a fruit-tree. If a person stripped 
off part of the bark of a growing oak-tree belonging to 
another — a thing sometimes done for tanning — he had not 
only to pay a fine, but also to secure the tree against 
injury from weather by covering the wound and two fingers 
beyond all round with a plaster made of fine clay, cow- 
dung, and new milk. The apple (ubhall, pron. ooal) 
appears to have been as much cultivated and used in old 
times as at the present. Apples, when gathered, were 
hoarded up to preserve them as long as possible : they 
were generally eaten uncooked. During the great festivals, 
the Culdees, though not permitted to increase the quantity 
of bread at meals, could use various condiments as an 
indulgence, and among them apples. % According to 

* S e Sullivan, Introd., 367. 

f Stokes and Strachan, Thesaurus, I. 529. See also Stokes, in Trip. 
Life, Preface, el. J Reeves, Culdees, 85, 


O'Donovan, the word abhall (fern.) was used in the best 
and most ancient Irish mss. to denote the apple-tree, and 
ubhall (masc.) its fruit* : but this distinction has long 
ceased to be observed. 

The hazel-nut was much used for food. This is plainly 
indicated by the high value set on both tree and fruit, of 
which we meet with innumerable instances in tales, poems, 
and other old records, in such expressions as " Cruachan of 
the fair hazels " : " Fidh-cruaiche of the white banquets, in 
which are shower-shaken hazels of white bark " : " Doire- 
na-nath, on which fair-nutted hazels are constantly found." 
The Brehon Law classes the hazel among the " noble " 
trees, partly on account of its nuts (for which see p. 287, 
infra) : a plain indication of the value set on them. 
Abundance of hazel-nuts was a mark of a prosperous and 
plenteous season. The year 1031 is mentioned by the 
Four Masters as of such abundance that, among other 
great bargains, you could buy in Armagh one-third of a 
sesedhach " of the nuts of the fair hazel-hedge " for one 
pinginn or penny. It is expressly stated in the Colloquy 
of the Ancients that part of the choice food of the Fena of 
Erin was " nuts from the hazels of Cantire " (cno do chollaib 
cintire).\ Among the blessings a good king brought on 
the land was plenty of hazel-nuts : — " O'Berga [the chief] 
for whom the hazels stoop " [with the weight of their 
fruit] : " Each hazel is rich from [the worthiness of] the 
hero " (HyF, 253, 221). 

In many similar entries nuts are mentioned without 
the hazel being specified : but there can be no doubt that 
hazel-nuts are meant. An old Irish poem enumerating 
the viands for the Mayday festival has among them cno 
mes, ' nut-mast. 'I A young man comes up to St. Patrick 

* HyF, 285, d. The distinction seems to indicate that the fruit was 
imported long before the tree was naturalised in Ireland. 
f Silva Gad., 119. 
% Hib. Minora, 49 : see also Br. Laws, v. 407, bottom. 


and his companions with a present of fruit : — " An armful 
of yellow-headed nuts, and of beautiful golden-yellow 
apples* : and on a certain occasion the cook of St. Mochta 
of Louth brought him a dish of nuts.f Nuts are often 
referred to as a dainty : a lady sends to her lover kernels 
(ettne), and apples, and many sweets. J In the Book of 
Leinster it is recorded that in the year 1056 there was a 
remarkable nut-harvest (cno-mess : " Trip. Life," 525). It 
has been stated elsewhere (at p. 121, supra) that hazel-nuts 
were sometimes used as an ingredient in making mead : 
but this was an exceptional and minor use. It appears 
that nuts were hoarded up for use like corn, another 
illustration of the value set on them. 

From all these references and quotations it may be 
inferred that hazel-nuts were regarded as an important 
article of human food. No doubt they were generally 
eaten raw, as they are at present. Tacitus tells us that 
some of the Celtic nations of Gaul ground acorns and other 
wild nuts into meal of which they made a sort of coarse 
bread : but I find no evidence that the Irish ground nuts 
for food. The Irish name for the hazel-tree is coll, gen. 
coill, caill, or cuill. A nut of any kind is cno or cnu, 
cognate with Latin nux and English nut, both of which 
have lost the initial hard c, which the Irish has preserved. 
A hazel-nut is called cno-coill. 

The sloe-tree or blackthorn was called droigen (Corm. 
60), modern droigheann [dree-an], which generally takes a 
diminutive form droigheannan [dreenan] : hence dreenan- 
donn or drinan-donn (donn, ' brown ') is a common name for 
the blackthorn, even among English-speaking people. The 
sloe is called dime [awrna] : a less usual word, grannmhuine 
[granwinna], is given in O'Clery's Glossary as meaning 
sloes. That sloes were used as food, or as an annlann or 

* Silva Gad., 112. f Stokes, Three Irish Homilies, 99. 

J Mac Conglinne, 4, , 3 . see also Three Sorrowful Stories, Atlantis, m. 
385, verse. 


condiment, and that the sloe bush was cultivated, is evident 
from the manner in which both are mentioned in Irish 
literature. Annagh Island in Lough Conn is called by 
way of praise " a district of sloes and apples."* The year 
103 1 was so plentiful that, as the Four Masters tell us, 
a large measure of black-red sloes could be bought for a 
penny. When King Domnall was endeavouring to placate 
his angry foster-son Congal, he offered him, among many 
other privileges [the produce of] an apple-tree and of a 
sloe-tree out of every homestead in a certain district :f and 
among the many dainties promised to Finn mac Cumail 
on a certain occasion, as related in the story of the Boroma, 
were sloes from one of the glens of Ebliu, now Slieve Felim, 
east of Limerick city. St. Brigit once came to a certain 
church round which grew abundance of apples and fragrant 
sloes ; and one of the nuns gave her a basketful of the 
fruit. I 

Strawberries (sing, sub, pi. suba : pron. soo, sooa) are 
often mentioned as dainties. In the passage above referred 
to from the Boroma, Finn was also promised strawberries 
from Sliab Bairrche, now Slieve Margy near Carlow. We 
are told in the Book of Rights (p. 9) that one of the pre- 
rogatives of the king of Erin was to have the heath-fruit 
(fraechmeas) of Brigh-Leithe (now Slieve Golry in Long- 
ford) brought to him. The fraechmeas was no doubt the 
whortleberry (called whorts or hurts in Munster), as is 
indicated by the fact that the whortleberry is now called 
fraechog and fraechdn, two diminutives of the same word 
fraech, heath. Most Dublin people have seen women with 
baskets of " froghans," as they call them, for sale, picked 
on the neighbouring mountains. The passage referred to 
shows that they were eaten in old times even by kings. 

Beechmast and oakmast were greatly valued for feeding 
pigs, which were kept in droves among the woods. The 

* HyF, 283. t Moyrath, 131. 

J Stokes, Lives of SS., 326. 


general name for mast was mes or mess. On one occasion 
the badb [bauv] or war-witch, predicting evils for Ireland, 
included among them " woods without masts."* (feda tin 
mess). The Four Masters signalise the year a.d. 835, for 
" great produce both of beechmast and acorns, which so 
choked up the brooks that they ceased running." And in 
the Brehon Laws (iv. 257 bot.) mast is coupled with grass 
and corn as a part of the valuable produce of the land. 
In the Bodleian copy of the Dinnsenchus we are told that 
" in the west of the plain of Macha there was a fruitful 
" oak wood, of which the odour was so fragrant, that when 
" the swine in the country all round smelled the wind that 
" blew over it, it was a heartbreak to them, and they 
" rushed quite mad to get to the wood." The same story 
is told in the Book of Leinster.f 

9. Fuel and Light. 

Fuel. — as the country abounded in forests, thickets, 
and brakes, the most common fuel for domestic use was 
wood. Firewood or " firebote " was called condud, or as it 
is now spelled, connadh [conna]. Two other names given 
in Cormac's Glossary (p. 73) are fochonnad and geltine. 
Firewood, made up in faggots, is mentioned in the Book of 
Rights as a portion of the tribute of the unfree tribes of 
Leinster to the king of that province. A bundle of fire- 
wood was called a brossna, a word found in the oldest 
authorities and used to this day all over Ireland, even by 
the English-speaking people, as meaning a bundle of 
withered branches, or of heath, for fuel. We read in the 
Tripartite Life (p. 10) that when St. Patrick was a boy, his 
foster-mother told him to bring her a brossna of withered 
branches to make a fire. 

Peat or turf was much used as fuel. The Senchus M6r 
speaks of the cutting of turf from a bank (port) and carti.ig 

* Rev. Celt., XII. III. 

■j- Folklore, in. 514 : LL, 169, a (" Srnthair Matha "). 


it home when dry ; and mentions a penalty for stealing 
it.* It is recorded in the Annals that Ragallach, king of 
Connaught in the middle of the seventh century, having 
exasperated some men who were cutting turf (oc buain 
monadh) in a bog, they fell on him and killed him with 
their sharp ruams or turf-spades, f The whole bog was 
the " commons " property of the fine or group of related 
families : but a single turf-bank might belong for the time 
to an individual. I The word ruam, used above, was a 
general word for any spade. At the present day the sharp 
spade used in cutting turf is designated by the special 
name of sleaghan [pron. slaan, the aa long like the a in 
star]. This word is a diminutive of sleg, modern sleagh, 
a ' spear ' (see vol. I., p. 108, supra) : sleaghan, a little 
spear ' — though a slaan is not very like one. 

Metal-workers used wood charcoal ; for neither plain 
wood nor peat afforded sufficient heat to melt or weld. 
We have seen (vol. I., p. 565) that charcoal made from 
birch afforded the highest degree of heat then available ; 
and was used for fusing the metals known at that time. 
Allusions to the use of charcoal — which in Irish is 
designated gual or cual — are met with in all sorts of 
Irish literature. In the Book of Rights (253) it is stated 
that the king of Hy Gabla was entitled to certain stipends, 
and, among other things, to " a ring of gold from the 
white [-hot] coal " (fail oir o'n gheal ghual) : and in the 
Crith Gabhlach we find mention of " a sack of coal (gual) 
for the irons. "§ In a poem in tne Book of Leinster the 
fuel kept in a blacksmith's forge is designated cual craing, 
a form of cual craind, ' coal of crand or wood,'|| which 
plainly points out its material. From the Crith Gabhlach 
passage, as well as from the special manner in which the 

* Br. Laws, 1. 133 ; also iv. 221, 223. 

f Silva Gad., 431 : Keating, 476. See also p. 92, supra. 

I Br. Laws, 1. 133, 165 : a law still observed. 

§ Ibid., iv. 311, bottom. 

!| O'Currv. Man. & Cust., i. T47 : LL, 35, a, 4 , 


cual craing is mentioned in this last entry, it appears that 
it was usual for smiths to keep it in sacks. The pit on 
which charcoal was made was called clas-guail or gual- 
chlais. The remains of some of the old charcoal-pits are 
still recognisable. I know one in which the soil is mixed 
up and quite black with quantities of charcoal-fragments 
and dust. In the story of " Fingal Ronain " a man says : — 
" If I were to be thrust into a cual-chlais tened (a ' fiery 
charcoal-pit.'), I would not do it " [a certain evil deed].* 

That coal from mines was used at some very early time 
is rendered certain by the fact that old coal-mines have 
recently been found exhibiting all the marks of extreme 
antiquity (chap. xxiv. page 289, infra). But as the word 
gual or cual, which is constantly used in the old accounts 
of the Irish metal-workers, will stand for either mine-coal 
or wood-charcoal, the literature alone — or that portion of it 
available — would not enable us to infer with certainty that 
mine-coal was used by the old Irish smiths and braziers, or 
used at all for fuel. 

A live coal from a turf fire was called smerdit or 
smeroid\ : a mixture of smerdids and hot ashes was, 
and is still, called griosach [greesagh]. A mass or fire of 
burning coals, especially of charcoal or coal, was called 
richis or rights, I a word which has long dropped out 
of use. O'Clery explains righis as lasair (flame), and 
O'Davoren has richis as meaning tine (fire). Straw, 
when plentiful and not otherwise wanted, was sometimes 
used as fuel in the absence of better. That this was so we 
know by a provision of the Book of Aicill, which mentions 
a penalty for stealing it.§ If it was intended for cattle- 
feeding, there is a certain fine : but if for burning, the fine 
was less. Probably the straw for burning was wheaten 
straw, and was subjected to some sort of preparation, such 
as trussing it up into wisps. Sometimes when fuel was 

* Rev. Celt., xni. 376, 377. % Rev. Celt., xi, 435, top. 

t Conn. Gloss., 149. § Br. Laws, III. 151. 


scarce, the poor people burned dried cowdung, gathered in 
the pasture fields in summer, as they do to this day : they 
call it boithredn [boraun], a word formed by suffixes from 
bo, ' a cow.'* 

Flint and steel with tinder were used for striking and 
kindling fire. In the ancient Latin version of the voyage 
of St. Brendan, so celebrated all over Europe in the middle 
ages, the old hermit Paul says that on a certain occasion 
he struck fire with flint and steel {silice ferro percusso) and 
cooked his fish.f And in the Mediaeval Irish Tract on 
Latin Declension, edited by Stokes (No. 720), igni ferrium 
(' steel-fire ') is glossed by the Irish teine-creasa, literally 
' fire of the girdle,' so designated because the whole 
kindling-gear, or tenlach-teined, as it is called, J i.e. the flint, 
steel, and tinder, was carried in the girdle-pocket, so as to 
be ready to hand. The spark produced by flint and steel 
was called tenlam, which in Cormac's Glossary (158) is 
derived from tene, ' fire,' and lam, ' the hand ' : that is to 
say, it means ' hand-fire ' : and O'Clery again explains this 
word tenlam by teine-creasa, ' girdle-fire.' Tinder was, and 
is, commonly called sponc [spunky which is obviously the 
same as the Latin spongia, English sponge. Spunk or 
tinder was sometimes made from the dried leaves of the 
coltsfoot, so that this plant is now always called sponc :§ 
but in recent times it was more usually made of coarse 
brown paper steeped in a solution of nitre and dried. 
" Spunk " is now used as an English word both in Ireland 
and Scotland : "a spunk of fire on the hearth." 

Light. — In the better class of houses dipped candles 
were commonly used. The usual Irish word for a candle 
is caindel or cainnel, which seems borrowed from the Latin 
candela : but there is also an old native word for it — innlis, 

* See O'Curry, Man. & Cust., 11. 65. 

f Brendaniana, 171 : Card. Moran, Act. Brend., 127. 

X Silva Gad., 302, top : Ir. Text, 267, s . 

§ See Kilk. Archaeol. Journ., 1868-9, p. 449. 



which O'Clery's Glossary explains by cainneal. There are 
numerous references to candles in ancient Irish authorities. 
The Senchus Mor mentions candles of " eight fists " (about 
forty inches) in length, made by [repeated] dipping of 
peeled rushes in melted tallow or meat grease* : from 
which we learn that the wicks of candles were, sometimes 
at least, made of peeled rushes : but other kinds of wicks 
were used. In the Tripartite Life (p. 53) St. Patrick is 
made to say, when about to present himself at Tara, that 
he would not make of himself a candle under a vat (caindel 
fo dabaich). In the Irish Life of St. Senan in the Book of 
Lismore (which however seems a comparatively late piece, 
though copied from older books), we are told that the 
saint — when young — once asked for a candle to light him 
while grinding corn at the mill : and the cook answered 
" I have no dipped candles [coinnle tumtha] just now but 
©ne : take this and you will get more if they are dipped, "f 
In the ancient tract called the " Law of Adamnan," it is 
stated that before the time of that saint one function of a 
cumal or female slave was to dip a candle (cainnel), four 
hand-breadths in length, in butter or lard, and to hold it 
in her hand to light the company at supper till they 
separated for bed.| 

As bees were so abundant, beeswax (Irish ceir, pron. 
care), as might be expected, was turned to account. Bees- 
wax candles must have been in use at some early period 
in the houses of the rich. In the Book of Rights (15 and 
xlvi) it is stated that one of the prerogatives of the king 
of Leinster was " to drink by the light of wax candles 
[coindle ciarriha] at Dinnree " ; which was one of the most 
ancient of the royal residences of Leinster (p. 94, supra). 
Add to this that beeswax " found in square masses, and 
" also in the form of candles, has been discovered under 

* Br. Laws, il. 251, 253. f Stokes, Lives of SS., line 1995. 

X Trip. Life, Introduction, xxii. See also for candles, Ware, Antiqq., 
183, bottom. 


" circumstances which leave no doubt as to the great 
" antiquity of such articles."* Several specimens of this 
ancient wax are in the National Museum, Dublin. 

Although, in very early times, candles were sometimes 
held in the hands of slaves, they were more commonly 
placed on candlesticks. The ancient Irish word for a 
candlestick is caindelbra, modern Irish coinnleoir [conlore], 
both of which are modified forms of the Latin candelabra. 
The Senchus Mor and the Crith Gabhlach mention a 
caindelbra as a usual article in a house ; O'Donovan here 
translates the word ' branch-light ' : and the old Irish com- 
mentator explains it as " the straight wand upon which the 
" beautiful light is placed, like a candle, in the house of 
" each person."! The ancient Latin Hymn of Secundinus 
makes mention of a light placed on a candelabrum% : and 
in the description of the Banqueting-House of Tara in 
the Book of Leinster it is stated that there were seven 
coindelbruig in it.§ 

It was usual to keep a rlchaindell [reehannel], or ' king- 
candle ' (ri, ' a king '), or royal candle, of enormous size, 
with a great bushy wick, burning at night in presence of a 
king : in the palace it was placed high over his head ; 
during war it blazed outside his tent-door ; and on night- 
marches it was borne before him. This custom is men- 
tioned very often in the records. We are told in the 
Annals that Cerbhall [Kerval], king of Ossory, coming 
out of his chamber in the middle- of the night to attack 
the Danes, a.d. 860, had " a large king-candle carried 
before him, the light of which shone far on every side."|| 
In Tain-bo-Fraich (O' Curry,. 11., 219), Froech visits his aunt 
with a spear shining like the candle of a king-house (caindel 

* Wilde, Catalogue, 255. See also Kilk. Archaeol. Journ., 1892, p. 184. 
t Br. Laws, 1. 126, s; 143, bottom ; also iv. 310, „. 
% Trip. Life, 387, 14. 

§ Petrie's Tara, p. 188. On Candlesticks, see also Reeves, Eccl. 
Antiqq., 210 ; and Joyce, Irish Names of Places, 11. 204. 
}| Three Fragments, 145. 

1 64 



rig-thigi) in the hand of each of his fifty attendants : and 
in Bruden Da Derga in LU, three heroes are described as 
sitting in presence of King Conare with the candle of a 
royal house burning over the head of each of them.* A 
hero's spear is sometimes compared (as above) to a royal 
palace candle for the brightness of its polished bronze 
headf ; and in this sense a spear is sometimes called, 

I it.. 213. Fie. 219. Fig. 220 

Rush and Candle-holders : found in different parts of Ireland. These are 10 or 12 inches high, 
and when in use were placed on a table. Those intended to stand on the floor were about 36 inches 
high. (From Proc. Roy. Ir. Academy for 1889-91, p. 629.) 

'figuratively, a caindell rigthaige; as in LL, 97, a, i 4 from bot.: 
See also Rev. Celt, xxiv., 128, note. In regard to size, the 
Four Masters are fairly definite in the passage already 
quoted, vol. 1., p. 62, supra, where the " king-candle " kept 
burning at night before Shane O'Neill's tent (a.d. 1557) is 
described as " a huge torch thicker than a man's body " : 
a passage which shows moreover that this custom continued 
till the sixteenth century. (See also Ware Antiqq., 183 bot.) 

* O'Curry, Man. & Cust., 11 140. 

| O'Curry, Man. & Cust., 11. 219, I9; and 220, note, 1st col., line 9. 


The poorer classes commonly used a rush-light, i.e. a 
single rush peeled (leaving one little film of rind the whole 
length to keep it together) and soaked in grease, but not 
formed into a candle by repeated dippings. It gave a poor 
light and burned down very quickly ; and it was known by 
two names, adann and itharna [ey-an : lharna]. It is well 
characterised in a verse ascribed to St. Colman, founder of 
Cloyne (died a.d. 604), quoted in Cormac's Glossary (p. 10), 
in which a warrior, praising his sword, is made to say : — 
" As blackbirds are to swans, as peasant women are to 
" queens, as an adann is to a candle [so is any other] sword 
" to my sword." The word adann means ' to kindle ' ; and 
hence adannadh was applied 
to a candle-lighter in a church. 
The other word itharna is 
also given in Cormac's Glos- 
sary (p. 92) as equivalent to 
adann : and it is derived, says FIG - -**■ 

,■• ^~>i t( -.. Ancient Irish bronze Lamp. Found in a 

tUe LrlOSSary, aCCOramg tO crattnog-e (i.e. an island-dwclling in a lake) 

~-~.*.:*.~i. ...„:4.^^« " 12 „ „„„:^„.j. in County Roscommon. The vessel held (he 

ancient writers [i.e. ancient oil , and the wick projected from tlie pipe . 
when the Glossary was written £*£*«**«** Lake Dweu^s m i rc i»»d. 

— ninth or tenth century), 

from ith, ' fat,' i.e. " the fat of the cattle they used to melt 
in the rushes " (simnib). There were simple holders for 
rushlights : and several specimens have been found, which 
however do not appear to be old : nevertheless they are 
probably the representatives of the holders of more ancient 
times. They are of iron or of wood, or of both combined, 
and are often so formed as that each will hold either a rush 
or a dipped candle, or both at one time.* (See last page.) 
Oil lamps of various kinds were used ; and they are 
often mentioned in the oldest recordsf under two names — 
lespaire [les-pe-re] and luacharnn or lochrann (from Lat. 
lucerna). Luacharnn occurs several times in the eighth- 

* See Kilk. Archaeol. Journ., 1890-1, p. 473 ; and Proc. Roy. Ir. Acad., 
1889-91, p. 626. f Corm. Gloss., 103 : Stokes, Lives of SS., line 342. 


century Glosses of Zeuss, as the equivalent of lampas and 
lucerna, which shows the remote time in which lamps and 
lanterns were used in Ireland. Some were made of bronze 
(fig. 221) : some of clay. A rude unglazed earthenware 
lamp, shallow, and with a snout to support a wick, was found 
some time ago among prehistoric remains near Portstewart.* 
It has been already stated (p. 30) that it was usual to 
light the principal apartment in a house from above by a 
row of lamps or candles suspended from the ridgepole. 

In modern times, long, dried slits of bogdeal have been 
used bj^ the peasantry instead of candles. Probably the 
same practice prevailed in early times, but I have found 
no notice of it. 

10. Free Public Hostels. 

This seems a proper place to give some information 
regarding the provision made for lodging and entertaining 
travellers and officials. Hospitality and generosity were 
virtues highly esteemed in ancient Ireland ; in the old 
Christian writings indeed they are everywhere praised and 
inculcated as religious duties (1. 330, supra) ; and in the 
secular literature they are equally prominent. The higher 
the rank of the person the more was expected from him, 
and a king should be lavish without limit. In the story of 
the Battle of Moyrath (p. 105), Erin is lauded for many 
virtues during the reign of King Domnall, among thern 
hospitality : — " Her habitations were hospitable, spacious, 
" and open for company and entertainment, to remove the 
" hunger and gloom of guests." The duty of dealing out 
plenty to guests was so universally recognised that even 
the Brehon Law (iv. 337) is careful to specify the cases 
where a king may be excused for deficiency of food if 
there should occur an unexpected arrival of a number of 
guests : — such as failure or refusal, at the wrong moment, 
of a tributary chief to send in the expected food-supply. 

* Kilk. Archaeol. Town., 1883-4, p. 318. 


Guaire, king of Connaught in the seventh century, is 
celebrated and lauded everywhere in tales and poems for 
his generosity and hospitality. He was so constantly 
stretching out his hand to give away that — as the legends 
have it — his right hand grew longer than his left : and to 
this day he is known as " Guaire the Hospitable." We 
often find generous persons praised in terms like those 
applied to Owen O'Madden, a Connaught chief, of whom it 
is said in an old Irish document (HyM, 141) that " he does 
" not refuse anyone gold or horses, food or kine. and he is 
" the wealthiest of the race of Gaedhal for bestowing them." 
Even St. Patrick himself is represented as blessing the 
district of Moy Rein in this fashion : — " I leave prosperity 
"to the place so that it shall provide for all [requiring 
" help] even though every cleric should be poor " — as much 
as to say that in case the clergy and the monasteries had 
not the means to discharge the duty of hospitality expected 
of them, the lay people of the district should be so prosper- 
ous that they could and would provide for all without any 
clerical help.* 

If by any accident a person found himself unable to 
discharge the due rites of hospitality, it was supposed that 
his face became suffused with a mice [rucke] or blush — a 
blush of honourable shame — called also enech-ruice or ainech- 
ruice, ' face-blush,' as it is explained in Cormac's Glossary 
(p. 66). The brewy or head of a hostel took care to have 
" the snout of a rooting-hog 'f — meaning he had plenty of 
p 0r k — " to break or prevent his face-blush " : when there 
was plenty there was no reason to blush. If anyone 
through the default of another ran short of provisions when 
visitors came, so that he had reason to feel ashamed of his 
scanty table, the defaulter had to pay him as compensation 
what was called a " blush-fine. "J 

* Book of Fenagh, 273. t Br - Law, iv. 311 et seq. 

% Br. Laws, I. 123, Ii; 129, note 1. ; iv. 345, middle; 347> I3: and 
Corm. Gloss., 103 (" Leos "). 


As illustrating what was expected of the higher ranks, 
the Brehon Law (iv. 237) lays down that " the chieftain 
" grades are bound to entertain [a guest] without asking 
'-' any questions " — i.e. questions as to his name, or business, 
or where he was bound for, and the like. It is added that 
the Feine or farmers were not so bound — i.e. they might 
make reasonable inquiry about a guest before entertaining. 
Once a guest had partaken of food in a house, his host 
was bound to abstain from offering him any violence or 
disrespect under any circumstances.* Bede's testimony 
as to the hospitality of the Irish has been already quoted 
(vol. 1., p. 414). 

This universal admiration for hospitality found its 
outward expression in the establishment, all over the 
country, of public hostels for the free lodging and enter- 
tainment of all who chose to claim them. At the head of 
each was an officer called a brugh-fer, or brugaid, or briuga 
[broo-fer, brewy, broo-a], a public hospitaller or hosteller, 
who was held in high honour. He was bound to keep an 
open house for the reception of certain functionaries — king, 
bishop, poet, judge, &c. — who were privileged to claim for 
themselves and their attendants free entertainment when 
on their circuits : and also for the reception of strangers. 
He had a tract of land and other large allowances to 
defray the expenses of his house : the names brugh-fer 
and brugaid indeed literally signify ' landholder,' from 
brugh, land, a farm of land. Brugaid was often used in 
the sense of a farmer merely, but we have here to do only 
with its special application to a keeper of a public hostel. 

The brewys were of two main classes. The lowest was 
the brugaid cedach or ' hundred hospitaller, 'f who should 
have at least one hundred of each kind of cattle, one 
hundred labourers, and corresponding provision for feeding 

* As illustrative, see how Branduff treated Glasdam, p. 483, below. 
t See Stokes, Rev. Celt., xv. 431 : HyF, 239, h : and Kiik. Archseol 
Journ., 1872-3, p. 47, verse xlii. 


and lodging guests. The brugaid cedach is constantly 
met with in all kinds of Irish writings. " But " — says the 
gloss on the Senchus Mor — " there is a brugaid who is 
better than this man " : this was the brugaid-lethech, who 
should have two hundred of each kind of cattle. His 
house should be supplied with all necessary furniture and 
appliances, including one hundred beds for guests ; for he 
was not allowed to borrow.* In order to be at all times 
ready to receive visitors, a brewy of either class was bound 
to have three kinds of meat cooked and ready to be served 
up to all who came ; three kinds of raw meat ready for 
cooking ; besides animals ready for killing. In one of the 
law tracts a brewy is quaintly described as " a man of 
three snouts " : — viz. the snout of a live hog rooting in the 
fields to break the blushes of his face ; the snout of a dead 
hog on the hooks cooking ; and the pointed snout of a 
plough : meaning that he had plenty of live animals and 
of meat cooked and uncooked, with a plough and all 
other tillage appliances, f He was also " a man of three 
sacks " : — for he had always in his house a sack of malt 
for brewing ale ; a sack of salt for curing cattle-joints ; and 
a sack of charcoal for the irons ; this last referring to the 
continual use of iron-shod agricultural implements calling 
for frequent repair and renewal. We are told also (iv., 
pp. 310, 311) that his kitchen-fire should be kept per- 
petually alight, and that his caldron should never be taken 
off the fire, and should always be kept full of joints boiling 
for guests. The whole description is a picture of lavish 
abundance, reminding one strongly of Chaucer's description 
of the Franklin : — 

" An householder, and that a grete was he 
Seint Julian J he was in his Contree, 
His brede, his ale, was alway after on ; 
A better envyned§ man was wher non. 

* Br. Laws, 1. 47, bottom. j Ibid., iv. 311. J St. Julian, the patron- 
saint of travellers and of hospitality. § Supplied with, wine. 


VVithouten bake mete never was his hous, 

Of fish and flesh, and that so plenteous, 

It snewed [snowed] in his hous of mete and drinke, 

Of alle deintees that men coud of thinke, 

After the sundry sesons of the yere, 

So changed he his mete and his soupere. 

Full many a fat partrich hadde he in mewe, 

And many a breme, and many a luce [a fish] in stewe. 

Wo was his coke but-if [i.e. unless] his sauce were 

Poinant and sharpe, and ready all his gere. 

His table dormant in his halle alway 

Stode redy covered all the longe day." 

There should be a number of open roads leading to 
the house of a brewy, so that it might be readily accessible ; 
and on each road a man should be stationed to make sure 
that no traveller should pass by without calling to be 
entertained ;* besides which a light was to be kept burning 
on the faithche [faha] or lawn at night to guide travellers 
from a distance. The noble brewy, Da Derga, mentioned 
below, kept his doors open day and night, except at the 
windy side of the house, f 

As visitors and their followers were constantly coming 
and going, the house-furniture and other property of a 
brewy were jealously protected by law from wanton or 
malicious damage, the various possible injuries being set 
forth in great detail, with the compensation for them. He 
was moreover a magistrate, and was empowered to deliver 
judgment on certain cases that were brought before him 
to his house : " He is a bo-air e for giving judgment." We 
have already seen (vol. I., p. 44) that a, court was held in 
his house for the election of the chief of the tribe. Keating 
says that there were ninety brugaids in Connaught, ninety 
in Ulster, ninety-three in Leinster, and a hundred and 
thirty in Munster, all with open houses ; and though it is 
not necessary to accept these numbers as strictly accurate, 

* Br. Laws, v. 17, , 7 . and 79, 22 . 

f Da Derga, 36. See also, about the brugatd, Br. Laws, v. 77, 79. 


they indicate at least that the houses of hospitality were 
very numerous. The house of a brewy answered all the 
purposes of the modern hotel or inn, but with the important 
distinction, that guests were lodged and entertained with 
bed and board, free of charge. With great probability the 
rule prevailed here, as in case of private hospitality, that 
an ordinary guest was supposed to be kept — if he wished 
to stay — for three nights and three days : after which the 
obligation to entertain ceased : but I have not found this 
specifically mentioned. 

There were a few brewys of a higher class than the 
preceding, who had large tracts of land and held a vety 
exalted position. They often entertained kings, chiefs, 
and warriors, of the highest classes, with whom also they 
were on terms of familiar intercourse. The hostel of one 
of these was called a brudin or bruden [now pronounced 
breen or bryan], a word which Mr. Crowe* connects with 
the Greek prytaneum, meaning the same thing as the Irish 
bruden — a house of public or state-endowed hospitality. 
In the time of the Red Branch Knights there were six of 
these " chief courts of hospitality in Erin," each situated 
at the meeting of four roads, f all of which figure in the 
Romantic Tales. The most remarkable of them was the 
" Bruden Da Derga," kept by the great hosteller Da 
Derga. The ancient story of the " Togail or Destruction 
of Bruden Da Derga," gives a detailed and very vivid 
account of the sack of this hostel and the slaughter of its 
people, including Conari I., king of Ireland — who happened 
to be staying in it at the time with all his retinue — by a 
band of Irish and British marauders, in the first century of 
the Christian era : in which however the assailants suffered 
still greater loss than those they attacked. This fine story 
has been lately edited and translated by Dr. Stokes in the 
Revue Celtique, vol. xxii. The narrative fixes the position 

* In Kilk. Archaeol. Journ., 1868-69, p. 326. 

t According to the poem referred to in note *, p. 173, infra 


of the Bruden Da Derga with great precision, as situated 
on the river Dodder, where it was crossed by the " Slige 
Cualann " (for which see chap, xxviii., p. 395, infra), the 
great road leading from Tara across the Liffey at Dublin, 
and on across the Dodder through the district of Cualann 
towards Bray. 

In 1879, during the preparation of a piece of ground for 
building near Donnybrook, a remarkable discovery was 
made, which, as in many other like cases, goes to confirm 
the truthfulness of the old saga. A large, low, earthern 
mound situated beside the Dodder on the south side, at a 
spot now called Mount Enroll, a little east of the Roman 
Catholic Church of Donnybrook, was levelled, in which 
vast quantities of human bones were found, not interred as 
in an ordinary cemetery, but flung in heaps and otherwise 
exhibiting unmistakable evidences of a general massacre. 
The whole mound and its contents were carefully examined 
by Dr. Frazer of Dublin, whose account of the exploration 
is given in the Proc. Roy. Ir. Academy, 1879-1886, p. 29. 
Sir Samuel Ferguson, in an instructive note to his poem of 
Conary, has rightly identified the place where this mound 
was situated, with the site of Da Derga's hostel. 

We have seen that each bruden was placed at the 
meeting of four roads : i.e. where two main roads crossed. 
The two roads here were the Slige Cualann and another 
leading from the old district of the once powerful tribe of 
Hy Donohue which lay along on both sides of the Dodder 
from Glennasmole down. This road passed by Boherna- 
breena and Rathfarnham till it crossed Slige Cualann at 
the Bruden, and on towards the mouth of the Liffey, then 
and afterwards a great resort of trading vessels. O'Curry 
(Man. & Cust., 11. 136) states that Bohernabreena took 
its name from the old Bruden or breen, in which he is 
undoubtedly correct : for " Bohernabreena " is the proper 
anglicised phonetic form of Bothar-na-Bruidne, i.e. the 
road of the Bruden, meaning the road leading to it. But 


he is certainly mistaken in asserting, as he does, that the 
present Bohernabreena was itself the site of Da Derga's 
Bruden ; as anyone may see who glances through the 
story. The destruction of this Bruden is recorded in the 
Annals of Tigernach. 

Another of these six hostels was Bruden-Da-Choca (or 
Choga), kept by the hospitaller Da Choca. This was the 
scene of another tragedy, in which Cormac Conlingeas, son 
of Concobar mac Nessa, perished, as related in the historical 
tale called " The Destruction of Da Choca's Bruden," which 
has also been translated by Dr. Stokes in the Revue Celtique, 
vol. xxi. The old fort, the only remaining relic of this 
bruden, is still well known. It is situated in Westmeath, 
a few miles from Athlone ; and to this day it retains the 
name " Bruden," in the phonetic form of Bryan. Forgall 
Monach, or Forgall the Wily, Cuculainn's father-in-law, 
kept another of these brudens at Lusca, now Lusk, north 
of Dublin, which figures much in the tales of the Red 
Branch. The remaining three were the bruden of Mac 
Datho in Leinster ; Bruden Blai Briuga (or Brugaid) in 
Ulster ; and Bruden mic Dareo (or Bruden da Ger) in 

Every bruden was a place of refuge for a homicide, 
where he might claim protection from the immediate 
vengeance of his victim's friends till he could obtain a fair 
trial before a brehon ; as appears from a statement in the 
" Destruction of the Bruden Da Choca " : — " Every Bruden 
is an asylum of the red hand," i.e. for the manslayer.f In 
this function the Irish bruden answered to the " asylum " 
of the Greeks : and Dr. Stokes has called attention to the 
curious correspondence of the six refuge brudens of Ireland 

* An account of the whole six may be seen in a short poem published 
and translated by Stokes, in Rev. Celt., xxi. 397 : of which a very corrupt 
version was published in Proc. Roy. Ir. Acad., 1870-76, p. 253. They are 
also enumerated in the Battle of Moyrath, pp. 51-53. 

t Rev. Celt., xxi. 315. In the Br. Laws, v. 319, a manslayer is called 
" a man of red weapons." 


with the six Jewish cities of refuge for manslayers against 
the avengers of blood.* 

The word bruden is still preserved in the names of 
several townlands in different parts of Ireland, from which 
it is probable that the term was applied to other houses of 
hospitality as well as to the six mentioned above. Or 
perhaps the multiplication of the name may have arisen 
from this circumstance : that bruden was — as we know — 
sometimes applied to any great banqueting-hall, as, for 
instance, to that of Tara and of Dun-da-benn, now the fort 
called Mountsandal, over the Bann, near Coleraine ; and 
also to the royal house of refuge for aged warriors at 
Emain (vol. I., p. 97, supra).] " Bruighean " — says Peter 
O'Connell's Dictionary — ' a sumptuous house, a court or 
palace. 'J 

There was another sort of public victualler called 
biatach or biadhtach [beetagh], who was also bound to 
entertain travellers, and the chief's soldiers whenever they 
came that way. In order to enable the betagh to dispense 
hospitality, he held a tract of arable land free of rent, 
called a baile-biadhtaigh or ballybetagh, equal to about 
1000 of our present English acres, with a much larger 
extent of waste land. The distinction between a brewy 
and a betagh is not very clear. They are distinguished in 
a passage in the Book of the Dun Cow,§ which, among 
other classes of people, mentions the briugaid and the 
biatach : but there was probably little substantial difference 
between them. The Four Masters record the death of 
several individual biataghs : thus at a.d. 1225 (p. 219) : 
" Auliff O'Boland, Erenach of Drumcliff, a wise and learned 

* Numbers xxxv. ; Deuteronomy iv. : Joshua xx. 

t See Petrie's Tara, 199, bottom : Mesca Ulad, 13, 8: and Rossnaree, 
20, last line. 

X Zeitschr., Celt., Phil. 1. 427. At the present day Bruden, "in its 
modern form bruighean [bree-anj, means a ' fairy-palace " ; for which, 
and for the local names derived from it, see Joyce's Irish Names of Places, 
vol. 1. 289, 290. § LU, p. 123, b, A and s from bottom. 


man, and a general biatach, died." In later times the 
English of the Pale used the word betaghe to denote a 
servile class of farmers, like the English villeins : those on 
the king's manors are indeed sometimes called ' the king's 
villeins or betaghes.' But from the descriptions of these 
persons given by Anglo-Irish authorities, it appears that 
they were a class of dependent tenants who held small 
farms, probably on betagh lands, from which in course of 
time they came to be erroneously called betaghs. It is 
clear they were not the biataghs of Irish records.* 

The Irish missionaries carried this fine custom to the 
Continent in early ages, as they did many others : for we 
are told, on the best authority, that before the ninth 
century they established ' hospitalia/ chiefly for the use of 
pilgrims on their way to Rome, some in Germany, but 
most in France, as lying in the direct route to the Eternal 


In the legendary history we read of female brewys. 
Just before Cormac mac Art's accession there was a ban- 
brugaid or ' she-brewy ' at Tara ; the lady already men- 
tioned (1., 216), whose sheep ate up the queen's crop of 
glaisin : and in Cormac's Glossary (p. 13c) is given the 
legend that through the country there were several female 
brewys who entertained Finn and the Fena on their 
hunting excursions. 

* About the Anglo-Irish betaghes, see Ware, Antiqq., chapter x-x. : 
\teg. of All Hallowes, xv. : and Statute of Kilkenny, pp. 4, 5. 
f See Lynch, Cambr. Ev., 11. 244-5. 

Ornament composed from the Book ef Kells 

Sculpture on a Capital : Priest's House, Glendalough : Be ranter, 1779. 
(From Petrie'; Round Towers.) 



Section i. The Person and the Toilet. 

arks of Aristocracy. — An oval face, broad above 
and narrow below, golden hair, fair skin, 
white, delicate, and well-formed hands with 
slender tapering fingers : these were considered 
by the ancient Irish as marking the type of beauty 
and aristocracy.* Among the higher classes the finger- 
nailsf were kept carefully cut and rounded : and beautiful 
nails are often mentioned with commendation. It 
was considered shameful for a man of position to have 
unkempt nails : among several opprobious terms applied 
by Conan Mael to the warrior Cairell is crechtingnech, 
' ragged-nailed.' Crimson-coloured finger-nails were greatly 
admired. In the Tain a young lady is described as having, 
among other marks of beauty, " regular, circular, crimson 
nails " ; and ladies sometimes dyed them this colour. 
Deirdre, uttering a lament for the sons of Usna, says : — - 

* All the above characteristics are mentioned so often in Irish writings 
that it is unnecessary to give references. 

f Inga or ionga, ' a finger-nail.' As to the nails, see Silva Gad., 381, 
with Irish text, 339, s . Tromdamh, 71, bottom : Hyde, Lit. Hist., 258, 
bottom ; Sons of Usna, 413, note 44 : and Sullivan, Introd., 72, 73. 



" I sleep no more, and I shall not crimson my nails : no 
" joy shall ever again come upon my mind."* 

Ladies sometimes dyed the eyebrows black with the 
juice of some sort of berry, as appears from the following 
expression in Cael's poem in praise of the lady Crede 
[cray] : — " A bowl she has whence berry-juice flows, with 
which she colours her eyebrows black, "f We have already 
seen (vol. I., p. 343) that the Irish missionary monks some- 
times painted or dyed their eyelids black. An entry in 
Cormac's Glossary plainly indicates that the blush of the 
cheeks was sometimes heightened by a colouring matter 
obtained from a plant named ruam. The Glossary thus 
explains the word : — " Ruam, i.e. ro-eim, an herb that 
gives colour to the face until it is red. "J The ruam was 
the alder : but the sprigs and berries of the elder-tree were 
applied to the same purpose, as appears by the " kenning " 
or figurative name — the " reddening of faces " — given to 
this tree in a passage in the Book of Ballymote.§ It is to 
be hoped that bedecking the face with an artificial blush 
was practised only by ladies : but the authorities do not 
enlighten us on the point : or perhaps it would be more 
correct to say they leave a sort of presumption that the 
practice was common to men and women. In connexion 
with all this, it is proper to remark that among Greek and 
Roman ladies the practice was very general of painting 
the cheeks, eyebrows, and other parts of the face. 

The Hair. — Both men and women wore the hair long, 
and commonly flowing down on the back and shoulders — a 
custom noticed by Cambrensis, and pronounced by him, 
in his narrow-minded way, barbarous, because among the 
Anglo-Normans of his time it was the fashion to trim the 

* Sons of Usna, Atlantis, ill. 413 : Ir. Texte, 1. 79, „. This is like the 
practice of the Egyptian ladies dyeing their finger-tips with henna. 

t O'Curry, MS. Mat., 309, 595, 6 . Silva Gad., 120 (Irish text, m). 
Berry-juice is here called sugh-subh, from sugh, ' juice,' and subh ' a berry.' 
\ Corm. Gloss., 144: Three Ir. Gloss. 39 (for the Irish). 
Kuno Meyer, in Rev. Celt., xm 220, note. 



hair short. The hair was combed daily after a bath, as is 
shown by the passage from the Battle of Rossnaree, quoted 
at p. 185, infra. The heroes of the Fena of Erin, before 
sitting down to their dinner after a hard day's hunting, 
always took a bath and carefully combed their long hair. 
The Irish derived this fashion of the hair from old times ; 
for we know that part of Gaul was called " Gallia Comata," 
from the long hair worn by the people.* 

Among the higher classes in very early times great 
care was bestowed on the hair : its regulation constituted 
quite an art ; and it was dressed up in several ways. Very 
often the long hair of men, as well as of women, was 
elaborately curled. Conall Cernach's hair, as described in 
Da Derga (p. 199), flowed down his back, and was done up 
in " hooks and plaits and swordlets." The accuracy of 
this and other similar descriptions is fully borne out by 
the most unquestionable authority of all, namely, the 
figures in the early illuminated manuscripts and on the 
shrines and high crosses of later ages. In nearly all the 
figures of the Book of Kells, for example (7th or 8th century) 
the hair is combed and dressed with the utmost care, so 
beautifully adjusted indeed that it could have been done 
only by skilled professional hairdressers, and must have 
occupied much time. Whether in case of men or women, 
it hangs down both behind and at the sides, and is com- 
monly divided the whole way, as well as all over the head, 
into slender fillets or locks, which sometimes hang down 
to the eyes in front. In some cases the fillets are combed 
down straight, though kept carefully separate ; but in 
others each is beautifully curled or twisted spirally the 
whole way down, which must have been done with a 
curling-iron of some kind. These descriptions apply to 
the hair of priests and nuns as well as to that of lay people. 
In the seventh century this elaborate arrangement of the 
hair must have been universal among the higher classes : 

* Ware, Antiqq., 176 



for the artist who drew the figures in the Book of Kells 
has represented the hair dressed and curled in the manner 
described, not only on the figures of men and women, 
whether lay or ecclesiastical, but even on those of angels. 
The three nuns represented on the Breac Moedoc [Brack 
Mogue : thirteenth century] have the hair hanging down 
on each side to the waist : not divided into fillets, but each 
a single mass twisted spirally. Two other figures from 
the same shrine, given here — both ecclesiastical — show 
very well how men had 
the hair and beard dressed, 
which is seen still better 
in the figure of the Evan- 
gelist at page 197, below. 
I do not find mentioned 
anywhere that the Irish 
dyed their hair, as was the 
custom among the Greeks 
and Romans. 

For women, very long 
hair has been in Ireland 
always considered a mark 
of beauty. For example, 
in the Tain, a lovely lady 
is described as having her 
yellow hair parted in four 
wreaths, three of them braided round her head, and the 
fourth hanging down to her ankles.* This admiration 
has come down to the present ; for you constantly find 
mentioned in the Irish popular songs of our own day, a 
maiden " with golden hair that swept the dew off the 
grass " — or some such expression. The long fair hair 
hanging down at the back was called cuilfhionn [coolin] ; 
from cul, the back of the head, and finn or fionn, white : 
whence the well-known anglicised word coolin or coolun, 

* O'Curry, Man & Cust., II. no 

Fig. aaa. 

Figures of two ecclesiastics worked in bronze 
on the Breac Moedoc, the ' Shrine of St. Maidoc or 
Moeue,' dating from abont the thirteenth century. 
See vol. I., p. 570, note •. (From Miss Stokes's 
Early Christian Art, p. 107.) 




which is often applied to a fair-haired person, but which is 
now better known as the name of a beautiful Irish air, and 
of Moore's exquisite song to it. 

In the fifteenth, sixteenth, and seventeenth centuries, it 
was usual, among the general run of people, for unmarried 
girls to wear the hair carefully combed out and hanging 
down loosely on the back : while married women more 
commonly bound it up round the head, with bright- 
coloured ribbons and long pins, in tasteful knots and 
wreaths : generally with a covering of some 
kind — a cap or folded kerchief.* 

The practice of braiding the hair must 
have been very general among men as 
among women. One test of the activity 
of a candidate for admission to the ranks 
of the Fena of Erin (vol. I., p. 87, supra) 
was that he should be able to run and 
escape from pursuers through a wood- 
without letting the braids of his hair be 
disordered by the branches, f It was con- 
sidered an accomplishment for a young 
man to be able to plait hair well. J 

Very often — especially in active life — 
the hair was bound up and confined with 
rings or circlets, called by various names, 
such as fdinne, flesc, buinne (or bunde), of gold, silver, 
or white bronze ; or with ribbons or fillets of different 
materials, or with thin flexible gold plates (called 
lann or niamlann : see pp. 249, 250, below). This mode 
of disposing of the hair — both of men and women — is' 
constantly referred to in the tales. § That these binders 
were often of gold we know, partly from the literature 

* On this see Lynch, Cambr. Ev., 11. 169. 

t O'Grady, Silva Gad., 100 : Keating, 350. J Three Fragm., 35. 

§ For example, Voyage of Bran, 1. 60, „ ; 72, 3I : O'Curry, Man. & Cust., 
J 59> 9: !69, top ; 188, bottom : Joyce, Old Celtic Romances, 248. Such 
references might be indefinitely multiplied. 

fig. 233. 

Portions of the plaited 
hair of a woman whose 
body, clothed in antique 
woollen costume, was 
found buried in hard 
gravel, under a bog, 4!4 
feet deep, in the County 
Down, in the year 1780. 
(From Proc. Roy. Irish 
Academy, IX. 102.) For 
more about this find, see 
note under illustrations, 
P. 352. in/ra. 



and partly from the testimony of the National and other 
museums, in which may be seen many long plates and 
ribbons of gold, most of them probably used to confine 
the hair. One beautiful golden ribbon of this kind in 
the National Museum, 5 feet long, will be mentioned in 
section 3 of this chapter. The forehead-band or fillet 
usually worn by a charioteer, sometimes of bronze or 
findruine, sometimes of a woven fabric, was called gipne* 
— a word also applied to a doctor's cupping-horn (I. 621, 

Fig. 224. 

Fig. 225. 

Fig. 226. 

Ancient Irish Combs, of bone, now in National Museum. Fig. 224 is 10 inches long : fig. 225 
2J4 inches : fig. 226, 2^ inches. (They are not drawn here to uniform scale.) Fig. 224 is of one 
single piece, with thin metal plates riveted on the sides. The other two have each two plates 
riveted together, with the teeth inserted between, and firmly riveted : so that if a tooth got broken, 
it could be withdrawn and a new one inserted. (From Wilde's Catalogue, p. 271.) 

supra). At the end of this chapter will be found a notice 
of the custom of suspending light, hollow gold balls from 
the ends of the hair- wreaths. In later times the long locks 
worn on the back and sides of the head by men were called 
glibbes by Anglo-Irish writers. t 

Combs. — From what precedes it will be understood 
that combs were in general use with men as well as with 
women : and many specimens — some made of bone, some 
of horn — some plain, some ornamented — have been found 

* Wooing of Emer, p. 72, i 9 : LU, p. 122, b, 8# , » 7 . 
t Ware, Antiqq., 176, bottom. 


in lisses, crannoges, and such like places. In the Crannoge 
of Cloonfinlough near Strokestown in Roscommon have 
been found combs in an incomplete state of manufacture, 
indicating a combmaker's workshop.* The comb — Irish 
cir or cior [keer] — is, as we might expect, often mentioned 
in ancient Irish writings. In the story of Maildune, two 
great birds are said to have " picked and sleeked the 
plumage [of another bird] as if it were done with a 

As long hair was so much admired, so baldness was 
considered a serious blemishj : and as showing the notice 
it attracted, we find it classified in Cormac's Glossary 
(p. 143, " Range ") into six different kinds, which the 
author names and describes. Mail, mael, or maol is the 
Irish word for bald ; and baldness is designated by matte 

The Beard. — The men were as particular about the 
beard as about the hair. The common Irish names for 
the beard were ulcha and feasdg [faissoge], of which the 
last is still in use. It was also called grend or greann 
(Conn. 90) : in O'Clery's Glossary greann is explained 
by " ulcha or feasdg." In very early times the men — 
especially the soldiers and higher classes of people — wore 
the full beard. The soldiers of King Domnall, marching 
to the Battle of Moyrath, had " tufted beards covering and 
" surrounding their cheeks and mouths, their faces and 
" their heroic chins : great is the length of their beards, 
" which reach to their navels." 

The fashion of wearing the beard varied. Sometimes 
it was considered becoming to have it long and forked, 
and gradually narrowing to two points below. King 
Concobar mac Nessa — like many of his attendant heroes — 
is described as having " a double-forked beard upon his 
chin " : and other kings and mighty heroes are constantly 

* Proc. Roy. Ir. Acad., v. 211. t Rev - Celt -» x - 77- 

t O'Curry, Man. & Cust., n. 144, bottom : Da Derga, p. 286. 


described as wearing their beards in this fashion. On several 
panels of the high crosses at Monasterboice and elsewhere, 
as well as on the shrines, and in the Book of Kells, are 
figures of men with full beards : in some the beard is 
forked ; in others it falls down in a single mass : while in 
a few it is cut rectangularly not unlike Assyrian beards 
(see the figures on St. Manchan's shrine, p. 204, below). 
Nearly all have a mustache, in most cases curled up and 
pointed at the ends as we often see now. In some there is 
a mustache without a beard : and a few others have the 
whole face bare. In many the beard is carefully divided 
into slender twisted fillets, as described above, for the hair. 
All this must have taken great trouble and a long time to 
arrange : but among the higher classes there was provision 
for it ; for kings and chiefs had their barbers (p. 184, infra). 
Indeed men must have given more time to this part of the 
toilet than women ; for they had both hair and beard to 
attend to. 

It was disgraceful to have the hair and beard trimmed 
short. When Cuculainn had his hair and beard cut off 
by Curoi mac Daire, who had vanquished him in single 
combat, and inflicted this humiliation on him, he remained 
in a hiding-place till both grew sufficiently long to be 
presentable. None others but nobles, chiefs, and warriors 
were permitted to wear the full beard : and those who wore 
it were bound by laws of honour to be brave and generous, 
never to retreat in battle, never to resort to mean ways of 
fighting, never to engage in manual or servile labour, and 
to be always ready to relieve distress. Working people 
were prohibited from wearing beards, so that they were 
expected to shave at least once a month.* 

The beard that grew on the upper lip, when the lower 
part of the face was shaved, was called crombeol (' stoop- 
mouth '), what we now designate a mustache. This term 

* All this is laid down in a short ancient Irish piece called Gets* Ulchai, 
or Prohibitions of Beard, edited by O'Looney, in Bee. Fola, 191. 




is often met with in Anglo-Irish writings in the form 
crommeal. This was the fashion sometimes adopted by 
soldiers marching to battle, who probably regarded 
the long beard on the chin as an encumbrance. 
Among the silly measures passed by the Anglo-Irish 
Parliament in 1465, was one prohibiting the crommeal — 
commanding all the Irish within the Pale to shave the 
upper lip like the English* 

That the ancient Irish used a razor 
(in Irish alt or altan) is proved by the 
fact that it is mentioned in our very 
oldest documents, and in such a way as 
shows it to have been a very familiar 
article. In a poem in the Book of 
Leinster, alt, ' a razor,' is mentioned 
twice : — " as sharp as a razor was his 
spear."f In Cormac's Glossary (p. 10) 
altan is derived from ail, ' edge,' and 
teinn, 'sharp-cutting.' In a still older 
authority, the Milan eighth - century 
glosses on the Psalms, the Latin raso- 
rium acutum is explained by the Irish 
commentator: — Amal inscrissid, edon, 
anial innaltain ndith : or in English, 
"as a scraper, that is, as a sharp razor. "J 
This shows moreover that the razor was then used in 
shaving as it is now, by a sort of scraping movement 
against the beard. In the Book of Leinster it is stated 
that the "man of shaving" {i.e. barber) to the sons of 
Miled was Maen of the Mighty Deeds, and that he 
received as fee for his office the district of Berramain, 
lying along the shore near Tralee in Kerry, which was so 
called — says the legend — from berrad, ' shaving,' and main, 

* See Lynch, Cambr. Ev., n. 219 : and "Ware, Antiqq., 176, bottom. 

t'MS. Mat., 481. 

J Stokes and Strachan, Thesaurus, I., p. 245. See also Zeuss, 657, i B . 

Fig. 227. 

Bronze cutting-instrument, 
believed to be a Razor. It is 
all of one piece, 3'A inches 
long, i'X inch wide, with the 
two edges very thin, hard, and 
sharp. In National Museum, 
where there are others like 
it. (From Wilde's Catalogue, 
P- 549) 


'riches or reward': — Berrad-main, 'pay for shaving.'* 
A razor was also called berr-scian : " shaving-knife." 

The Bath. — Bathing was very usual, at least among the 
upper classes, and baths and the use of baths are constantly 
mentioned in the old tales and other writings. The bath 
was a large tub or vat usually called dabach, but sometimes 
ammor, omur, or lothomur : bathing — taking a bath — was 
designated jothrucud.\ People bathed daily, generally in 
the evening ; and it was usual to prepare a bath for a 
guest. In the " Battle of Rossnaree " (pp. 33, 35), we read 
that when King Concobar's army were encamped for the 
night, after a day's march, " fires were kindled, food and 
" drink were prepared, they went into baths where they 
" were carefully cleansed, their hair was smooth-combed, 
" after which they had supper." In one of the houses that 
Maildune came to he found a bath ready : and a certain 
person who entertained St. Brendan and his companions 
in their voyage, had a bath prepared for them on their 
arrival.J In the story of Goll and Garb we read that 
when Concobar and his retinue were entertained at the 
house of Conall the brewy, a bath was prepared for them 
after supper.§ In the commentaries on the Senchus M6r 
it is incidentally mentioned that Fergus mac Leide, king 
of Ulster, took a bath every day. There was a bath for 
the use of visitors in the guest-house of every monastery : 
when St. Cairnech of Tuilen came on a visit to the 
monastery of Duleek, a bath was prepared for him in a 
dabach : and we are told in the Crith Gabhlach that every 
brugh-fer or brewy had in his house a bathing-vessel (long- 
foilcthe).\\ Kings and chiefs were in the habit of bathing 

* Silva Gad., 525, middle, and 478, 33 . LL, 167, b, 6 . and " Contents," 
43, b, middle. Other references to razors and shaving, Br. Laws, 1. 125, 3 . 
and 133, 23: Ir. Texte, 1. 277, 3: Moyrath, 20, 3i and 21, 4 . 

f Mac Conglinne, 11, M: Corm. Gloss., 73. J Brendan., 144. 

§ Rev. Celt., xiv. 417. 

|| Br. Laws, iv. 311, , 4 . See Foilcim in Glossary of Atkinson, Pass. & 


and anointing themselves with oil and precious sweet- 
scented herbs : as we find in the case of Cellach, king of 
Connaught, who, before a battle, bathed and anointed 
himself in this manner.* So Ulysses bathes and anoints 
himself with olive oil after being shipwrecked on the coast 
of Phaeacea (Odyss. VI.). A king of Leinster died at Naas 
while in a bath, as sometimes happens people at the 
present day. 

Every bath was furnished with a number of round 
stones. Among the articles of furniture in the guest-house 
of the Cork monastery was " a bath-tub with its stones, "f 
They must have been moderately small : for in the Gloss 
on the Senchus Mor we read that Fergus mac Leide being 
once angered by the bond-woman Dorn, he killed her with 
a bath-stone {cloch-fothraicthe)% : no doubt by flinging it at 
her. The bath stones are constantly referred to in all sorts 
of Irish documents : but what the use of them was is a 
puzzle. It has been suggested that the water was heated 
by throwing the heated stones into it : and this view 
receives countenance from an incident related in Jocelin's 
" Life of St. Patrick " (Cap. C), that the saint cursed the 
stones of Ushnagh, after which " they cannot be heated by 
" fire, nor when plunged into water [after coming out of the 
" fire] do they hiss like other stones " : which seems to 
show that Jocelin believed the stones were used to heat 
the water. In the far older Tripartite Life, where the 
same circumstance is told, nothing is said about heating 
the stones, but merely that after the curse, nothing good 
is made of them, " not even bathing-stones." 

But it seems incredible that they heated bath-water 
in this uncouth fashion ; for we know they often heated it 
in the ordinary way. In the Battle of Moyrath (275, 20 ; 

* Three Fragments, 107, top : and Silva Gad., 443, bottom. Other 
references to baths and bathing, Silva Gad., 77, bottom : Miss Hull, 
Cuch. Saga, 130, J2 (LL, 59, 7 ) : Mesca Ulad, 47 : Ir. Texte, 1. 295, 6 

t Mac Conglinne, 10, 2fi . % Br. Laws, I. 69, ,,. 


277, 3 ) we are told that in the house of a certain chief, the 
women were preparing a bath " for washing and bathing " ; 
and that they heated the water with firewood. In the 
tale of the Sons of Usna (Atlantis, in., p. 409), 
represented as preparing a bath for Naisi " over the fire " : 
and such examples might be multiplied.* 

Cormac's Glossary distinguishes between fothrucud, 
bathing the whole body, and indlot (or indluf) , washing the 
feet or hands ; and this distinction is generally recognised 
in the old writings. As the people had a full bath some 
time down late in the day, they did not bathe in the 
morning, but merely washed their hands ; for which 
purpose they generally went out immediately after rising 
and dressing, to some well or stream near the house. This 
practice is constantly referred to. " At early morning," 
says Mac Conglinne (70 : also 58), " I rose and went to the 
well to wash my hands." In the Sick Bed of Cuculainn, 
Eochaid Iuil goes out early in the morning to wash {do 
ilnuf) his hands at the spring ; and a better-known example 
is where, as we are told in the Lives of St. Patrick, the two 
daughters of King Laegaire came out in the early morning 
to the well of Clebach near Cruachan " to wash their hands 
as was their custom. "f 

In both washing and bathing they used soap (sleic, 
pron. slake). In the Crith Gabhlach we are told that 
foulness is washed away from a person's, honour as the 
face is washed with soap (sleic) and water and a linen 
cloth. I 

* In O'Grady's Silva Gadelica, p. 283 (transl.), the quarrel between 
King Fergus and his queen is related " anent precedence in the use of the 
bath-stone." But in the corresponding part of the Irish text, as given in 
the first volume (p. 250, 24 ) — tre imremim fothraicthe — it will be observed 
that the word for " stone " (cloiche) does not appear — perhaps a printer's 
error. The original manuscript is inaccessible to me : and, as matters 
stand, this passage teaches us nothing on the point. 

f Trip. Life, 101, top line : see also Tain bo Fraich, 165. 

J Br. Laws, iv. 319. 


Small Toilet Articles. — Mirrors of polished metal must 
have been common from very early times, for they are 
often mentioned ; generally by one or the other of the 
two names, scathdn [skahan] and scadarc, this last spelled 
variously scaiderc, scadarcc, and the oldest form scaterc [all 
now pron. sky-ark]. The great antiquity of the article is 
shown by its mention in Zeuss's Glosses (854, is), where 
scaterc glosses lucar, i.e. speculum, and where it is derived 
from scdth-derc, 'shadow-seeing,' or a 'shadow see-er.' 
From scdth [skaw], ' a shadow,' is also derived the other 
name scathdn, which is merely a diminutive form. In the 

Senchus M6r, among many house- 
hold articles, is mentioned a scadarc, 
which is explained by the other 
word scathdn ; and a man is spoken 
of as looking into it to see his 
image (scdth)* In the same autho- 
rity (I., 235, 3 from bottom) the 
a ^.dbox^Jinches ac^: white cloth and "the nitairic, i.e. 
c-S";:^;: the scathdn or mirror," are men- 
SS£ mSZSltJ™ tioned as among the articles of the 
toilet : which gives another name 
for the mirror — nitairic. In both these entries certain 
regulations are laid down against removing the mirror 
when one is using it, " looking at his image." In the 
romantic story of the Death of Fergus mac Leide, king 
of Ulster, we are told that this king was struck with 
a deformity in his face which he was not aware of, and 
care was taken that a mirror {scathdn) should not be left 
in his way. But one day when he and his queen had a 
quarrel, she, in her anger, brought him a mirror, in which 
he saw his face with his mouth all awry.f 

Small articles of the toilet, and especially combs, were 
kept by women in a little bag which they carried about 

*Br. Laws, 1. 124, i 3 ; 125, n ; 138, 30; 139, 34. 
t Silva Gad., 283. 


with them, called a ciorbholg [keerwolg], i.e. ' comb-bag ' 
(cior, ' a comb ' : bolg, ' a bag '). The Book of Aicill lays 
down that a woman is exempt from liability in certain cases 
of quarrel, if she shows her comb-bag and her distaff (cuigel) 
in presence of her guardian (Br. Laws, in. 291, top). 

2. Dress. 

Materials. — Woollen and linen clothes formed the dress 
of the great mass of the people. Both were produced at 
home ; and elsewhere in this book the mode of manufac- 
turing them will be described. Silk and satin, which were 
of course imported, were much worn among the higher 
classes, and we find both constantly noticed in our literature. 
The flags and banners used with armies were usually made 
of silk or satin, as we find mentioned in the Book of Rights 
and in many of the historic tales. The ordinary word for 
silk was sida [sheeda] ; but it was also sometimes called 
striae, a word coming from the same origin as the English 
silk, and Lat. sericum : from which again comes the adjec- 
tive sirecda, ' silken.' But siriac was also occasionally 
applied to satin. The common word for satin is srdll 
[srole], both in the old and in the modern language. 

The furs of animals, such as seals, otters, badgers, 
foxes, &c, were much used for capes and jackets, and 
for the edgings of various garments, so that skins of all 
the various kinds were valuable. They formed, too, an 
important item of everyday traffic, and they were also 
exported.* St. Molaise of Devenish wore a hood of 
badgers', skins, which, after his death, was cherished as a 
relic, and called the brocainech [bruckanagh], from broc, ' a 
badger. 'f In 1861 a cape was found in a bog at Derry- 
keighan in Antrim, six feet beneath the surface, made alto- 
gether of otter skins. " The workmanship of the sewing " 
— says Mr. Robert Mac Adam, who gives an account of it J 

* Wilde, Catalogue, 279. f Silva Gad., 21, bottom. 

J In Ulst. Journ. Archaeol., ix. 294. 


— " is wonderfully beautiful and regular : and the several 
" parts are joined so as not to disturb the fur, so that from 
" the outside it looks as if formed of one piece." 

In Scotland the tartan is much used — a sort of cloth, 
generally of wool, sometimes of silk — plaided or cross-barred 
in various colours. In some English dictionaries the word 
is conjectured to be derived from the French teretaine, 
which is not a good guess : but both the material and the 
name originated in Ireland. The original Gaelic name is 
tuartan, as we find it used several times, both in the 
Senchus Mor, and in the glosses on it, where tuartan is 
defined to be a sort of material " containing cloth of every 

Colours. — Before entering on the particular forms of 
dress it will be well to say a few words on colours. The 
ancient Irish loved bright colours. In this respect they 
resembled many other nations of antiquity — as well indeed 
as of the present day ; and they illustrated Ruskin's saying 
(speaking of poppies) : — " Whenever men are noble they 
" love bright colour, . . . and bright colour is given to 
" them in sky, sea, flowers, and living creatures." The 
Irish love of colour expressed itself in all parts of their 
raiment : and in chapter xxvi. (p. 356), below, it will be 
shown that they well understood the art of dyeing. 

Everywhere in our ancient literature we find dress- 
colours mentioned. Cahirmore, king of Ireland, saw in 
his sleep a vision, namely, the daughter of a brewy, with a 
beauteous form, and every colour in her dress, f Ailill and 
Maive, king and queen of Connaught in the first century, 
when showing off their treasures — as related in the " Tain " 
in the Book of LeinsterJ — brought forward their stores of 
apparel, " purple, blue, black, green, yellow, speckled, grey, 
brownish-grey, pied, and striped." In the Ulster army, as. 

* Br. Laws, i. 188, l8 . 189, 24> 25; 239, <. 
t Kilk. Archasol. Journ., 1872-3, p. 31. 
X LL, 54, a, 36: Man. & Cust., 90. 


described in the Tain, was one company with various- 
coloured mantles : — " some with red cloaks ; others with 
" light blue cloaks ; others with deep blue cloaks ; others 
" with green, or blay, or white, or yellow cloaks, bright and 
" fluttering about them : and there is a young red-freckled 
" lad, with a crimson cloak in their midst."* 

The king of Tara, as recorded in the Book of Rights, 
was bound to give, as stipend every year to the provincial 
king of Emain, who was his subordinate, " twelve suits of 
clothes of every colour " : and in the same book we often 
find notices of such articles as a purple cloak, a cloak with 
purple border, " ten red cloaks and ten blue cloaks." 

King Domnall, in the seventh century, on one occasion 
sent a many-coloured tunic (inar-ildathach) to his foster- 
son Prince Congalf : like Joseph's coat of many colours. 
The fashion of dyeing a single, cloak variously was so usuaV 
that we sometimes find it specially mentioned, as a thing 
worthy of notice, that a man's cloak had only one colour. J 
Colours are also depicted in the few drawings that have 
come down to us, as for instance in the Book of Kells ; but 
Dr. Keller is of opinion that no inference can be drawn 
from these figures as to the hues of the several garments 
in real life ; for he thinks that the colours are often flung 
in any way, according to the fancy or caprice of the 
artist.§ In several of the figures in this same book the 
upper mantle has a uniform pattern consisting of little 
spot-clusters about three inches asunder, irregularly 
placed : each cluster formed of three little white circular 
spots close together like a shamrock — but not a shamrock. 
This pattern is seen in the figure of the Evangelist at 
p. 197, infra, and it so often occurs that in this case at 
least we may conclude it represents a fabric often worn in 
real life. 

* MS. Mat., 38, bot. For another good example, see Rev. Celt., xrv. 413. 
t Moyrath, 39. + Voyage of Bran, p. 72, 19 , 20 . 

§ Ulst. Journ. of Archa^ol., vm. 229. 


We are told in our legendary history that exact regu- 
lations for the wearing of colours by the different ranks of 
people were made by King Tigernmas [Teernmas] and by 
his successor, Eochaid Edguthach (' Ochy the cloth- 
designer '), many centuries before the Christian era : — a 
slave was to be dressed in clothes of one colour ; a peasant 
or farmer in two ; and so on up to a king and queen and an 
ollave of any sort ; all of whom were privileged to wear six 
(FM, a.m. 3656, 3664). Whatever degree of credence may 
be accorded to this legend, it is certain that in historic 
times there was some such arrangement : for the commen- 
tator on the Senchus M6r lays down with some detail the 
colours to be worn by children in fosterage, the clothes of 
those of kings and high-class chiefs having more varied, 
brilliant, and costly colours than those of the lower grades.* 
All people, young and old, wore brightly-coloured clothes, 
so far as they could afford, or were allowed them : and we 
may infer from this Brehon Law example that the distri- 
bution of colours among various classes of people in 
ordinary life was subjected to some sort of supervision 
and regulation. 

At the present day green is universally regarded as the 
national colour : but this is a very modern innovation. It 
is well known that at the Battle of the Boyne in 1690, the 
Irish wore little strips of white paper in their caps, while 
the Williamites wore sprigs of green. In ancient times 
some colours were preferred to others. Purple, for 
instance, was a favourite with kings, for no other reason 
apparently than its great scarcity and expensiveness 
(for which see p. 363, below). On this account it is now 
sometimes designated the imperial or royal colour : but 
its preference had certainly nothing to do with nation- 
ality : and as a matter of fact the ancient Irish had no 
national colour. 

• Br. Laws, IX. 147, 149 


Classification of Upper Garments. — The upper garments 
worn by men were of a variety of forms and had many 
names : besides which, fashions of course changed as time 
went on, though, as I think, very slowly. Moreover, the 
several names were often loosely applied, like the English 
words " coat," " mantle," " frock," &c. ; so that it is often 
impossible to fix exact limitations. But the articles them- 
selves were somewhat less vague than their names ; and 
so far as they can be reduced to order, the upper garments 
of men may be said to have been mainly of four classes : — 

1. A large cloak, generally without sleeves, varying in 
length, but commonly covering the whole person from the 
shoulders down. 

2. A short tight-fitting coat or jacket with sleeves, but 
with no collar. 

3. A cape for the shoulders, commonly, but not always, 
carrying a hood to cover the head. 

4. A sort of petticoat, the same as the present Highland 
kilt. There was nothing to correspond with our waistcoat. 

Sometimes only one of those was used, viz. either the 
outer mantle or the short frock — with of course in all 
cases the under and nether clothing ; but often two were 
worn together ; sometimes three ; and occasionally the 
whole four. 

1. Loose Upper Garment. — The long cloak assumed 
many shapes : sometimes it was a formless mantle down 
to the knees ; but more often it was a loose though shaped 
cloak reaching to the ankles. This last was so generally 
worn by men in out-door life that it was considered 
characteristic of the Irish. It had frequently a fringed or 
shaggy border, round the neck and down the whole way on 
both edges, in front ; and its material was according to the 
rank or means of the wearer. Among the higher classes it 
was of fine cloth edged with silk or satin or other costly 
material. Sometimes the whole cloak was of silk or satin ; 
and it was commonly dyed in some bright colour, or more 



often — as we have said — striped or spotted with several 
colours. In the numerous figures in the Book of Kells (7th 
or 8th century) the over-garment is very common : some- 
times it is represented full length, but often only as far as 
the knees or the middle of the thigh. 

The large outer garment of whatever material was 
known by several names, according to shape, of which 
the most common was brat or bratt, gen. and pi. bruit, 
dat. brut : which appears to have been a general term for 
any outer garment, and which is still in common use, 
though somewhat altered in meaning. The word fallainn 
[foiling : from Eng. f aiding] was applied to a loose cloak or 
mantle, reaching about to the knees : but it has nearly 
dropped out of use. This was the name given to Giraldus 
Cambrensis by his informant in 11 85 : and he gave it the 
Latin form phalingium : — " Under which [i.e. under the 
" hooded cape : see p. 200, infra] they wear woollen 
" phalingiums instead of palliums or cloaks "* (that is, 
instead of the long cloaks that Giraldus was accustomed to) . 
There were other names for a mantle, which evidently 
point to some difference in material or make. The 
lummon was a cloak or ' wrap ' of coarse material. In 
Peter O'Connell's Dictionary the word is explained " a 
coarse cover, a large great coat, sackcloth " : and it was 
sometimes applied to a sack. According to the Dinn- 
senchus legend, Limerick derived its name from the 
circumstances that a high tide once flowed over a number 
of men standing on the brink, and carried off their loose 
luimne or lummons ; whence the place was called Luimnech, 
i.e. a spot full of lummons. Mac Conglinne, the ecclesiastical 
student, on his way to the Cork monastery, travels in his 
cloak, which he calls more than once his " white lumman "f; 
and in the story of Mongan in LU, a lay student is described 

* Top. Hib., in. x. ; see also Book of Rights, 38, note / : and Lynch, 
Cambr. Ev., n. 201. \ Mac Conglinne, pp. 9, 27 . and 25, IS . 


as wrapped in his lummon while learning his lesson.* In 
Cormac's Glossary (p. 104) the word lend, which was in 
common use for some sort of coat or mantle, is derived in 
this manner : — " Lend, the name for a white brat or mantle, 
from lee-find, ' white wool.' " This entry, and Mac Con- 
glinne's " white lummon," point to the practice, which we 
know from other sources (vc 4 . 1., p. 343, supra) was common 
enough, of making these coarse garments from undyed 
wool — the natural colour just as it came from the sheep's 
back. If the sheep from which the wool was taken were 
black, then of course the lummon was black ; and black 
sheep must have been — at one time at least — very general, 
for Giraldus says that nearly all the woollen clothes the 
Irish wore in his time were black, " that being the colour 
of the sheep in this country " (Top. Hib. in. x.). 

The fuan or fuaman was a loose brat or mantle : 
O'Clery's Glossary explains it as meaning a kind of brat. 
In the " Demon Chariot " Cuculainn is described as wearing 
a fuan of bluish-crimson around him with borders of pure 
white, silver, The word matal was applied to some sort of 
loose cloak, apparently — as O'Donovan believesf — another 
name for the fallainn. The term seems borrowed from 
Norse mottul, both words being masculine, and both pro- 
bably connected with Lat. mantelum, the n of this dropping 
out, as usual, in the transfer. In the Lebar Brecc the 
garment worn by our Lord is called a matal. The outer 
garment was called by another name, tlacht : " tlacht, that 
is, a brat or cloak," says O'Clery's Glossary. Momera, the 
Spanish princess, made a tlacht or cloak of bright-coloured 
wool for her intended husband Eoghan Mor, king of 
Munster.J Still another name for the brat was fola, which is 
given in O'Clery's and Cormac's Glossaries (Corm. 73). 

It was a very common fashion to have, on the loose 
cloak, five folds or plaits, called in Irish c6ic diabail : 

* Voyage of Bran, i. 54. f In Book of Rights, p. 38, note /. 

X Moylena, 163. 


from cdic, five : diabal, a plate or fold : diabul, duplex 
(Z., 980, 3 i). In the story of Laegaire Liban, a warrior is 
seen emerging from the mist wearing a purple five-fold 
brat* : Mac Conglinne's cloak is in one place called lummon- 
coic diabalta, a five-fold lummon : while Cuculainn, in the 
" Courtship of Emer," is described as wearing a beautiful 
five-fold fuan around him. The folds apparently ran 
across, not lengthwise. 

Women wore similar cloaks called by the same names. 
The woman that was to wait on Mac Conglinne (p. 96, i 4 ) 
was to have a purple five-fold bratt about her : and in the 
Tain bo Fraich in the Book of Leinster, the fifty women 
from the shee are described as wearing purple tunics (inar), 
green head-dresses, and brooches of silver. f Women 
often wore a variously-coloured tunic down to the very 
feet, with many folds and much material — twenty or thirty 
yards — which was different from the bratt and from the 
hooded cloak mentioned below. Under this was a long 
gown or kirtle. Linen, whether used by men or women, 
was commonly dyed saffron. The long cloak worn by 
women had often a hood attached at top which commonly 
hung down on the back over the cloak, but which could 
be turned up so as to cover the head at any moment when 
wanted. A woman represented on one of the crosses at 
Clonmacnoise appears with a hooded-cloak of this kind, 
the hood hanging down behind : and the country-women 
wear this sort of cloak to the present day all through 

The loose cloak, of whatever shape or by whatever 
name called, was almost always fastened at the throat by a 
brooch. Cloaks in their various forms and with their 
several names were an important commodity of inter- 
change, and very often constituted part of the tribute 
given by or to kings. 

* Stokes, Lives of SS., xxxiv. 
t Tain bo Fr., 149. 



It is difficult or impossible to embrace all varieties of 
clothing in any formal classification : and as a matter of 
fact there was another article of full-covering dress worn 
in very early times by both men and women, hardly 
included in any of the preceding descriptions. In the 
Book of Kells (7th or 8th century) 
a large number of the figures, both 
of men and women, have the usual 

FIG. M«. FIG. a*). 

Figure 239. representation of an Angel. (From the Book of Kells : Dr. Abbott's Reproductions, Plate XIV.) 
Figure 330, representation of one of the Evangelists. (From same, Plate XVIII.) 

outside mantle generally reaching to about the knees, and 
under it a long narrow garment like a petticoat (but not 
a kilt), from the shoulders down to the insteps, widening 
towards the bottom, yet so narrow that it would obviously 
interfere with the free movement of the feet in quick walking. 
I do not find this mentioned in the written records any- 


where — at least so as to be recognisable ; but^ it is depicted 
so often in the Book of Kells that it must have been in 
general use. It is quite conspicuous on the Blessed Virgin 
and on the Infant Jesus, on angels and evangelists, on our 
Lord in the representation of His arrest, and on the two 
soldiers who are in the act of arresting Him.* It is well 
shown here in figs. 229 and 230, both from the Book of 

Distinct apparently from the preceding over-mantles 
was the loose-flowing tunic — worn over all — usually of 
linen dyed saffron, commonly called Mine [2-syll.], which 
was in very general use and worn by men and women 
in outdoor life. This is noticed by Spenser (p. 102) as 
prevalent in his time. It had many folds and plaits and 
much material — sometimes as much as thirty yards ; so 
that in later times the Anglo-Irish Parliaments made laws 
forbidding the use of more than a certain quantity of stuff. 
It has been already remarked that the Irish generally wore 
this garment in battle. It seems to have been the lend- 
brat which the royal army are described as wearing in the 
Battle of Moyrath (p. 181, note c), intermediate between 
the loose brat and the close-fitting lend : made not too 
loose to impede the action of the limbs in fighting, with 
probably a waistbelt. Part of the Boroma tribute con- 
sisted of " three times-fifty hundred lend-brats." The 
saffron-tunic was used in the Hebrides down to the time of 
Martin : there it took twenty-four ells of material. It 
was he says, called leni-croich, because it was dyed with the 
crock or saffron-plant ; and when worn it was tied round the 
waist with a belt. 

The outer covering of the general run of the peasantry 
was just one loose sleeved coat or mantle, generally of 
frieze, which covered them down to the ankles ; and which 
they wore winter and summer. This is the garment that 

* Hodges & Figgis's Reproductions, Plates xxvu., xxvin., xxxi., 

XLI., XLII., L. 

chap, xxnj dress And personal adornment 199 

drew from Spenser (View, 87) the well-known denunciation 
that stands to this day an astonishing example of blind 
irrational prejudice and intolerance. Here are a few 
extracts ; which are applied, be it remembered, to an 
ordinary everyday garment, worn with no more malignant 
intentions than are our present overcoats : — 

" It is a fit house for an outlaw, a meet bed for a rebel, and an apt 
cloke for a theife. First the outlaw being for his many crimes and 
villanyes banished from the townes and houses of honest men, maketh 
his mantle his house, and under it covereth himselfe from the wrath 
of heaven, from the offence of the earth, and from the sight of men. . . . 
Likewise for a rebell it is as serviceable. For in his warre, when he still 
flyeth from his foe, and lurketh in thicke woods and straite passages, 
it is his bed, yea and almost his household stuff. . . . Lastly for a theife 
it is so handsome [convenient] as it may seem it was first invented 
for him, for under it he may clearly convey any fit pillage that cometh 
handsomely in his way. Besides this, he, or any men els that is dis- 
posed to mischief or villany, may under his mantle goe privily armed 
without suspicion of any, carry his head-piece, his skean [dagger], or 
pistol if he please, to be always in readiness." 

And so he continues, throwing in a passage about the 
uses to which women turned it, that could not well be 
reproduced her. 

2. Tight-fitting Upper Garments. — The tight-fitting 
sleeved upper garment was something like the present frock- 
coat ; but it had no collar, and was much shorter, usually 
reaching to about the middle of the thigh, and often only a 
little below the hips ; with a girdle at the waist. It was 
generally called by one or another of three names : — lenn or 
lend, inar, and Mine, line, or leinid : but Mine was also 
applied to a shirt, as well as to the saffron-dyed loose tunic. 
Persons are very often described as wearing this short 
coat with a brat or mantle over it : as for example in the 
Bruden Da Derga.* Cuculainn's charioteer wore a tight 

* O'Curry, Man. & Cust., 147, note 218 : LU, 95, a, 3 and . 




inar of leather, over which was a for-brat (' over-brat ') or 
loose mantle * Sometimes the tight coat was fastened at 
the throat with a brooch like the loose over-cloak. But 
lend, like many other terms for garments, was often used 
vaguely. The short coat is very well represented in the 
figures given below, which, however, belong to a com- 
paratively late time, but serve to show how this garment 
held on in fashion. 

A sleeve, no matter to what article of dress it belonged 
was denoted by two names :— lamos and munchille, both 

Fig. 231. 

Figures carved on a bone book -cover, Now in National Museum. Probably of the thirteenth 
or fourteenth century. "Five figures" — says Wilde — "engaged in some sort of game. . . . The 
external figures are represented in the act of throwing rings or quoits." The tight-fitting inar or 
jerkin well represented here : with striped sleeves and plaited skirts : confined by a waist-belt : all 
probably parti-coloured. (From \Y'ilde*s Catalogue, p. 320.) 

given in Cormac's Glossary (pp. ioo, 116). He derives 
lamos from lam-fhoss, the foss or case of the lam or arm. 
The other word he derives from man, the hand, and call, 
a case: mun-chille, equivalent to man-cail, 'hand- or 
arm-keeper.' Muinchille is the present Irish word for 
a sleeve. 

3. Cape and Hood. — The short cape, with or without a 
hood, was called cuchull or cocholl, corresponding in shape 
and name with the Gallo-Roman cucullus, English cowl: 

•Kilk. Archaeol. Journ., 1870-71, p. 423, 5, 8 , 36, 40 : LU, 79, a, top line 
and those that follow. 



but this English word cowl is now often applied to a hood 
simply. The cochull just covered the shoulders : and it is 
quite usual to find in the tales persons described as wearing 
" a short cochull reaching as far as the elbows."* Some- 
times in old writings the diminutive cochline [3-syll.], 
" little cochull," is used. Cuculainn wears a cochline ettach 
immi con urslocud for a dib n-ulendib,\ " a winged cochline 
about him with openings at the 
two elbows." Here the word ettach 

Fig. 232. 

Meeting of Mac Murrogh Kavanagh and the Earl of Gloucester in 1399. (From an illuminated 
contemporary English manuscript. Reproduced in Gilbert's Facsim. National Manuscripts, 
from which this illustration was copied,) The English appear on the left-hand side. 

' winged,' refers to the loose extremities of the sleeves 
flying open at the elbows, where they terminated. Both 
cape and hood were dyed in colours : Giraldus says 
that in his time they were made with variously-coloured 
pieces of cloth sewed together. The hood was called 
cenniud [ken-yu], from cenn, the head ; or more usually 
culpait. It covered the whole head except the face. The 
followers of Art mac Murrogh KavanaghJ (fourteenth 

* Man. & Cust., 138, l6 : Da Derga, 181 : Rev. Celt., xn. 87. 

t LU, 122, b, 28 , 29 : Demon Chariot, 376, last line ; 379, top line. 

X For whom see Joyce's Short History of Ireland, p. 323. 


century) are shown wearing these odd-looking hoods : and 
it is worthy of remark that the English standing beside 
them wear head-dresses and capes not very different. 
This hood was generally attached to the cowl or cape so 
as to form part of it ; as appears clear from the following 
examples. In the Bruden Da Derga, Ingcel, describing 
certain persons that he saw among many others in Da 
Derga's hostel, says that each wore a little cochall or cape, 
and a white hood {cenniud) on each cape, and a red tuft 
(cuirce, pron. cur-ke) on each cenniud, and an iron brooch 
in each cochull* [fastening it at the throat in front]. 
The three Pictish kings who were in the same hostel 
are described by Ingcel as having each a short black cowl 
with a long hood on itf ; and again he says, about three 
others, that they wore three short black capes reaching 
to the elbows, and hoods on the capes. This fashion 
continued long, for we find it mentioned in the story of 
O'Donnell's kern as in use in the fifteenth century. J In 
this same century, too, the hood was sometimes worn in 
the Scottish Highlands.§ Still later, Thomas Dineley 
(in 1675) observed that the men, in parts of Ireland, 
covered their heads with their cloaks. || Cochall is now 
applied, as anciently, to any short cape covering the 

The word cochall, like many other terms designating 
articles of dress, in ancient as well as in modern times, 
was often used loosely. It was applied to a monk's cowl : 
and the chasuble worn by a priest was sometimes called a 
" cassal or cochall."^ We know that the long leathern 
cloaks, reaching down to the feet, worn by the soldiers of 
Murkertagh mac Neill during his celebrated circuit through 
Ireland in the winter of a.d. 941, were called by thisname, 

* Man. & Cust., 11. 150, top paragraph. f Stokes, Da Derga, 181. 

J Silva Gad., 315, top line : Irish Text, 279, I0 . 
§ Rob Roy, Introduction. || Kilk. Archaeol. Journ., 1856-7, p. 186. 

UTrip. Life, 384, 4: 399, a3l LB, "Contents," p. 6, bottom. 


whence he is known in history as Muirchertach na g-cochall 
g-croicenn, i.e. ' of the leathern cochalls or cloaks.'* The 
fact that these long cloaks were called cochalls renders it 
pretty certain that they were furnished with hoods to 
protect the head ; a thing we might expect under the 

4. The Kilt. — The Gaelic form of this name is celt [kelt], 
of which " kilt " is a phonetic rendering. In Cormac's 
Glossary (p. 47) celt is vaguely explained by the Latin 
vesta, and also by the Irish edach, ' raiment ' : and in 
another old authority quoted by O'Donovan in his trans- 
lation, it is said to be " anything used as a protection." 
This seems nearest to its primary meaning : for celt means 
' concealing.' The word occurs so seldom, and is used so 
vaguely, that we might find it difficult to identify the 
particular article it designates, if the Scotch had not 
retained both the article itself and its name : for the High- 
land kilt is the ancient Irish celt. In Ireland the garment 
itself was very common, though it was seldom called a 
celt. On one of the panels of Muiredach's cross at 
Monasterboice are represented three soldiers dressed in 
kilts reaching to the kneesf : and all the figures on the 
shrine of St. Manchan — a work of about the eleventh 
century — are similarly attired — the kilts here being very 
decided and characteristic, as well as highly ornamented. J 
The kilt — commonly falling to the knees — is very fre- 
quently met with on the figures of manuscripts, shrines, 
and crosses, so that it must have been very much worn 
both by ecclesiastics and laymen. The kilt and the bratt 
outside it are seen in some of the figures of the illustration 
in vol. 1., p. 59, where also, as in all other representations, 
the plaits run up and down, like what we see at the present 
day. The present Highland article of dress is called kilt 

* Joyce, Short History of Ireland, pp. 197, 198. 
t Kilk. Archaeol. Journ., 1872, p. 109, Ia> 

* See on this shrine Stokes's Petrie, 285. 




everywhere except among the Highlanders themselves, 
whc usually designate it by another Gaelic term, filleadh, 
or more generally filleadh-beg (' little garment '), anglicised 

FIG. 233. 
The figures on one face of the shrine of St. Manchan (for which see vol I., p. 564, supra) : dating 
from about the eleventh century. They all represent laymen, and they diminish in size to the right 
to suit the shape of the panel. (From Kilk. Archxol. Journ. for 1874-5, P- I45-) 

In the story of the Tain we read that one of the games 
in which the boys of Emain contended was tearing off 
each other's outer garments — truly a rough play. The 
little boy Cuculainn entered the field against a number of 
them, and while they were not able even to disturb his 


brooch, he tore off the de-chelt from a number of them.* 
This de-chelt or ' double celt ' was a loose jacket and celt 
combined, as it is defined in Cormac's Glossary (p. 47) : — 
" Dechelt, that is to say, a brat and a leine " [joined] : 
whereas the celt proper extended only from the waist 

In several passages of the Bruden Da Derga persons 
are described as wearing berbroca, a term which both 
O' Curry and Stokes translate aprons : though Stokes in 
one place — and only one (Da Derga, p. 57) — makes it 
' drawers ' — apparently on the authority of Zimmer. The 
word is always used in LU in the plural number — berr- 
broca : but whether the singular is the same, or berbroc, is 
at present uncertain. The name of the article seems to 
indicate that it was an apron — or possibly some special 
sort of kilt — reaching down to the broca or shoes. For 
ben means to shave : berr-broca, ' shave-brogues,' because 
it just brushed them with its lower hem : exactly as the 
word tond, ' a wave,' is said in Cormac's Glossary (p. 161) 
to be derived from tondeo, ' I shave,' " because it shaves 
[berrad] the grass from off the sea-marsh," where, it will be 
observed, the same Irish word (berr) is used.f 

Of the four upper garments hitherto mentioned, 
Giraldus (a.d. 1185) notices two : — the cochall and the 
fallainn, with the trousers (to be presently dealt with 
here). He says : — " It is their custom to wear small 
" tightly-fitting hoods (caputium is the word he uses) 
" hanging the length of a cubit below the shoulders [i.e. 
" the cape to which the hood was attached hung so far] 
" and generally made of variously coloured-strips sewn 
" together."^ Three of them are mentioned in an ancient 

*Miss Hull, Cuch. Saga, 139: LL. 63, a, 6 from bottom: Kuno 
Meyer, Ventry, 83, ,. 

f On berr-broca see O'Curry, Man. & Cust., n. 147 (twice), 148, 149, 
183 : and Stokes, Da Derga, 57, 289, 309. Several of the original 
passages where the word is used will be found in pp. 94, 95, of LU. 

I Top. Hib., in. x. : Book of Rights, 38, note /. 

Fig. 238. 

F10. 237. 

Fig. 236. 

Fig 234 

10. 235- 


Fig. 239. 

Fig. 240. 

Fig. 241. 

Bronze pins and button : all very ancient. Figures 234, 235. and 236, drawn natural size. Those with 
circular disk-heads are generally very long: figure 240 is I3>4 inches, with a disk 2^ inches in diameter 
Figure 241, drawn natural size, a hijfhly-decorated bronze button, enamelled in red and green, with a 
a small metal fastening-loop behind. (All from Wilde's Catalogue, pp. 555. 537, 53 8 . 572') 


Irish poem copied at Armagh in 1139 by Mael-Brigte hua 
Mael-Uanaig, noticed by Stokes,* which states that on 
each of the Magi who came to visit the Infant Jesus were 
three [upper] garments {tri-etaige im each fer dib : " three 
garments round each man of them ").- The three were no 
doubt the cape with hood, the loose cloak, and the short 
tight-fitting coat (with of course the nether garment to 
correspond) : the Irish writer attributing to outsiders the 
fashion of his native country. 

Fasteners for Upper Garments. — The over-garments were 
fastened by brooches, pins, buttons, girdles, strings, and 
loops. Brooches will be treated of next section. Simple 
pins were generally ornamented, head, or shank, or both, 
as seen in the annexed figures, of which the originals are 
all in the National Museum, with many others. 

Nether Garments. — The ancient Irish wore a trousers 
which differed in some respects from that worn at the 
present day. It generally reached from the hips to the 
ankles, and was so tight-fitting as to show perfectly the 
shape of the limbs. When terminating at the angles it 
was held down by a slender strap passing under the foot, 
as seen in one of the figures in the Book of Kells.f Like 
other Irish garments it was generally striped or speckled 
in various colours. The usual Irish name was triubhas 
[truce], which is often correctly anglicised trews, and from 
which the modern word " trousers " is derived. The 
people of other ancient nations wore parti-coloured trousers 
as well as the Irish ; the Gauls and Britons for example ; 
among whom it was called braccae. The Romans saw this 
article of dress in general use for the first time among the 
Gauls : so that they gave the name Gallia Braccata to a 
part of Transalpine Gaul. It would seem, that the Irish 
and British trousers were also called braccae, I from which 

* In Rev. Celt., vm. 346. t Abbott's Reproductions, Plate 1. 

J Ware, Antiqq., 176 : Lynch, Cambr. Ev., 11. 213 : De Jubainville, 
Cours, Litt. Celt., vi. 371, 372 




comes the modern word breeches : and which, as some 
think, is itself derived from the Celtic brecc, speckled. 

In the time of Giraldus Cambrensis, the Irish trousers, 
hose, and shoes were all one garment : — " The Irish "—he 
writes — " wear breeches ending in shoes, or shoes ending 
in breeches"*: a fashion also described by Lynch in his 
" Cambrensis Eversus " (II. 209) : — " The breeches used by 
" the Irish was a long garment, not cut off at the knees, 
" but combining in itself the sandals, the 
" stocking, and the drawers, and drawn 
" by one pull over the feet and thighs. 
"It was not flowing (to use the words of 
" Tacitus), but tight, and revealing the 
" shapes of the limbs : not 
" unlike what Sidonius 
" describes : — 

A closely-fitting dress their limbs compresses, 
No trailing robe their legs conceals.' " 

This fashion continued in use to 
the time of Lynch, who tells us 
(II. 211) that the people generally 
abandoned it before 1641, partly in 
consequence of the exhortation of 
the clergy — which implies that that 
part of the Irish dress bordered on indelicacy — and partly 
of their own accord. But I do not find any statement 
that this combined garment was used in very early ages. 
It may be worth mentioning that there was an old canon 
of the Irish church, which is still extant, forbidding the 
clergy to wear those close-fitting trousers. The figures 
on the shrine of St. Manchan (p. 204, supra) have no 
trousers : but they wear long kilts reaching below the 
knees, with legs and feet bare. 

Fie. 242. 

Showing the titflit trews or trousers, 
with a/allaiim or short cloak, dyed 
olive -green. (From an illuminated 
copy of Giraldus of A.D. 1200. From 
Wilde's Catalogue, p. 311.) 

*Ton. Hib. ill. x. 


The trousers, as has been said, usually went below the 
ankles. But in some figures on the high crosses it termi- 
nates immediately below the knee, like the Irish knee- 
breeches of our own day : and two of the figures of the 
S.-E. cross of Monasterboice wear breeches terminating 
just above the knees, where they are closely bound, and 
fitting skin-tight on the thighs. 

Leggings of cloth, or of thin soft leather, were worn, 
probably as an accompaniment to the kilt. They were 
called ochra or ochrath. In " Mac Conglinne's Vision " (88) 
a person is mentioned as having " ochra encircling his 
shins." It will be observed that this word ochra is almost 
identical with the Latin term ocrea applied by the Romans 
to their leggings. The Irish leggings were laced on by 
strings tipped with findruine or white bronze, the bright 
metallic extremities falling down after lacing, so as to form 
pendent ornaments. Bove Derg's cavalcade had all of 
them strings [with tips] of findruine hanging from their 
ochraths* The ochra reached about to the ankle : for in a 
passage in one of the ancient Gaelic Triads it is mentioned 
that there was a has or handbreadth between the shoe and 
the lower edge of the ochrath.\ 

I think it likely that the trousers and kilt were not 
worn together : at least in all the kilted figures that have 
come down to us the legs are bare. As bearing on this 
point it is worthy of remark that there are many passages 
in our ancient literature showing that it was pretty usual 
with those engaged in war to leave the legs naked : a 
fashion perpetuated by the Scotch to this day. In the 
ancient account of the battle of Mucrime (fought a.d. 250) 
the jester Dodera says to Maccon, the leader of one side : — 
" Eoghan [one of the leaders of the opposing army] will 
" seek thee through the battle, and if he catch sight of thy 
** legs [colptha, legs or calves] he will strike thee down."J 

* O'Curry, Man. & Cust., II. 157. t #>«?., 107. 

X Silva Gad., 350 : Rev. Celt., xm. 441. 



And a little farther on we are told that " through the host 
" Eoghan saw Maccon's two calves which were as white as 
" the snow of one night " — the whiteness being noticed as 
a matter deserving praise. 

That it was customary to leave the legs naked is also 
shown by such personal names — or nicknames — as Niall 
Glunduff (black-knee) which was the name of a brave king 

Fig. 243. 
Croup showing arms and costumes of the sixteenth century. Irish soldiers and 
peasants, from a drawing by Albert Durer in 1521, preserved at Vienna. Over the 
two soldiers is an inscription in German : " Here go the war-men of Ireland beyond 
England." Over the three peasants : " Here go the poor men of Ireland beyond 
England." Between the two is the date a.d. 1521. (From Kilk. Archocol. Journ., 
1877. p. 396, where the original drawing is reproduced.) 

of Ireland (a.d. 916-919) : Amergin Glungel, of the white 
knee (gel, white) ; Brocshalach Crion-Ghluinech,* of the 
withered knee ; Irial Glunmar (big-knee) f : Gliin-iarainn 
(iron-knee). Eber Glunfhind was so called — says the 
Book of Lecan — " because he had white marks on his 
knee " {find, white). J Scott gives a corresponding expla- 
nation of the cognomen of one of the Mac Gregors of 
Scotland of two centuries ago, " Gregor Ghlune Dhu, Black 

* Silva Gad., 527, top line : Rev. Celt., xvi. 273. 

f Rev. Celt., xvi. 411, \ Ir. Texte, in. 409, 



" Knee, from a black spot on one of his knees, which his 
" Highland garb rendered visible," like Nial GlundufL* 
It was considered a blemish to have dark-coloured knees, 
as we see from a passage where it is said of a splendid- 
looking young man, that his dark-coloured knee was his 
only blemish. On the other hand, to have white legs 
and knees was considered a 
point of beauty, as in case of 
Lugaid Maccon mentioned 
last page. 

As illustrative of all that 
precedes, two series of cos- 
tumes of the years 1521 and 
1600 respectively are pre- 
sented here : pp. 210, 211. 

Fig. 24s. Fig. 246. 

Irish Costumes, A.D. 1600, (From map of Ireland published by Speed in 1611.) Figure 244, 
gentleman and lady of the high classes. Figure 24s. persons of the middle rink. Figure 246, 
peasants (Speed, after the fashion of Englishmen of the time, calls them " wilde Irish.") 

Underclothing. — Both men and women wore a garment 
of fine texture next the skin. This is constantly mentioned 
in the tales, and, whether for men or women, is denoted by 
the word Mine or line [2-syll.], which is now the common 
Irish word for a shirt. It was usually made of wool or 
flax. It is said of St. Columkille that — by way of morti- 
fication — he never wore linen or wool next his skin.f But 
sometimes it was made of silk, occasionally of satin, highly 

* Rob Roy, Introduction. t Three Irish Homilies, 123. 


ornamented. One party of Queen Maive's forces wore 
" pure- white shirts [Unti glegela] next their skin "* : and 
such expressions are very common. Sometimes a silken 
shirt was shot with threads of gold. In the Battle of 
Moylena, (p. 129) the hero Fraech Mileasach who was 
surprised sleeping in his tent, started up and had to fight 
for life in his " shirt of many devices ornamented with 
threads of gold." Very often the shirt is called caimse, 
a word which in Cormac's Glossary (p. 33) is derived 
from the Latin camisia. All these notices about shirts 
refer to the higher classes : whether the lower order of 
people wore shirts is a matter on which I have found no 

Girdles and Garters. — A girdle or belt (Ir. criss) was 
commonly worn round the waist, inside the outer loose 
mantle, and it was often made in such a way as to serve 
as a pocket for carrying small articles. We read in the 
Tripartite Life that on one occasion St. Patrick met six 
young clerical students having their books in their crisses 
or girdles. Sometimes a bossan or purse (also called 
spardn) was hung from the girdle, in which small articles 
were kept, such as rings. f The girdles of chiefs and other 
high-class people were often elaborately ornamented and 
very valuable. In the Brehon Law the value of a bo-aire 
chief's girdle is set down as three seds or cows, which 
might represent £40 or £50 of our present money : and 
those of higher chiefs and kings were still more costly. J 

Garters were worn, sometimes for use, and sometimes 
for mere ornament, or to serve both purposes. There are 
two words for a garter, ferenn and id ; and the use of the 
article is made quite clear by the explanation given in 
Cormac's Glossary (p. 72) : — " Ferenn, a garter (id) which 
is round a man's calf." Cormac goes on to say that the 

* Miss Hull : Cuch. Saga, 119 : LL., 55, a, bottom, and b, top. 
t Kuno Meyer, in Rev. Celt., XII. 460 : LL, 250, a, 23 , 3i , 
I Br. Laws v. 417. 


ferenn was made of different materials according to the 
rank of the wearer, and he instances the garter of a king 
as made of gold. This agrees with an expression in an 
ancient panegyric written on th? hero Couri mac Daire, 
king of South Munster in the first century, in which it is 
stated that he gave his bard, among other valuable pre- 
sents, garters of gold.* But this word ferenn was also 
used, according to Cormac, to denote a girdle : — " Ferenn " 
— he says — " is also a name for a girdle (criss) that is 
round the man " : and he gives as an instance " the snow 
reached the men's girdles " (ferna), referring to the snow 
the druids brought down by magic in their contest with 
St. Patrick at Tara. The Tripartite Life relates this 
incident fully, which bears out the correctness of Cormac's 
reference : the expression used in the Tripartite being 
cotoracht [in snuhta] fernu fer, " so that [the snow] reached 
the ferna or girdles of the men."f 

Gloves. — That gloves were commonly worn is proved 
by many ancient passages and indirect references. The 
common word for a glove was lamann or laminn, from 
lam [lauv], ' the hand or arm ') ; which is the word still in 
use. Cormac (p. 100) clearly defines laminn, when he 
derives the word from laim-inn, i.e. ' arm-end,' because — 
as he says — " the end of the arm is clothed by it." And 
in the mediaeval tract on Latin declension, edited by 
Stokes (p. 4, No. 34), the meaning is made equally clear, 
where lamann glosses the Greek word " chirotheka." 
Sometimes the word used was lamagan, which is a 
diminutive form. Another word for a glove was braccaile, 
which Cormac's Glossary (p. 19) also explains : — [From] 
" brace, a hand, and cail, a case [a case for the hand], 
i.e. a lamann or glove " : exactly corresponding with the 
Greek word given above, chirotheka, ' hand-case.' 

We often find notices of people wearing gloves. In 
one passage of the old tale called the Acallamh it is 

* O'Curry, Man. & Cust., 11. 152 f Trip. Life, 56, line 2. 


incidentally mentioned that two persons closing a contract 
by joining hands had first to remove their gloves, showing 
how usual it was to wear them in common life (see vol. I., 
p. 182, supra). In the " Vision of Mac Conglinne " (p. 90), 
an imaginary personage is spoken of as having " two 
glove < on his hands" (cona di lamainn bd lamaib). St. 
Patrick, when traversing the country in his chariot, wore 
gloves when necessary.* 

They appear to have been common among all classes — 
poor as well as rich. One of the good works of charity laid 
down in the Senchus M6r is " sheltering the miserable," 
which the gloss explains, " to give them staves and gloves 
and shoes (lorga, lamanna, cuaraind) for God's sake."f 
The evangelist depicted in the Book of Kells (fig. 230, 
p. 197, supra) wears gloves, with the fingers divided as in 
our present glove, and having the tops lengthened out 
beyond the natural fingers. 

Sometimes gloves were highly ornamented. In the 
Voyage of Maildune we are told that a certain lady in one 
of the islands visited by the voyagers, wore gloves on her 
hands " with gold embroidery " {lamanna co n-6rphill imma 
lamaib). % Besides the two names already given for gloves, 
there were two others, which appear to be very old words : — 
Bracand, which is given in Cormac's Glossary (p. 27), and 
derived from brae, a hand : and mana ma in O'Clery's 
Glossary (from man, a hand : Corm. 108) : both which the 
glossaries explain as meaning gloves. 

As to material : probably gloves were made, as at pre- 
sent, both of cloth and of animal skins and furs. We have 
an example of this last, where gloves were on one occasion 
made from part of a fur mantle worn by St. Molaise of 
Devenish.§ The importance and general use of gloves as 
an article of dress are to some extent indicated by their 
frequent mention and by the number of names for them. 

* Trip. Life, 295, 9 . % Rev. Celt., x. 65. 

f Br. Laws, in. 19. § Silva Gad., 33. 


Head Gear. — The men wore a hat of a conical shape, 
without a leaf, called a barred [barraid], a native word, of 
which the first syllable, ban, signifies top, and according 
to some, the second, id, is from the Irish eda, dress ; in 
which case barred would mean ' head-dress ' or " head 
cover.' But the word exists in several other languages, 
as French, barrette ; Spanish, birreta ; Italian, berretta ; 
and German, pirete ; all meaning a head-covering of some 
kind : which makes one doubt that the second syllable of 
the old Irish name is from eda. The word at also occurs in 
Old Irish documents for a head-covering, and of course 
is the same as the English hat, both derived from Norse 
hattr : but at has several other meanings in Irish which will 
be referred to farther on (p. 240).* The helmet and its 
designations have been treated of under " Warfare." 

Among the peasantry, the men, in their daily life, 
commonly went bare-headed, wearing the hair long behind 
so as to hang down on the back, and clipped short in 
front. Sometimes men, even in military service, when not 
engaged in actual warfare, went bare-headed in this 
manner. In the panels of one of the crosses at Clonmac- 
noise are figures of several soldiers : and while some have 
conical caps, others are bare-headed. Camden describes 
Shane O'Neill's galloglasses, as they appeared at the 
English court in the sixteenth century, as having their 
heads bare, their long hair curling down on the shoulders 
and clipped short in front just above the eyes.f 

Married women usually had the head covered either 
with a hood (caille, pron. cal-le) or with a long web of 
linen wreathed round the head in several folds. This last 
is probably what was designated in ancient writings by the 
word callad [cal-la] : a term different from caille, though 
no doubt derived from the same root. In the Feast of 
Dun-nan-gedh it is related that a certain queen cried 

* See also Ware, Antiqq., 177. 

t See Joyce, Short History of Ireland, p. 409. 




aloud in a fit of grief, and wrung her hands, " and cast her 
royal callad into the fire in presence of all." This word 
is now obsolete in Ireland, though it is retained in Scotch 
Gaelic to signify a cap or wig. But the other word caille 
is still used in Ireland for a hood or veil, from which again 
comes caillech (the veiled or hooded one), a nun : a different 
word from caillech, an old woman. It should be remarked 
that the veil was in constant use among the higher classes, 
and when not actually worn was usually carried, among 
other small articles, in a lady's ornamental hand-bag. 

There was another word for a 
woman's head-covering — melt 
— which had grown old in the 
ninth century, and which in 
Cormac's Glossary (page 120) 
is explained by cop-cailli, a 
woman's hood. The head- 
dress of a woman was also 
sometimes called cenn - barr 
(' head-top ') as in the Tain 
bo Fraich (p. 148, 2 o) : and in 
a Gloss on a portion of the 
Brehon Laws (iv. 28, 22) it is 
called cenbar no caille — ' head-dress or head- veil.' 

Foot- Wear. — The most general term for a shoe was 
brdc, brdcc, or brog (plural broga), which was applied to a 
shoe of any kind : it is still the word in common use, and 
it is correctly perpetuated in sound by the well-known 
Hiberno-English word brogue. The brog was very often 
made of untanned hide, or only half-tanned, free from hair, 
and retaining softness and pliability like the raw hide. 
This sort of shoe was also often called cuardn or cuarog, 
from which a brogue-maker was called cuardnaidhe 
[cooraunee]. A fox once stole St. Ciaran's brogues and 
proceeded to make a meal of them ; .but was caught just 
when he had eaten the ears and thongs : these must have 

Fig. 247. 
Portion of " a light gauzy woollen veil, 
of the most delicate texture" (Wilde). 
Found on the body of the woman men- 
tioned at p. 180, sufra. From Proc. Roy. 
Ir. Acad., IX. 103.) 


been of untanned hide* Shoes of untanned hide are worn 
to this day in the Aran Islands. Mac Conglinne (p. 8), 
before beginning his journey to Cork, made for himself two 
pointed cuarans of * brown leather " of seven doubles — > 
meaning seven folds or layers of hide in the sole — for his 

Fig. 249. 

Ancient Irish shoes, of tanned leather, in National Museum, Dublin. Figure 248 is a complete 
shoe, formed of a single piece. Figure 249 represents the upper only : it had a separate sole, 
which is gone : the ornament at top is a separate figure, and is merely an enlargement of the 
decoration at the top of the heel. The opus Hibernicum on the strap in the middle of this shoe 
betokens Christian origin but great antiquity. (From Wilde's Catalogue, p. 284.) 

long walk. When he arrived at the guest-house of the 
Cork monastery he took off his cuarans, and having washed 
his feet in the bath-tub, he next washed the shoes in it — 
the usual plan of cleaning the mud off shoes of this kind. 
The cuaran had generally a single sole ; but sometimes 
two or more thicknesses were used, as we see in the case 

* O'Grady, Silva Gad., 3, bottom. 


of Mac Conglinne. This shoe had no lift under the heel. 
The whole shoe was stitched together with thongs cut 
from the same hide.* 

There are two other words for a shoe common in 
ancient writings, as or ass and maelan : but these appear 
to have been applied to a more shapely shoe than the 
cuaran ; made of fully tanned leather, and furnished with 
a serviceable sole and heels. Many passages could be 
quoted showing that shoes were made of tanned leather : 
but the subject of tanning will be taken up again in 
chapter xxvi., p. 367. One example will suffice here : — 
We find it related that on one occasion St. Molaise gave 
some students, among other articles of clothing, " thick 

bark-soaked brogues as 
if of tanner's leather. "f 
Most of the shoes pre- 
served in the National 
Fic.aso. Museum are of tanned 

Small portion of a panel, showing the sandals under leather " but SOUie are 

the feet, with the rosettes. (From Book of Kells: 
Dr. Abbott's Reproductions, Plate XXXIV.) of Untamied llide.j 

Maelan is a diminutive of mael, blunt, and means a 
shoe or sandal with a wide rounded top — not pointed. A 
shoe of this shape is sometimes called ass or as (pi. assa, 
assai) and occasionally we find mael-assa, i.e. blunt-topped 
assa or shoes. Another compound of this word applied 
to a shoe is folasai, which is given in Cormac's Glossary 
(p. 76) : and O'Clery gives folasa as equivalent to " shoes." 
A shoe was also sometimes called iallachran (iall-acrann, 
' thong-shoe ') which in O'Clery's Glossary is given as 
equivalent to broga, i.e. shoes. Still another name for a 
shoe, according to O'Clery's Glossary, was foirtchi, 
connected with fortcha, clothing of any kind. 

Most of the figures depicted in the Book of Kells and 
on the shrines and high crosses have shoes or sandals, 

* See Ware, Antiqq., 178. t Silva Gad., 33. 

\ Wilde, Catalogue, 280. 


though some have the feet bare. One wears well-shaped 
narrow-toed shoes seamed down along the instep, some- 
thing like the shoes here represented (figs. 248, 249), but 
much finer and more shapely. Some have sandals con- 
sisting merely of a sole bound on by straps running over 
the foot : and in all such cases the naked toes are seen. 
On many of the sandals there are what appear to be little 
circular rosettes just under or on the ankles, one on each 
side of the foot — perhaps mere ornaments. They are seen 
in the figure of the angel, p. 197, supra ; and more plainly 
in fig. 250 on the opposite page, both from the Book of Kells. 

From a passage in the story of Da Derga's Hostel 
(p. 189) we may infer that the shoes or sandals were 
often fastened to the feet by two or three or more straps 
across the instep. In this passage the sandals (brdic) of 
a gigantic warrior are compared to two currachs or hide- 
boats, each with five thwarts or cross-benches, referring 
evidently to the five fastening straps : a record which, 
as we see, is corroborated by the figures in the manuscripts 
and on the crosses. The shoes of the higher classes were 
often highly and beautifully ornamented ; as we know, 
partly from the records, and partly from the specimens 
preserved in museums : as illustrated in figures 248, 249, 
and 251. 

In the tales we often find it mentioned that persons 
wore assa or maelassa or sandals of silver or of findruine 
(white bronze) . On one of the islands visited by Maildune 
and his people they see a lady richly dressed approaching 
them, with two sandals (da maelan) of silver on her feet.* 
Dermot, king of Ireland (a.d. 656-664), saw a lady in a 
chariot with two pointless shoes (da maelassa) of white 
bronze on her feetf : and any number of such references 
might be given. Such sandals must have been worn only 
on special or formal occasions : as they would be so 
inconvenient as to be practically useless in real everyday 

* Rev. Celt., ix. 491 ; and xxiv. 129, I0 . f O'Curry, Man. & Cust., 11. 160 




life. This seems also to be indicated by the fact that in 
at least one shoe — namely that of which measurements 
are given below — the sole was fastened on by leaden 
solder, and would at once give way under any rough 
usage, such as walking. As confirming this idea of 
temporary and exceptional use, we have in the Museum 
a curious pair of (ordinary leather) shoes — shown in the 
illustration — connected permanently, so that they could 

only be used by a per- 
son sitting down or 
standing in one spot. 
In whatever way 
and for whatever pur- 
pose the metallic 
shoes were used, they 
must have been pretty 
common, for many 
have been found in 
the earth, and some 
are now preserved in 
museums. There 
were tradesmen, too, 
who made and dealt 
in them ; as is proved 
by the fact that about 
the year 1850 more than two dozen ancient bronze shoes 
were found embedded in the earth in a single hoard near 
the Giant's Causeway. One of these was presented some 
years ago to the Museum of the Royal Irish Academy 
(now the National Museum), the dimensions of which 
were: — 13^ inches long; breadth, 4.3 inches; height at 
heel, 2\ inches ; height of instep, 3f inches ; weight, 9f 
ounces. This was larger than an- ordinary shoe or slipper, 
no doubt to allow for a thick woollen stocking ; or a wisp.* 

* Proc. Roy. Ir. Acad., V. 27, 28. For another bronze shoe, see Ulst. 
Journ. Archseol., iv. 23. 

Fig. 251. 

A pair of shoes permanently connected by straps: the two 
soles and the straps are cut out of one piece. Most ingeniously 
and beautifully made. (From Wilde's Catalogue, p. 287.) 


The custom of using bronze shoes descended far into 
Christian times. In the National Museum, Dublin, may 
be seen a shoe of this kind, of thin hammered bronze, 
engraved all over with an ornamental pattern (not the 
ofus Hibernicum) , and with the name " St. Brigit, patroness 
v of Kildare " (in Latin), dated 1410. 

The finding of bronze shoes, and in such numbers, is a 
striking illustration of how the truthfulness of many old 
Irish records, that might otherwise be considered fabulous, 
is confirmed by actually existing remains. 

Sometimes people placed soft wisps of hay or fine 
straw in the shoes under the feet. This we know from the 
Senchus Mor, which provides that if a man is delayed in 
the performance of certain legal functions by such neces- 
sary things as " changing the wisp of his shoe " (dlui 
n-assa : see for dlui, vol. I., p. 224), i.e. removing the old 
wisp and putting in a fresh one, he is exempt from blame 
and responsibility. The gloss on this makes the matter 
clear : — " That is, while the cleric is changing the wisp of 
" his as or his curan, i.e. a wisp of straw which is between 
" his foot and his shoe [brog) when his shoe (curan) is 
" hurting him."* A passage from an old Irish tale is 
quoted by 0'Curry,f in which Dill, the famous blind druid 
of Ossory, is made to say, "lam putting incantations on 
the wisp which is in my shoes ; " preparing some spell 
(see vol. 1., p. 224). 

Laws relating to Dress. — Whenever dress came under 
the eye of the Law there was much particularity. This 
happened sometimes when it became necessary to set forth 
the privileges of persons of different classes, as in the case of 
children in fosterage. On this point the following regula- 
tions are laid down in the Senchus Mor. Whenever a boy 
had clothes of washing materials he should have two suits, 
so that one might be worn while the other was in the wash. 

* Br. Laws, i. 268, 7 ; 269, 8 ; 301, 2S . 

f Man. & Cust , 1. 207, aa (and correction, p. xviii, bottom). 


The sons of kings, when in fosterage, were to have satin 
mantles, dyed scarlet, purple, or blue : the scabbards of 
their little swords should be ornamented with silver, and 
there should be brass rings on their camdns or hurling 
sticks : while the sons of lower grade chiefs had tin 
scabbards. The children of the ard-ri or of a provincial 
king should have their mantle fastened with a brooch 
ornamented with gold, bearing a crystal : and those of 
inferior kings with ornaments of silver. The sons of chiefs 
were to be dressed in red, green, and brown clothes, and 
those of inferior ranks in grey, yellow, black, and white. 
The Law goes on to lay down many other arrangements 
for the dress of foster-children of various social grades, the 
quality depending on the grade : and all were to dress 
in their best on Sundays and festival days.* We must 
suppose that the regulations made compulsory here for 
children in fosterage were merely what were commonly 
carried out by their fathers and mothers in their own 

3. Personal Ornaments. 

Legendary Origin. — In the ancient Irish tales and other 
records, referring to both pagan and Christian times, gold 
and silver ornaments, especially gold, are everywhere men- 
tioned as worn by the upper classes : and these accounts 
are fully corroborated by the great numbers of objects of 
both metals found from time to time in various parts of 
Ireland, and now preserved in the Dublin Museum, and 
elsewhere. Gold naturally figures more prominently in 
the old literature than silver : and so well was the general 
custom of wearing gold ornaments recognised, that the 
legendary annalists, after their . manner in such cases, 
thought it necessary to assign a distinct origin for it. We 
are told that King Tigernmas (who first smelted gold in 
Ireland : vol. 1., p. 69) was the first to introduce orna- 

* Br. Laws, n. 147, 149. 


merits of gold and silver : that another king, Muinemon, 
first caused necklets of gold to be worn round the necks 
of kings and chiefs ; and that a third, Fail-derg-doid 
(whose name signifies " of the red-ring-arms or ringers "), 
was the first to cause rings of gold to be worn on the 
hands of chiefs in Ireland. Perhaps these records are a 
dim traditional memory of the institution by these 
monarchs of certain orders of nobility or knighthood 
distinguished by peculiar gold ornaments : for we know 
that in Ireland there were knightly orders marked by 
some such badges (see vol. I., p. 99) . Most of the ornaments 
described in this chapter are mentioned as in use in 
Christian as well as in pagan times ; and records of them 
are found in ecclesiastical writings as well as in the lay 
literature. The manufacture of gold and silver ornaments 
was of native growth. M. Salomon Reinach, a Continental 
scholar, who has carefully examined this question, says of 
the gold ornaments in the Dublin National Museum : — 
" Of objects of gold attesting imitation of Greek or Roman 
models there is no trace."* 

In the National Museum there is a great collection 
of ancient artistic ornamental objects, some of pure gold, 
some of silver, and some of mixed metals and precious 
stones. All, or nearly all — of whatever kind or material — 
are ornamented in various patterns, some simply, some 
elaborately. Those decorated with the peculiar patterns 
known as opus Hibernicum or Irish interlaced work 
(described in vol. 1., p. 545) were made in Christian times 
by Christian artists, and are nearly all of mixed metals 
and precious stones. Those that have no interlaced work, 
but only spirals, circles, zigzags, lozenges, parallel lines, 
&c, are mostly of pagan and pre-Christian origin, many 
of them dating from a period long antecedent to the 
Christian era. Nearly all the gold objects, except closed 
rings and bracelets — and most even of these — belong to 

* Rev. Celt , xxi. 75. 




this class — made in pagan times by pagan artists. All 
the articles of gold are placed in one compartment of the 
Museum, and they form by far the largest collection of the 
kind in the British Islands : eleven or twelve times more 
than that in the British Museum.* 

Rings and Bracelets. — Among the high classes the 
custom of wearing rings and bracelets of gold, silver, and 

findruine (white bronze) 
on the fore-arm, wrist, 
and fingers — including 
the thumb — was uni- 
versal, and is mentioned 
everywhere in ancient 
Irish literature. The 
words for a ring, 
whether for finger or 
arm, are fail (pi. fdilge) : 
fdinne [faun-ye] : nasc, 
which was applied to a 
ring, bracelet, collar, or 
tie of any kind — obvi- 
ously cognate with Latin 
nexus, a tie : and some- 
times flesc and tinde.-f 
The word id was applied 
to a ring, collar, circlet, 
or chain ; thus Moran's 
judgment-collar was 
called id M or din. The ordnasc and ordus were rings for 
the ord or thumb : the dornasc was for the wrist (from 
dorn, the shut hand) : the fiam was worn round the neck. 
Still another name for a bracelet or circlet or ring was 

Fig. 253. 

l-'igure 252, Irish Bracelet or Armlet, of solid gold. 
It is double the size of the picture, of beautiful shape 
and Workmanship, and weighs 3H oz. Many of the Irish 
bracelets were of gold, like this : but many also were 
bronze, one of which is shown in figure 253. Both ill 
National Museum. From Wilde's Catalogue, pp. 53 
(Gold) and 570. 

* At the end of this chapter will be found a short list of the gold objects 
in the National Museum, Dublin. See vol. I., p. 556, for a comparison of 
the gold collections of the Dublin and British Museums. 

t Tinde, Corm. 58, under "Doss": flesc, Keat., 162. 


buinne or bunne [2-syll.]. These several names were no 
doubt applied to rings of different makes or sizes : we 
know for instance that fail and fdinne are distinguished 
in the Tain bo Quelna* : but these distinctions have been 
in many cases lost. 

A passage in the Bruden Da Derga describes nine 
harpers, each with a crystal ring {fail) on his hand, and a 
thumb-ring (ord-nasc) on his thumb. The lady Bee Fola, 
going through a wood, sees a young warrior whose two 
arms were covered with bracelets (failge) of gold and 
silver up to his elbows, f According to the Book of Rights 
(p. 7) when the provincial kings attended the supreme 
monarch at the meeting of Ushnagh, each 
was bound to wear on his hand a bunne 
niad d'or dearg, a ' hero's ring of red gold,' 
which, on the breaking up of the assembly, 
he left on his seat as a sort of tribute 
and mark of respect to the high king. So fig. 2S 4. 

jealous were the monarchs of this privilege ££!!£*£ 
that on one occasion a provincial king was j" the Na "°" al , Museum - 

* O (From Wildes Catal., 

expelled from the assembly for neglecting Go,d 'P- SL| 
to bring his ring. It was the custom with- some warriors 
to wear a ring of a certain kind for every king they had 
killed in battle. Lugaid Laga (or Lewy Law), a famous 
Munster champion, wore seven bunne in commemoration 
of seven kings he had slain at different times : whence 
King Cormac (whose father Art was one of the seven) 
says of him : — " In case of Laga his hand does not conceal 
that he has slain kings. "J 

Both men and women belonging to the highest and 
richest classes — as for example King Nuada's wife, a 
Leinster lady — had the arm covered with rings of gold, 
partly for personal adornment and partly to have them 

* O'Curry, Man. & Cust , 90 : LL, 54, a, 33. 
t O'Curry, Man. & Cust., II. 161. 

% ibid., 11. ^56, x^^. 


ready to bestow on poets, musicians, story-tellers, and 
ollaves of other arts, who acquitted themselves satisfac- 
torily.* Cailte, once travelling through Connaught with 
the little musician Cas Corach, meets a chief's wife with 
her attendants. She asks him who the little man was : 
" The best musician in Erin or Alban," said he. " He must 
be very skilful," replied she, " if his music is as good as 
his countenance." So the little harper took his timpan 
and played a tune, which so charmed her that she gave 
him the two gold fails she had on her wrists, f This custom 
is like that of ladies of the present day, who often wear 
many thin bracelets together on the wrist — though not to 
give them to poets or musicians. Circlets of gold, silver, 
or findruine were also worn round the legs above the ankle : 
but these have been already noticed. Fully answering to 
all these entries and descriptions we find in the National 
Museum in Dublin, and in other museums, gold and silver 
rings and bracelets of all makes and sizes : some pagan, 
some Christian. 

Precious Stones and Necklaces. — Ireland produced gems 
of many kinds — more or less valuable — which were either 
worn as personal ornaments by themselves — cut into shape 
and engraved with patterns — or used by artists in orna- 
mental work. Precious stones are often mentioned in 
ancient Irish writings, the term commonly used being lec- 
logmar or lia-logmar : lee or lia, ' a stone ' : Idgmar, ' very 
costly, precious ' (from log or luach, ' price '). In Kerry 
were found — and are still found — " Kerry diamonds," 
amethysts, topazes, emeralds, and sapphires : and several 
other precious stones, such as garnet, were found native 
in other parts of the country. J In crannoges and pre- 
historic sites in various parts of the country, have been 
found beads, rings, and other small ornaments, of such 

* Man. & Cust., 169 : Bee Fola, 196, 197. 

f Man. & Cust., 11. 169, 170. 

% Ware, Antiqq., 172 : Petrie, Tara, 195. 


stones as red jasper, rose-coloured quartz, jet, amber, 
diorite, &c* 

A pearl was usually designated by the word sed [shade], 
old form set : but this word, as we shall see in chapter 
xxvii., sect. 4, was also applied to a cow regarded as an 
article of value or exchange ; and it was often used to 
designate a gem or jetvel of any kind. Sed or sead is still 
in use in this last sense. Several Irish rivers were formerly 
celebrated for their pearls ; and in many the pearl mussel 
is found to this day. Solomon Richards, an Englishman, 
who wrote a description of Wexford about the year 1656, 
speaking of the Slaney, says : " It ought to precede all 
" the rivers in Ireland for its pearle fishing, which though 

Fig. 255. Fig. 256. Fig. 257. 

Beads or studs of jet. In National Museum. Used as buttons or fasteners, or strung together 
for Necklaces. (From Wilde's Catalogue, p. 241.) 

" not abundant are yet excellent, for muscles are daily 
" taken out of itt about fowre, five, and six inches long, in 
" which are often found pearles, for lustre, magnitude, and 
" rotundity, not inferior to oriental or any other in the 
" world."f O'Flaherty (Iar C, 53) states that in the Fuogh 
river or Owenriff, flowing by Oughterard in Galway, 
" muscles are found that breed pearles " ; and to this day 
they are often found in the same river. In Harris's Ware 
(Antiqq., 172) it is stated that pearls are found in the fresh- 
water mussels of the Bann, as well as in those of several 
streams of Tyrone, Donegal, and elsewhere ; and Harris 
goes on to show that a present of an Irish pearl was made 
to Anselm, archbishop of Canterbury, by Gillebert, bishop 

♦See an article by Mr. Knowles in Kilk. Arch. Joum., 1879-82, p. 522. 
fKilk. Arch. Joum., 1862-3, p. 0.1. 


of Limerick, about 1094. The same authority (Antiqq., 
p. 178) quotes a record of Nennius that the kings of the 
Irish wore pearls in their ears : but he gives no reference, 
and I have not been able to find the original passage in 
Nennius. Petrie says that he has not found the use of 
pearls in any Irish ornament older than the fourteenth 
century* : and I do not remember seeing them mentioned 
— or at least any stones that could be identified with 
pearls — as used in personal ornament, in any of the old 
Irish writings ; though there are several rivers and places 
in Ireland that derive their names from scds or precious 

Of the various ornaments worn on the person, the 
common necklace was perhaps the earliest in use. Neck- 
laces formed of small 
shells are common among 
primitive people all over 

the world, and they have 
Fio. 258. Fig. 259. . , , . 

been found with skele- 

Gold Beads: portions of Necklaces: natural size. 
In National Museum. (From Wilde's Catalogue, Gold, tOnS Under Cromlechs in 
P- 37) 

several parts of Ireland, 
of which specimens may be seen in the National Museum 
in Dublin, belonging to prehistoric ages. In historic times 
necklaces formed of expensive gems or of beads of gold 
were in use in Ireland. The word for necklace in Cormac's 
Glossary is cuibrech-braiget (or cuimriug-braigef), i.e. 
' neck-binder ' {cuibrech, a yoke or binder : brage, modern 
braghad, the neck, gen. braigef) : but this has long gone 
out of use, the present term being ursgar-bhraghaid 
[ursgar-vraw-id]. Cormac notices the necklace under the 
word base, which he states was an old term denoting ' red ' ; 
and he goes on to say that it was also a name for a neck- 
lace, but that the necklaces called base were properly those 

* Stokes's Life of Petrie, p. 305. 

t For more information on Irish pearls and on places taking their names 
from them, see Joyce, Irish Names of Places, 11. 375. 


made of " draconic beads " (dona mellach dracondai).* 
This alludes to the draconite or dragon-stone, a red jewel- 
stone, which, according to the legend perpetuated by Pliny 
and Solinus, was taken from the brain of a living dragon. 
These dragon-stones are mentioned in even an older Irish 
authority than Cormac's Glossary — the Fled Bricrenn : and 
in a manner that shows it was considered very beautiful 
and valuable. When Cuculainn visited Ailill and Maive 
in their Connaught palace, a golden goblet of wine was 
given to him, and two dragon-stones the size of his two 
eyesf : and elsewhere in the same old tale a splendid- 
looking young hero is described with ocht n-gemma deirg 
dracondai for lar a da imlisen : [the brightness of] " eight 
" gems of red dragon-stone in the middle of his two eye- 
" pupils." 

Whether real dragon-stones found their way to Ire- 
land in those early ages is a question that cannot now be 
determined with certainty. Indeed the words of Cormac's 
Glossary, quoted above, would seem to imply that the 
necklaces in use in his time with the name' of base were 
made of gems which were not real dragon-stones, but 
only stones like them. Still we see, from the Glossary 
and from the other passages quoted above, that this stone 
was known in Ireland : and as there was communication 
with the Continent from very early ages, it is quite possible 
that real dragon-stones may have been occasionally used 
among the higher classes of the Irish. 

I do not know if carbuncle is mentioned as having 
been worn on necklaces in Ireland : but according to the 
records it was much used in artistic metal work : and it 
is very often noticed in Irish writings. The Irish name 
was carmogal, with some slight varieties of form : — 
carmhogal, carbunculus, Zeuss, 42, 8. The house built 

* Corm. Gloss., 20 ; Ir. Text in Three Ir. Glossaries, p. 7. 
f Ir. Texte, 1. 284, 285 : Henderson, 79 : see also Voyage of Bran, 
1. 8 f verse 12. 


by Bricriu for Concobar mac Nessa was ornamented with 
gems of carmogal (co n-gemaib carrmocail)* In Tara, as 
we find mentioned in the old account (Petrie, 192), there 
were a hundred and fifty drinking-vessels ornamented 
with gold, silver, and carmogal ; and part of the stipend 
paid by the king of Ireland to the king of the Gailenga 
consisted of twenty splendid bridles adorned with red 
bronze and carmogal.] According to Petrie (Tara, 195) 
the word carmogal was " applied loosely by the ancient, 
" Irish to any shining stone of a red colour, such as garnet, 
" a production of the country " : but as in the case of 
dragon-stone, real carbuncle may have found its way 
hither ; though no doubt in the greater number of cases 
the stones called carmogal were only imitations. 

There was a sort of necklace called an episle or epistil. 
In the " Vision of Mac Conglinne " (39, 10) certain persons 
are mentioned as sitting round a fire, each having seven 
episles round his neck : and the celebrated judgment-collar 
of the just judge Morann, which is called a sin in Cormac's 
Glossary (p. 152), and id eleswhere, is called " a sin or an 
epistil that was round the neck for declaring truth," in a 
later copy of the same Glossary. It is also called eibistil 
in an ancient treatise on Irish ordeals in the Book of 
Ballymote, where moreover there is the legend that Morann 
got it from the Apostle Paul.J It seems obvious that this 
term made its way into Irish literature under Christian 
influences : and that it originally meant a few verses from 
one of the Epistles enclosed in a little case or box, and 
hung round the neck as part of a necklace : as Roman 
Catholics now wear a scapular. 

Torques or Muntorcs. — Besides the necklaces properly 
so called, there were various kinds of gold and silver 
ornaments for wearing round the neck, of which perhaps 

* Fled Brier., 2, I?: Ir. Texte, 1. 254, 1St l6 . 

f Bk. of Rights, 267, verse 5. For other examples, see Man. & Cust., 
II. 190, top ; and Tain bo Fraich, 137. J * r - Texte, in. 208. 


the best known was the torque (Ir. tore). The torque was 
often formed of a single square or triangular bar of gold 
from which the metal had been hollowed out along the 
flat sides, so as to leave four, or three, ribbons along the 
corners, after which it was twisted into a spiral shape, 
something like a screw with four, or three, threads. There 
is one in the Museum only half made, having three leaves 
or ribbons the whole length, untwisted. But they were 
formed in other ways, as may be seen by an inspection of 
those in the Museum. Torques are repeatedly mentioned 
in our literature, sometimes by the simple name tore, but 
generally by the word muntorc, 
i.e. ' neck-torque,' from muin, 
the neck. When the great 
King Cormac mac Art (a.d. 
254) was arrayed in his kingly 
robes at the Feis of Tara, he 
wore his muntorc of gold about 
his neck* : and we read that 
when King Eochaid Airgthech 

(A.D. 298) WaS buried, hiS jeWelS Gold torque: in National Museum: 

1 j 1 • , r-r i5# inches in diameter : found in 1810 in 

were placed on his stone comn, a mound at Tara . From P e tri e S Ta ra. 

among which was his silver p,l8l ' ) 

muntorc.] In the Irish version of the Aeneid it is stated 
that one of the presents Aeneas gave Dido was a muntorc 
oir : Williams, the editor, translates it ' golden necklace ' : 
but it was not a necklace properly so called. J There 
are in the National Museum in Dublin many muntorcs 
or various shapes and sizes. Some are barely the size 
of the neck, while others are so large that when worn 
they extended over the breast almost to the shoulders : 
and there are all intermediate sizes. A number of gold 
torques are figured in a group at p. 13, vol. I. . of this 
book, of which the two large outer ones were found at 

* Q'Curry, Man. & Cust., n. 180. f Voyage of Bran, I. 48, 52. 

J Zeitschr. Celt. Phil., 11. 434, „ 


Tara in the year 1810. (The largest is shown separately 
here in figure 260.) The one represented in figure 261 is 
of unusual make, being formed by twisting a single 
plate of gold, and having two apples or balls of gold 
at the ends. The custom of wearing torques, as well as 
rings and bracelets, was in ancient times very general, not 

Fig. 261. 

Gold Torque, half the size of the original, which is now in the National Museum: 
found near Clonmacnoise. (From Wilde's Catalogue, Gold, p. 74.) 

only among the Irish, but among the northern nations, 
both of Europe and Asia, especially the Gauls, as all who 
have read Roman history will .remember. The statue of 
the Dying Gladiator has a torque round the neck, almost 
the same as some of those in our museums.* 

• A fact first noticed by Robert Ball, ll.d., in a paper read before the 
Roy. Ir. Acad, in 1854, and published in the Proc, VI. J53. For torques, 
see Petrie's Tara, 181-4 : the article in Kilk. Arch. Journ., 1883-4, p. 182 ; 
and Wilde's Catalogue, Gold, 


Crescents, Gorgets, or Necklets. — The word muince 
[moon-ke] denotes a neck-circlet, from muin, the neck. 
It was used in different applications, as for instance to 
denote the ferrule of a spear, and also the collar round 
a greyhound's neck : but it is as a personal ornament 
that the muince concerns us here. The word seems to 
have been applied to almost any kind of neck ornament. 
Thus Maildune, entering a certain house, saw " a row of 
" muniorcs of gold and silver, like [in size] to the hoops of 
" a vat " : but a few lines farther on, in the paragraph 
where the same incident is re-told, one of these muntorcs 
is called a muince* Nevertheless the necklets that we 
find constantly mentioned in the ancient tales by the 
names muince are to be generally understood as golden 
gorgets or collars for the neck, worn by both men and 
women, now often conveniently called " crescents." Thus 
a lady is described as wearing " a muince of burnished gold 
round her neck "f : and when Conn the Hundred Fighter 
was preparing to engage in the Battle of Moylena, he put 
on the various articles of his kingly apparel : and among 
the rest, his muince round his neck.J 

These golden crescents are of three main types. The 
first is quite flat, thin, and brightly burnished. Most 
of those of this kind are ornamented in delicate line 
patterns, which are thus described by Wilde (Catalogue, 
Gold, 10) : — " The ornamentation, which is very minute 
' and elaborate, was in this, as in almost all similar 
' specimens, evidently effected by a series of fine chisel- 
' edged punches, the indentations made by which can in 
' some instances, be observed on the plain reverse side. 
' The lines which surround the edges would however 
' appear to have been produced by the graver." But it 
is probable that all the lines were produced by punches, 
as Mr. Johnson — an experienced goldsmith — has stated 

* LU, 23, a, 34 and b, 4 : also Rev. Celt., ix. 477, 478. 

j Man. $c Cust., n. 160 and note. J Ibid., 179. 


was the case with the line-ornamentation of another 
class of gold objects (vol. L, p. 566, supra). Crescents 
of this kind are often called by the name lunula or 
lunette. Figures 262 and 263 represent two of those 
beautiful objects, of which there are now more than 
thirty in the National Museum. 

Fig. 262. 

Gold Crescent, Mttinct, or Necklet of the first type, one continuous bright plate : sometime* 
called lunula or lunette. Diameter 9 inches: opening for the neck, %yi inches: weight, ■$% oz. 
Found near Killarney. Now in National Museum, Dublin. (From Wilde's Catalogue, Gold, p. II.) 

Any thin strip or plate of metal, whether of gold, 
silver, or findruine (white bronze), was called land or lann, 
i.e. 'blade': and if it was smooth and polished, it was 
usually designated niam-lann [neev-lann], ' lustrous-blade.' 
This term was often applied to those bright flat crescents 


or muinces now under consideration. Each of the seven 
horsemen who formed the retinue of the fairy chief Bove- 
Derg wore " a niam-lann of radiant gold round his neck."* 
It will be seen farther on that a long strip of gold or other 
metal for the forehead was also called lann. 

The SECOND type, and by far the most elaborate, is 
dish-shaped in general make, convex on one side, concave 

Fig. 263. 

Another specimen of gold Crescent of first type : 7 inches in diameter : the 
opening, S/i inches : weight, 18 dwts. In National Museum. (From Wilde's 
Catalogue, Gold, p. 14.) 

on the other : covered all over with ornamental designs. 
The illustrations (on next two pages) give a good idea of 
the general shape, but represent the ornamentation only 
imperfectly. There are five specimens of these gorgets 
in the Museum, all of very thin gold. Both the general 
convex shape and the designs were produced either by 
stamping, or more probably by hammering with a mallet 

♦Man. & Cust., II, 157, note, col 2, line 1. 




and punches on a shaped solid mould. The designs are all 
raised from the surface (with corresponding hollows on 
the back) ; and in this respect they differ from those of 
the other two kinds of crescent in which the lines are 
indented. The patterns and workmanship on these are 
astonishingly fine, showing extraordinary skill of mani- 
pulation : they are indeed so complicated and perfect that 

Fig. 264 

Cold Muincc, Crescent, or Gorget of the second type : the largest and most 
beautiful of this kind in the collection. " The arched or lunated portion " — says 
Wilde — "consists of three elevated rolls, with rows of conical studs on each— four on 
the upper, and three on each of the two others. A very minute rope-shaped fillet 
occupies the sunk space between each two elevations." Diameter n inches : weight, 
16^ oz. Found in County Clare. Now in National Museum. (From Wilde's Cata- 
logue, Gold, p. 12.) 

it is difficult to understand how they could have been 
produced by mere handwork with moulds, hammers, and 
punches. Yet they could have been done in no other way. 
The circular bosses at the ends of these gorgets deserve 
special notice. Two of them are shown half-size at page 
238. They were made separately from the general body 


of the crescent, to which they are securely fastened : and 
the ornamentation on them is of extraordinary delicacy 
and beauty. Each of the circular ornaments forming the 
rows between centre and edge consists, in one specimen, of 
three delicate raised concentric circles, in another of six, 
and in a third of seven, each series of circles round a 
central conical stud or button, with point projecting out- 
wards : and in the centre of the whole boss is a large 

Fig. 265. 

Another specimen of gold Crescent of the second type: now in National Museum. 
Nearly n inches in diameter : weight, j'/i oz. (From Wilde's Catalogue, Gold, p. 25.) 

projecting stud of the same shape : all of pure gold. Each 
boss consists of two saucer-shaped discs, fastened (not 
soldered) together all round the edge, with the convex 
sides* outwards, so as to enclose a hollow space. Wilde 
thus describes the disc shown in fig. 266 {i.e. the fr,ont disc 
of the two that form the complete boss) : — " It is com- 
" posed of a very thin plate, most elaborately tooled, and 
" hammered into a high centra} umbo, surrounded by nine 




' cones, each encircled with a series of minutely-raised 
" lines of the most delicate tracery. A transversely 
" decorated bur surrounds the edge, and another of a like 

'description encircles the central elevation."* 

Of the five gorgets of this class in the Museum, Wilde 
truly observes : — " It ma)' with safety be asserted that 
" both in design and execution, they are undoubtedly the 
" most gorgeous and magnificent specimens of antique 
"gold work which have as yet been discovered in any 
" part of the world."f In weight they vary from four to 

Fig. 266. 

Fig. 267. 

Two of the gold Bosses (front view) at the ends of the Crescents of second type : described on 
pp. 236-7-8. Drawn half size. (From Wilde's Catalogue, Gold, p. 26.) 

sixteen ounces : and taking material and workmanship 
into account they must have been of immense value in 
their time. 

The necklets of the THIRD kind, of which the Museum 
contains five specimens, are of a semi-tubular make, the 
plate being bended round so as to form, in some specimens, 
about a half tube, in others less than half. The gold is 
much thicker than in those of the other two types. The 
one represented in fig. 268, which is the largest and most 
perfect of the five, is ornamented at the ends with a 
punched herring-bone pattern. In an adjacent case of 

* Wilde, Catal., Gold, p. 26. 

f/did., p. 19. 


the Museum are five models of the type of these five real 
ones, of which the originals — all pure gold — were found in 
Clare in the great hoard mentioned below. 

All the muinces of the three types were intended, and 
were very suitable, for the neck. The inside circular- 
opening is in every case of the right size, and on account 
of the flexibility of the plates they can be put on and 

Fig. 268. 

Gold Crescent or Necklet of the third type: in National Museum: described 
above: 7 J4 inches across on the outside: opening, $yi inches; weight, a little over 
70Z. (From Wilde's Catalogue, Gold, p. 34.) 

taken off with perfect ease, even though the opening at 
the ends is only a couple of inches, or less. What Wilde 
says of those of this third kind (fig. 268) — which he set 
down as gorgets — applies to all : — " As, owing to its shape 
" and material, it is very flexible, it can be easily passed 
" round the neck by bringing one end forward and pressing 
"the other backwards."* This he found by actual trial — 
as any one else may do. 

'* Wilde, Catal., Gold, p. 34. 


As to the splendid crescents of the second type (figs. 
264, 265), the elaborateness as well as the exquisite beauty 
of the bosses at once settles the question as to how they 
were worn. When on the neck the ends were in front, so 
as to exhibit the bosses to full advantage : and of course 
all the other necklets were worn in the same manner. 
Indeed the more simple necklets — those of the first and 
third types — show, of themselves, that this was the manner 
of wearing them : for where there is ornament at all, the 
ends are much more highly, decorated than the rest, as 
may be seen in the illustrations. 

The opinion of Sir William Wilde and others, that the 
crescents of the first and second types (figs. 262 to 265) were 
diadems and worn on the head, will be examined farther 
on (p. 251). 

The Do-at and the Muince-Do-at. — In connexion with 
this part of our subject, we now come to an Irish 
descriptive epithet, do-at, used as part of two compound 
terms, and — so far as I am aware — of only two : — muince- 
do-at and bunne-do-zt, which are often met with in the 
old tales to designate, respectively, two — and only two — 
classes of objects. 

At each extremity of all the muinces or crescents of 
the three types is a disc or boss or button — seen in 
the illustrations — generally circular, or nearly so : very 
elaborate in one of the types, simple in the other two. 
Their primary use was as fasteners, to catch the orna- 
mental string by which the necklet was secured. These 
terminal appendages were known in ancient Irish records 
by the name of at. In Zeuss, 67, 21, ait glosses ' tuber ' : 
and to this day it has the same meaning, namely, a 
' swelling ' of any kind : but the special sense here is a 
terminal knob, button, or disc. In accordance with this we 
find these gorgets — of whatever kind — designated muince- 
do-at, ' the necklet of the two ats or terminal discs ' (do, 
two : do-at has the same form in the nom. and gen.) 


Ferceirtne the poet, lamenting the death of his master 
Curoi-mac-Daire, king of South Munster, states that he 
received as presents from him, among other precious 
articles, many a muince-do-at* 

The Bunne and the Bunne-do-at have now to be con- 
sidered. The word bunne [2-syll.], with various forms^ 

Fig. 269. 

Fig. 270. 

Fig. 271. 

Three examples of the gold Bunne-do-at or fibula, all drawn half size: all in National Museum. 
Figure 269, hollow; weight, 2% oz. Figure 270, solid: over 3^ oz. Figure 271, hollow : 5;^ or, 
(From Wilde s Catalogue, Gold, pp. 53, 55, 57.) 

buinne, buinde, bunde, bouinde> has several significations. 
It denotes : — I, a wave or stream : 2, a branch : 3, a tube 
or anything like a tube or cylinder ; for instance the pipe 
of a spout, a musical pipe, the snout or horn of an anvil 
(ur-buinde), the round, thick edging on the top of a wicker 
basket, or any such twisted or corded rim ; and in the 
Rennes Dinnsenchus it is used to denote a man's shin- 

♦Man. & Cust., 11. 179. 




bone* : 4, a ring or circlet, either completely closed up 
or having a narrow opening or gap at one side. We 
are concerned here only with the last two applications 
(3 and 4). That bunne was applied to a hand- or finger- 
ring is placed beyond doubt by this fact : — that while in 
one part of the Book of Lecan we are told that Lugaid 
Laga had seven bunncs of gold on his hand or fingers, in 
another part of the same book, when the statement is 
repeated, they are designated seven rings {fdilge) of gold. f 
Under this name gold and silver rings of various forms 
are often mentioned. Among the treasures promised by 

Fig. 272. 

Gold Bimnt-do-at : in the National Museum : the largest of the whole collection : drawn one-third 
size. The buttons, or discs, or do-ata, arc- unusually large — each 5 inches in diameter. The whole 
ornament is 11 inches in length : weight, nearly 17 oz. (From Wilde's Catalogue, Gold, p. 58.) 

Queen Maive to Ferdiad to fight Cuculainn is " a great 
reward in rings" {buinne)%: and in the Book of Rights 
(p. 6, 16) we read of a bunne of red gold for a king's hand. 
In the same Book of Rights (p. 75) we find mentioned 
mantles, each with a bunne, d'or, which O'Donovan explains 
" a ring-clasp of gold " : and in the Voyage of Maildune 
a splendidly -dressed lady wears a bunne of gold round 
her hair. This brings us to the consideration of a class 
of gold articles in the National Museum : open rings with 
ats or buttons at the two ends, now commonly called 

. *Rev. Celt., xvi. 44. 

t O'Curry, Man. & Cust., II. 177, bot. : see also p. 225, supra. 
%lbid., 414, 415, bot. 



fibula, of which three typical illustrations are given on 
p. 241. Other examples are on pp. 242-3-4. 

These have been somewhat of a puzzle : all the more 
so inasmuch as they are far more numerous and more 
varied, both as to shape and size, than any other class of 
articles among our gold antiquities. There are such 
numbers of them that they must necessarily have been 
mentioned in the records, as well as the other gold 
articles : but no one has hitherto identified their Irish 
name. There are altogether about 150 of them, varying 
in size from the great specimens pictured in figs. 272 and 
275 down through all gradations to the two diminutive 
ones shown in their real size in 
figs. 273 and 274. As in case 
of other articles of native manu- 
facture, some have been found 
half made — left for some reason 
unfinished, like the torque men- 
tioned at p. 231, supra* But 
however they vary in pattern 
all have the disc or button on 
each of the two ends, already noticed in connexion with the 
crescents. Seeing then that the terms muince-do-at and 
bunne-do-at were anciently applied to two classes of gold 
objects, and that there are two, and only two, classes now 
in the Museum to which the epithet do-at would apply, one 
}he muinces or necklets, the other the objects now under 
consideration, it obviously follows that the " fibula " are the 
very ornaments called bunne-do-at in the ancient writings. 
A plain ring, whether closed or open, was a bunne or 
buinne simply : an open ring with the two terminal 
buttons was a bunne-do-at. f 

* See for example Journal of Cork Archaeological Society, 1902, p. 230. 

•f It is right to state that, so far as I am aware, this explanation of 
muince-do-at and butme-do-at, and their identification, with the actually- 
existing objects, have not been given before, and that I am entirely- 
responsible for them. On all this see also Appendix, infra. 

Fig. 273. Fig. 274. 

Two specimens of the very small gold 
Bunne-do-at: full size: originals in the 
National Museum. (From Wilde's Cata- 
logue, Gold, p. 63.) 


That the bunne-do-at was used as a personal ornament 
is obvious from the way in which it is mentioned in the 
tales : the size and value of course depending on the rank 
and means of tne wearer. In the list of precious articles 
belonging to the usurping King Eochaid Airgthech (a.d. 
298), and buried with him, are included his two fdils* 
wrist-rings or bracelets, his muintorc, and his two bunne- 
do-ats* : and each individual in the retinue of Bodhbh 
Derg wore a bunne-do-at worth thirty ungas or ounces of 
gold.f From the two last entries we may conclude that 
these articles were sometimes worn singly and sometimes 

Fie. 275. 

Solid gold Bunne-do-at, drawn one-third size. Now in the Museum of Trinity 
College, I Hii.lm : 33 oz. So far as is known, the heaviest of its kind in existence. 
<From Wilde's Catalogue, Gold, p. 60.) 

in pairs. The old writer, when describing this retinue, 
probably exaggerates its magnificence : but yet — in regard 
to the " bunne-do-at worth thirty ungas of gold," equal 
22} oz. Troy (see chap, xxvii., sect. 3) — it should be 
remarked that there is not necessarily any exaggeration, 
inasmuch as the Trinity College bunne-do-at figured here 
is much heavier, weighing 33 Troy ounces. The bunne- 
do-at was used partly as an ornament and partly as a 
sriark of affluence like many valuable articles of the 

* Voyage of Bran, I. 48, u ; 52, 7. 

t Man. & Cust., II. 157, -where bouinde-do-at is translated 'twisted 
ring.' There is no English word to translate this Irish term : " two-disc- 
ring," though exact, is cumbrous. It would probably be better to transfer 
it to English as it stands— bunne-do-at; as Petrie does, with his habitual 
caution. He queries whether the bunne-do-at is not a bracelet : Roun<) 
Towers, 109. 


present day. It was probably worn on the breast at 
one side, suspended from a button like that shown at 
p. 206, supra, fig. 241, or if there was a pair, one was placed 
on each side. The question as to its use as money will be 
discussed in chap, xxvii., sect. 4. 

Circular Gold Plates. — Among the gold ornaments in 
the National Museum are a number of very thin circular 
plates, with raised ornamental patterns punched from the 
back, varying in diameter from 1^- inch up to 4 inches. 
Fig. 276 represents one of these, 3-5- inches in diameter, 
found near Ballina in Mayo. 
All of them have the two holes 
at the centre for fastening on 
the dress. According to Wilde 
they are often found in pairs : 
and they were worn on the 
breast, like the bunne-do-ats, as 
mere ornaments, and as a mark 
of opulence. Petrie, in an article 
in the Dublin Penny Journal 
U. 244), says: — "The figures of 
" the kings sculptured in relievo 
* on the great stone cross at 

" Clonmacnoise, are represented with round plates of this 
"description, placed upon the breast."* 

Brooches. — The brooch was worn by both men and 
women, and was the commonest of all articles of jewellery. 
It was used to fasten the mantle at the throat and was 
fixed crosswise. In the descriptions of the warriors in the 
Tain in the Book of Leinster, nearly all wore brooches of 
gold, silver, fitidruine, or iron. The value of the brooch — 
like that of the bunne-do-at— depended on the rank and 
means of the wearer. The poorer people wore a plain 

Fig. 276. 

Circular gold Plate. One of those in the 
National Museum. (Prom Wilde's Cata- 
logue, Gold, p. 83.) 

* For Bishop Gibson's story of the finding of one of these plates near 
Ballyshannon, through the description given by an Irish bard in a song : see 
Wilde, Catalogue, Gold, 82 : and Dub. Pen. Journ., 1. 244. 


one of iron or bronze, with little or no ornamentation : 
but kings, queens, and other persons of high rank wore 
brooches made of the precious metals set with gems and, 
in Christian times elaborately ornamented with the peculiar 
Irish interlaced work. These must have been immensely 
expensive. That the descriptions given of brooches in old 
Irish writings are not exaggerated we have ample proofs 
in some of those now preserved in our Museums, of which 
the Tara Brooch, figured in vol. I., p. 562, is the most 

What is called the Dalriada brooch was found in 1855 
by a man digging in a field near Coleraine : it is now pre- 
served in the National Museum. It is chiefly interesting 
as being of pure gold, in which probably it is unique. The 
circle is 2\ inches in diameter, and the pin is 5 inches 
long : total weight 2\ oz. The ornamentation is of the 
usual Christian Irish character, but not at all so elaborate 
as that of the Tara brooch. In Dr. Petrie's opinion it is 
not older than the end of the eleventh or the beginning of 
the twelfth century.* It is figured and described in the 
Ulster Journ. of Archaeol., iv., p. I. 

The general run of brooches had the body circular, 
from two to four inches in diameter, with a pin from six 
to nine inches long. But some were much smaller, while 
others again were larger and longer, and reached in fact 
from shoulder to shoulder. These great brooches are 
often noticed in the records. In the Story of Etain in 
the Book of the Dun Cow,- a certain mounted warrior is 
seen with a brooch (ed) of gold in his cloak reaching to 
his shoulders at both sides. In the Bruden Da Derga, 
Keltar of the Battles (whose residence is figured in vol. I., 
p. 85) had his cloak fastened with a cuaille or stake- 
brooch, which reached from one shoulder to the other. 
These descriptions are corroborated by the Brehon Law, 
which mentions a fine for injuries caused by the points of 

* Proc Roy. Ir. Acad., vi 302. 


brooches extending beyond the shoulders* : and still more 
decisively by the existence of very large brooches in the 
Museum, as mentioned in vol. I., p. 22. 

These large brooches were generally heavy. Queen 
Maive's brooch of gold (i.e. ornamented with gold) which 
fastened her mantle, weighed according to O'Curry's cal- 
culations about 4 lb. troy (or more correctly about 3 fb.J.f 
This of course cannot be insisted on as historical fact : 
but it shows that the writers were familiar with large 
and heavy brooches. 

The various names applied to a brooch will be 
exemplified in the following notices. One of the most 
common, and probably the earliest, was dele or delg, 
which primarily signifies a thorn. In the story of the 
Tain, Laegaire the Victorious is described as having a 
delg of gold fastening his cloak at the breast. When 
Queen Macha of the Golden Hair gave orders for the 
building of Emain she took the golden eo [yo] or brooch 
from her neck, and with the long pin marked the outline 
of the palace-rath. £ This term eo was in very general use. 
Another word was cassan, a diminutive of cas, a twist, 
which evidently refers to some peculiarity of make, 
probably in the pin : King Concobar's son Causcraid had 
his cloak fastened with a silver cassan. In another part 
of the Tain, three warriors are described as having 
tanaslaidhe [tonnaslee] of gold in their cloaks : and the 
nine comrades of Cormac Connlingas had their mantles 
fastened with nine tanaslaide.% The brooch with this 
name must have been long and slender : for tana means 
slender and thin. There was a very large circular brooch 
which was called a roth [ruh], i.e. a wheel, either from its 
great size, or because it was made with radii or spokes 
like a chariot-wheel. In the story of the Bruden Da Derga 

* O'Curry, Man. & Cust., II. 163. f Ibid., 102. 

X Corm. Gloss., 63 (under " Emain " Keat. 247. 
§ Stokes, Da Derga, 175. 


we meet with notices of such brooches. Conari Mor, king of 
Ireland in the first century, is described in a prose passage 
as wearing a mantle [of silk] which " is even as the mist 
" of Mayday. Diverse are the hue and semblance each 
" moment shown upon it : lovelier is each hue than the 
" other." This cloak was fastened with a roth or wheel- 
brooch of gold, so large that it covered his whole breast, 
from chin to waist. In the accompanying poetry the same 
brooch is called a delg* 

Another word for a brooch much in use is bretnas. 
Queen Maive once saw a lady wearing a speckled cloak 
in which was a round heavy-headed bretnas] : on one of 
the islands discovered by Maildune, he saw a lady richly 
dressed wearing in her mantle a silver bretnas with chains 
of goldj : and in the Palace of the Little Cat he saw a row 
of bretnases of gold and silver with their pins fixed in the 
wall and their heads outward. The word duille, which 
literally means a leaf, was applied — commonly with the 
termination nd or nn — to a large brooch of a special make. 
Queen Maive's great golden brooch was named duillend- 
delc, ' leaf-brooch ' : and in a passage already quoted 
Cuculainn wears a great dulenn at his breast. The last 
term I will mention is milech : in the story of the Bruden 
Da Derga one of the champions is described as wearing a 
silver milech in his cloak§ ; and Mac Conglinne (p. 9, 27) 
arraying himself roughly for his journey, fastens his cloak 
in front with an iron milech. 

Illustrations of brooches of the usual Irish type have 
been already given (vol. I., pp. 21, 562). Two others of 
a different make are represented here, figs. 277, 278. 

That elaborate and costly brooches continued to be 
made at least as late as the end of the twelfth century is 
proved by the following stanza (quoted by O'Curry) of an 
Irish poem written about 11 90 by an Ulster poet, Gillabride 

* Stokes, Da Derga, 202, 203, 204. For the Roth croi, vol. I., p. 59. 
t Man & Cust., 11. no. J Ibid., 11. 159. § Ibid., 137, 13S. 


mac Conmee : — " The gold brooch (dealg-oir), though it 
"gets the praise when the artist makes it lustrous by his 
"art, it is to the artist the praise is really due, who has 
" beautified the brooch."* 

The brooches mentioned in this article are all Celtic, 
some pre-Christian, others made in Christian times ; and 
they were as common among the Celts of Scotland as 
among those of Ireland. Numerous beautiful specimens 

Fig. 277. 

Fig. 278. 

Specially-shaped Brooches. Figure 277, bronze brooch, natural size, pin turning on 
a hinge: one of the most beautiful bronze articles in the Museum, both as to design and 
workmanship. Ornamentation on the ends produced by punching or hammering from 
behind. Fonnd in a crannoge III Roscommon. Figure 278, bronze spring brooch, 
serpent pattern, natural size, also in National Museum. Both of great antiquity. 
(From Wilde's Catalogue.) 

have been found in Scotland, many of which are figured 
and described in Dr. Joseph Anderson's work " Scotland 
in Early Christian Times " (Second Series). 

The Lann, Blade, or Plate. — It was customary to wear 
a band or ribbon of some kind round the forehead to 
confine the hair. It was generally of some woven fabric ; 
and it will be mentioned farther on that a charioteer wore 

*Man. & Cust., II. 168. 


a bright yellow gipne or fillet in this manner as a distinctive 
mark. Among the higher and richer classes the band was 
often a very thin flexible plate, strip, or ribbon of burnished 
gold, silver, or findruine. This was what was called a 
lann or land, i.e. ' blade,' or more commonly niam-lann, 
' bright-blade ' : and we have seen (p. 234) that the same 
terms were applied to the bright golden crescents for 
the neck. The lann is very often referred to in the old 
literature. In the story of Da Derga (p. 289) three men 
are mentioned as wearing tedra lanna 6ir for airthiur a 
cind, " three plates of gold on the front of their heads " : 
in the next page (290) nine others wear similar ornaments : 
and in the Acallamh (Ir. Texte, iv. 185) a woman is 
mentioned having a lann d'dr buidhe re hetan, a " blade 
of yellow gold on her forehead." In one of the eighth- 
century Irish Glosses on St. Paul's Epistles to the 
Corinthians (I., chap, ix., ver. 24), the Irish commentator 
thus explains the Latin word brabium (the prize in a 
race) : — " i.e. the lann which is sought therein is the 
remuneration of the soldier's service."* Several long 
thin gold plates are to be seen in the National Museum, 
no doubt of the kind and use referred to in the records. 
There is one beautiful thin ribbon less than a quarter of 
an inch wide and five feet long, with an at or button at 
each end for fastening, which was probably wound round 
and round the head, passing in front over the forehead, 
to confine the hair. 

The minn or diadem (treated of below) and the lann 
or plate were always distinguished : for example, in the 
Brehon Law (v. 382, 18; 383, 22, 23), a lady's ornamental 
tiag or hand-bag is mentioned as containing, among other 
articles, a mind 6ir and land Sir, i.e. a " golden diadem," 
and a " golden lamina or thin plate " (either for neck or 
forehead) . 

* Stokes and Strachan, Thesaur., 1. 565. 


The Minn, Diadem, or Crown. — Kings and queens wore 
a diadem or crown, commonly called minn or mind : often 
designated minn oir or mind n-oir, ' diadem of gold ' : 
but often also called ban or cenn-barr [kan-bar]. This 
last, meaning ' head-cover,' was also, as we have seen, the 
usual name of a helmet. That minn means ' diadem ' 
there can be no question, for it is used to explain 
" diadema " in the eighth-century Glosses of Zeuss. On 
account of the designation minn oir, ' golden diadem,' it is 
usually described as made wholly of gold : but this, as we 
shall see, has arisen from a misconception. The minn, 
nowever, was not confined to kings and queens, but was 
worn by men and women belonging to all the higher 
classes, probably indicating rank according to shape and 
make, like the coronets of modern nobility. It was not 
worn in common, but was used on special occasions : a 
lady usually carried her minn-oir in her ornamental work- 
bag, along with other such valuable or ornamental articles, 
ready to be used at any moment.* 

As there has been much misconception regarding the 
Irish minn, it will be well to look into the question somewhat 
closely here. Wilde, in his Catalogue (Gold, pp. 12 et seq), 
assumes that the crescents of the first tw) types already 
described (figs. 262 to 265, pp. 234 to 237) are the objects 
designated by the Irish term minn or mind : and states 
his opinion — an opinion not originated by him however — 
that they were worn as diadems ; in which he is followed 
by Dr. Frazer of Dublin, and by others : but those of the 
third kind (fig. 268, p. 239) he sets down as gorgets. He 
does not however put forward the diadem idea quite 
positively, as is indicated by several expressions of doubt, 
such as : — " While the precise use and mode of wearing 
" the lunulas or moon-shaped plates (fig. 262, p. 234) are 
" questions open to discussion, no doubt can exist as to 
" the object of the articles termed ' gorgets ' " (fig. 268, 

* See O'Curry, Man. & Cust., 11. 113, 


p. 239).* But none of these Irish crescents are formed for 
head-wear, for they do not make any approach to the 
shape of the head, as any one who tries to fit them on will 
find out for himself. And to make matters worse, in order 
to force one of them into position, it would be necessary 
to have " the flat terminal plates applied behind the ears "f 
[and turned inwards] ; which, besides the unnatural wrench 
necessary to bend them back on each occasion, and the 
obvious violent incongruity of the whole adjustment, would 
have the effect — as M. Reinach observes in the passage 
quoted below — of completely hiding the most beautifully 
ornamented portions of the crescents (as in fig. 264). More- 
over, it is incomprehensible why a use should be assigned 
to one type of crescent different from that for the others, 
as they have similar flat outlines — differing only in non- 
essential details : all three are equally suitable for the neck, 
and all three equally unsuitable for the head.| 

But it is needless to follow this matter farther : one of 
these crescents — of whatever type — is no more fitted to 
be worn on the head than a stocking is to be used as a 
glove. Hear what a common-sense and learned foreigner 
— M. Salomon Reinach — who has no preconceived notions, 
says on this point : — " The Irish crescents should be con- 
" sidered as collars or gorgets. Frazer, following [Wilde 
" and] others, has had the idea that they were diadems, 
" and assimilated them to decorations of that kind which 
" ornament the heads of the Roman empresses on coins. 
" But the form of the extremities suffices of itself to con- 

* Wilde, Catal., Gold, p. 30. 

+ These are Wilde's words. Ibid., p. 12. 

% Sir William Wilde is sometimes mistaken in his opinions regarding 
Irish antiquities, as he is in the present case : for Irish antiquarians of 
those days had not the advantages and facilities now available to us. 
But he has done great service to Irish archaeology by the publication of 
his Catalogue of Irish Antiquities, and of his two Essays on Irish Medical 
Science, in connexion with the Census Reports : service which I think 
has hardly received due acknowledgment. How much I owe him is well 
evidenced in this book. 


" demn this explanation."* O'Curry too, who has at great 
length examined this question,! pronounces decisively that 
the minn is not a crescent. His words are (Man. & Cust., 
II. 193) : — " That the mind-oir was not an ordinary land, 
" that is, a frontlet or a crescent of gold, must be at once 
" acknowledged, when we find both mentioned together as 
*' different articles belonging to one and the same person, 
"■ and when, besides this fact, it will be shown that, whilst 
" the land was worn either at the neck or on the forehead. 
*' the mind invariably covered or surrounded the whole of 
4t the head." 

Let us now inquire what the Irish minn really was. 
There are two circumstances that have helped to lead 
antiquarians astray on this question. One is the repre- 
sentation of Roman empresses on coins — as mentioned by 
M. Reinach above — in which they wear a crescent on the 
front of the head : from which some have jumped to the 
conclusion that the same fashion must have prevailed in 
Ireland : and to illustrate this supposed Irish custom there 
is actually in Gough's Camden (in. 476) a picture of a 
lady Wearing a crescent-ornament or diadem evidently 
drawn from a Roman coin, and having nothing to do with 
Ireland. The other misleading circumstance is the term 
minn-oir, ' diadem of gold,' which is constantly used, and 
which many persons took to mean that the minn was 
made wholly of gold — no mixture of any other material : 
and as the crescents are all pure gold, this was considered 
to indicate that a crescent was a minn. But the term 
minn-oir merely means that it was ornamented with gold ; 
a. mode of expression found everywhere in all sorts of 
Irish literature. We read of steeds having bridles and 
reins of gold : Cuculainn's chariot had " a frame of tin ": a 
chariot had " iron wheels " (meaning of course shod with 

* Translated from M. Saloman Reinach's " Les Croissants d'or 
Irlandais," Rev. Celt., xxi. 75. 

t In Lects. xxvm. and xxix. of his Man. & Cust. 


iron rims) : camans or hurling-sticks are described as "of 
silver " : chariot-shafts are " of white bronze " : all which 
mean that the several articles were mounted or ornamented 
with, or partly made of, the metals — the unmixed metals 
being impossible. Such expressions are so numerous that 
references are needless. 

The minn was an article wholly different from a 
crescent of any kind. It was not a plate of gold, but a 
regular crown or cap of elaborate workmanship, made of 
a combination of various materials, and so formed as to 
cover the whole head : all which will be obvious from the 
ollowing quotations and references. The barr or minn of 
Brunn the son of Smetra, mentioned in the " Adventures 
of Nera,"* and designated as the " mionn n-oir which the 
king wears on his head," was a wonder of workmanship, 
one of " the three chief articles of manufacture in Erin." 
This same mionn n-oir is described in another story 
relating to the same adventure, as " a cathbarr of the pure 
" purple of eastern countries, with a ball of gold above it as 
" large as a man's head, and a hundred strings around it of 
" mixed carbuncle, and a hundred combed tufts of red 
" burnished gold, and stitched with a hundred threads of 
" findruine."-\ Here it will be observed that the mionn 
n-oir is described as a diadem made of mixed materials, not 
of gold only. It is not meant of course that this gorgeous 
description should be accepted as literally true : but it 
shows that the writer had something in view very different 
from a crescent of gold. 

In the Tain it is related that on one occasion Cuculainn, 
seeing a number of maidens coming towards him headed 
by one beautiful lady with a mind-n oir on her head, whom 
he took to be Queen Maive, flung a stone from his sling, 
which struck the golden minn, and broke it into three 
-pieces. I How could this be if it was a plate of tough gold ? 

* Rev. Celt., x. 218, line 71. 

| O'Curry, Man. & Cust., 11. 200, 201, 202. J Ibid., 196, top. 


The manner in which Queen Maive's diadem is spoken of, 
too, indicates an article delicately made and easily soiled. 
During the march of the Connaught army north to prose- 
cute the war of the Tain, King Ailill and the queen each 
wore a golden diadem — mind-oir. Queen Maive, as the 
story tells us, had nine splendid chariots for herself and 
her attendant chiefs, her own in the centre, with two 
abreast in front, two behind, and two on each side, right 
and left ; and — in the words of the old tale — " the reason 
" for this order was, lest the clods from the hoofs of the 
" horses, or the foam-flakes from their mouths, or the dust 
" raised by that mighty host should, strike and tarnish the 
** golden diadem [mind-oir] on the head of the queen."* 
All this elaborate precaution could hardly be needed for a 
simple plate of gold. 

But a still better idea of the size, general shape, and 
use of the minn-oir is given in the very pretty legend of 
Queen Mairennf : which shows that it was made to encircle 
and cover the whole head. Dermot, king of Ireland (a.d. 
544-565), had two wives, Mugain, who was barren, and 
Mairenn Mael, who had children. Mairenn was quite bald, 
and always wore a diadem — minn-oir — to hide the blemish 
(see p. 182, supra) : and the barren Mugain was filled with 
jealousy and hate for the fruitful Mairenn. Once upon a 
time at the assembly of Tailltenn, when all were seated in 
state according to rank, the men on one side with the king, 
and the women apart on the other side with the two queens, 
Mugain, burning with jealousy, called to her a bitter-minded 
female satirist or jester [bancdinte, vol. I., p. 454, supra), and 
promised her whatever reward she asked, to pull the diadem 
off Queen Mairenn's head where she sat before all the 
assembly. The satirist went to Queen Mairenn and asked 
for a present (a usual request from a poet) ; but the queen 
said she had nothing to give. " You shall have this for me 

* LL, 59, last four lines. 

f O'Curry, Man. & Cust., II. 193 : Silva Gad., 89. 


then," said the satirist, seizing the diadem and lifting it 
off her head. The poor queen in her agony and shame 
at being thus publicly exposed, cried out : — " God and 
St. Ciaran help me in this need ! " whereupon, quick as 
lightning, says the legend, and before anyone had time to 
notice the blemish, a beautiful crop of bright golden hair 
sprang from her head in ringlets and fell in glorious masses 
down over her shoulders, so that the whole assembly were 
struck with wonder and delight. For a long time after 
that incident Queen Mugain was in disgrace. 

In numerous other old authorities the minn or mind is 
mentioned in terms implying that it covered the whole 
head. In the tenth-century version of the story of King 
Labra the mariner, who, like Midas, had horse's or ass's 
ears, we are told that he constantly wore a " mind n-ordha 
upon his head " to hide the deformity,* which certainly 
could not be done if the minn was a crescent placed on 
the top front of the head. Moreover, a little farther on in 
the tale, the same mind n-6rdha is called a cathbarr or 
helmet : cathbarr imom cenn, " a helmet round my head,"- 
as the king expresses it. 

All that has been said is borne out by the form of 
expression often used when the minn is mentioned. In 
the Acallamh it is related that St. Patrick and his com- 
panions see a band coming towards them with two warriors 
at their head. One wears a cathbarr or helmet, the other 
a mind-6ir : and in both cases the same words are used, 
im a chenn, " round his head," " encircling his head "f : 
im or imm, circum, circa, Z., 654, 20. This shows that the 
minn was not a mere lunula or ornament on the front of 
the head, but a covering for the whole head. 

Fully confirming the preceding literary testimonies, 
comes the most unquestionable evidence of all, the actual 

* Prof. Kuno Meyer in " Otia Merseiana," vol. hi., 1903. P- 46, where 
he edits the story (about King Eochaid) with translation. 
j Stokes, Acallamh, Ir. Texte, IV., pp. 162, 235. 



representation of a native crown worn by an Irish king, 
seated, carved on one of the panels of the south side of 
the Durrow high cross, which was erected about a.d. 
1010.* It is reproduced on the cross in Miss Stokes's 
book, " The High Crosses of Castledermot and Durrow," 
from which the illustration here has been copied. The 
original crown of which this is a representation was about 
five inches high, quite flat on top, with a slender band 
all round, above and below, the 
two bands connected by slender 
little fillets or bars, about two 
inches asunder. It covers the 
whole head like a hat, and 
there are two bosses over the 
ears, three or four inches in 
diameter, f 

The Irish crown varied in 
shape however ; but in no case 
did it resemble a crescent or 
lunula. It is pretty certain that 
some had rays or fillets standing 
up detached all round. Crowns 
of this kind, belonging to the 
O' Conors, kings of Connaught, 
as represented in the thirteenth- 
century fresco-painting in Knockmoy Abbey, are shown 
in vol. 1., p. 59. They are probably native Irish, though it 
is just possible that these particular forms might have 
been adopted under Anglo-Norman influence. That some 
such crown however existed in Ireland at an earlier 
period is shown by the literature. In the " Vision of Mac 

* For this date, see the above-named book of Miss Stokes, p. 11 bot., 
and 12 top. 

f See description of this figure in Miss Stokes's High Crosses of Castle- 
dermot and Durrow, p. 10 : but she is certainly mistaken — led astray I 
suppose by Wilde — in calling this crown a lunula : to which it bears not 
the least resemblance. 

Fig. 279. 

Crowned Irish Kin?, seated, with 

shield, sword, and spear ; a dog on each 

side. (From the High Cross of Durrow.) 




Conglinne " (p. 89, 10) a person is described as wearing a 
crown of seven corns, ' horns,' or fillets ; and in another 
part of the same tale (123, 31 ; 152, 2 s), a crown of twenty- 
seven fillets is mentioned. 

The crescents, then, are not minus. There is in fact no 
such thing as a tninn or diadem in the National Museum ; 
and I suppose there never will be, for the good reason 
that such a complex and delicate object would not hold 
together if buried in the ground. 
The metallic parts would indeed 
remain ; and there is good reason 
to believe that two relics now in the 
National Museum are of this class. 
One of them — a beautiful enamelled 
article — is figured by Miss Stokes 
in Plate xix., fig. 2, of vol. xxx., 
Trans. Royal Irish Academy, and 
described by her at p. 290, same 
vol., where she records the opinion 
that it is a portion of an Irish 
radiated crown. It is figured in 
outline here : but a proper idea of 
its exquisite workmanship and 
beauty of colouring can only be 
obtained by viewing either the 
object itself or Miss Stokes's coloured 
representation mentioned above. Mr. Kemble (quoted by 
Miss Stokes, Early Christian Art, p. 53) says of this and 
the other corresponding object : — " For beauty of design 
" and execution [they] may challenge comparison with 
" any specimen of cast bronze work that it has ever been 
" my fortune to see." 

It is said that an " Irish crown " was found in 1692 at 
the Devil's Bit Mountain in Tipperary. This " crown " 
was first figured by Dermot O'Connor in the Preface 
(p. v.) to his translation of Keating's History of Ireland, 

Fig. 380. 
Enamelled metallic object in the 
National Museum : believed to be 
a ray or fillet of a crown : drawn 
half size. (From Miss Stokes, In 
Trans. Royal Ir. Academy, XXX„ 
Plate XIX., fig. 3.) 


from which it was reproduced in the Dublin Penny Journal 
(i., p. 72, a.d. 1832). But we know that O'Connor wilfully 
perverted Keating ; so that no reliance can be placed 
either on the story told by him or on the picture. The 
illustration here shows O'Connor's delineation, which is 
not like any Irish crown. Indeed, Wilde plainly hints 
his opinion that it is a perverted picture of a drinking-cup 
(Catal., Gold, p. 8, note). 

The word Asionn appears for the first time as meaning 
a diadem in O' Flaherty's Ogygia, from which it was copied 
into various other standard works, including O'Brien's and 
O'Reilly's Irish Dictionaries. But in a communication to 
the Athenaeum of 24th 
August, 1 90 1, 1 have shown ^I^^^n. 

that there is no such word J^^^^^^^^> 

as asionn. It was merely J^ m^M^^S^m ^=,^ 

a printer's error. €^KBi P^^^^^^^^^^P 
O' Flaherty, in his manu- ^^i ^^^^^^^^^^^^ 
script, wrote Mionn : but ^^^ ~"*^ 

as the Irish capital M is FlG - 28 '- 

Conjectural drawing of an object erroneously 
Often Written and printed supposed to be an Irish crown. Stated to hare 

been found at the Devil's Bit Mountain in 1693. 

in such a form as to re- 
semble the Irish As, the printer changed O'Flaherty's 
Mionn to Asionn : and so this spectre of a word has 
haunted Irish literature ever since. Of course O'Flaherty 
did not see a proof ; for such a glaring error could not 
have escaped him. 

Earrings. — Men of the high classes wore gold earrings. 
This custom has been recorded by Nennius, as we have 
seen (p. 228, supra) ; and it is noticed in Cormac's Glossary 
(p. 8) under the word aunasc, an earring, a word which he 
correctly derives from nasc, a clasp or ring, and aue, a form 
of 0, an ear. After this he remarks that the aunasc was 
" a gold ring which is round the fingers or in the ears of 
the sons of nobles." As au-nasc properly means " ear-ring," 
O'Curry thinks the insertion here of " the fingers " a 


corruption of Cormac's text.* But it is not necessary to 
suppose this, as such extensions of meaning are common 
in all languages. The passage as it stands merely 
shows that similar rings were often worn on ears and 
fingers, so that the name of one got extended to the 
other. In the ancient tales these ornaments are often 
mentioned. In the Bruden Da Derga, King Conari's 
juggler wears " ear-rings (u-nasca) of gold round his 
ears."f And a little farther on the king's nine harpers 
wore ear-clasps of gold round their ears : but here the 
word for ear-ring or ear-clasp is au-chuimriuch. 

From the names given to 
earrings, as well as from the 
manner in which they are 
mentioned, it is plain that the 
ears were not pierced ; but a 
thin elastic ring was clasped 
round the ear ; and from the 
lower extremity of this another 
little ring was suspended (like 

Ancient Irish fold Earrine, one of a pair that represented in fig. 282) . 
found in County Roscommon. ■»-> • • 1 1 J 

For cuimnuch, as already 
remarked (p. 228), means literally ' binder ' : au-cuimriuch 
' ear-binder ' : and au-nasc means much the same thing : 
nasc, Lat. nexus, a bind or tie (see page 224, supra). 
Accordingly, as the Da Derga story tells us, each of King 
Conari's nine harpers wore an au-chuimriuch round (im) 
each ear "J : and the chief juggler wore ear-clasps (u-nasca) 
round (im) his ears.§ In Cormac's Glossary the au-nasc 
is defined nasc-aue, ' binder of the ear ' : and it is said to 
be worn " in the ear " : but this latter expression is quite 
consistent with all that is said here. 

The mention of Irish earrings by Nennius, and by 
Cormac under their proper name aunasc, may be classed 

* Man. & Cust., n. 186. % Man. & Cust., 11. 147, note 214. 

f Ibid., 11. 145, note 206. § Ibid., n. 145, note 206. 


among these remarkable confirmations of the accuracy of 
the Irish historical romances, so far as incidental details 
are concerned, mentioned in vol. 1., p. 9, supra, and in 
other parts of this book. 

Golden Balls for the Hair. — Both men and women some- 
times plaited the long hair ; and at the end of the plait 
they fastened a thin, light, hollow ball of gold, which was 
furnished for the purpose with little apertures at opposite 
sides. Sometimes these balls were worn singly — probably 
behind — and sometimes in pairs, one on each side. In the 
Book of the Dun Cow the fairy king Labraid (who was 
sitting in state) is described as having yellow hair, with 
" an apple (ubull) of gold enclosing it."* And in another 
part of the same book we are told that Cuculainn had 
" spheres (cuache) of gold at his two ears into which his 
hair was gathered, "f The lady Bee Fola, going through 
a wood, saw a young warrior magnificently attired, and 
among other ornaments he had " two balls of gold [da 
" ubuill oir] on [the ends of] the two divisions of his hair, 
" each the size of a man's fist. "J Ladies followed the same 
fashion : but they had several very small spheres, instead 
of one or two large ones. Eochaid Feidleach once saw a 
lady richly attired : on her head were two golden yellow 
tresses, each plaited into four locks, on the end of each of 
which was a mell, i.e. a little ball or bead : so that this 
lady wore eight little balls altogether.§ 

As corroborating all these accounts, there are in the 
National Museum a number of these golden balls, found 
from time to time in various parts of Ireland. They are 
all hollow and light, being formed of extremely thin gold : 
and each has two small circular holes at opposite sides by 
means of which the hair was fastened so as to hold the 
ball suspended. Each is formed of two hemispheres, 

* Sick Bed, Atlantis, 11. 103 : also Man. & Cust., 11. 192. 

f Kilk. Arch. Journ., 1870-1, p. 377, last line : and Fled Brier., 65. 

% Bee Fola, 177. § Da Derga in Rev. Celt., xxir. p. 14. 


which are joined with the greatest accuracy by being 
made to overlap about the sixteenth of an inch, and very 
delicately soldered — so that it requires the use of a lens to 
detect the joining. The largest of these balls is & inches 
in diameter, and weighs 2^oz. (figured separately in vol. I., 
p. 21) : so that we see the old story-teller was right enough 
in describing the two balls worn by the young hero as "the 
size of a man's fist." Those in the Museum vary in size 
from 3^ inches down to about two. 

Fig. 283. 

Hollow gold Balls, as described in text : erroneously represented as a necklace in Wilde's Catalogue, 

Gold, 35, from which this illustration has been taken. 

Wilde* conjectures that these balls were worn as neck- 
laces, which they certainly were not : for no such form of 
necklace is referred to in any of the records ; and in order 
to support the conjecture he gives an engraving of eleven 
of them (found in various localities) arranged, according to 
size, and connected by a string, as a necklace, as seen here. 

The corroboration of the truthfulness of the old records 
by existing remains has been frequently noticed through- 
out this book ; and this is a very striking example, inasmuch 
as the custom of wearing gold balls on the hair seems 
so strange that it might not unnaturally be set down as 
the invention of story-tellers, if their statements were not 
supported (see " Corroboration " in Index). 

* Catalogue, Gold, p. 34. 


4. Short rough classified List of the Gold Objects in the 
National Museum, Dublin. 

More than 30 crescents of the first type (figs. 262, 263) ; five of the 
second (figs. 264, 265) ; five of the third (fig. 268). 
Seven hollow balls for the hair (fig. 283). 

Great numbers of bracelets and rings of various shapes and sizes 
(figs. 252, 253, 254). 

A number of long thin bright plates and ribbons. 

About 150 open rings called Bunne-do-at (figs. 269 to 275). 

About 50 very small open rings without the ats or buttons (mentioned 
at p. 385, below.) 

About a dozen thin circular plates with patterns, all with two holes 
for fastening (fig. 276). 

About two dozen torques of different sizes (fig. 260). 
A number of small ornamental beads for necklaces, of various 
shapes (figs. 258, 259). 

Four amulets (vol. 1., p. 385). 

An open spiral, i\ inches long and 1 inch in diameter, with nine 
spires, formed of one square wire. 

Besides these there are a number of small objects not classified. 

(The total weight of all these articles is about 590 oz., which is 
twelve or thirteen times the weight of the collection of gold antiquities, 
from all England and Scotland, in the British Museum. See vol. 1., 
P- 556.) 

Models. — In 1854 an immense collection of gold articles were found 
in a stone cist under a small clay mound near Quin in the County 
Clare, most of them slender delicate rings of the kind called bunne- 
do-at. In one glass-case of the National Museum there are gilt-brass 
models of a portion of this find, consisting mainly of about 100 bunne- 
do-ats, and five crescents of the third type.* 

* In 1896 a number of important gold antiquities were found at 
Broighter in Deny. They were purchased by Mr. Robert Day, f.s.a., 
m.r.i.a., of Cork, who sold them to the British Museum. They are at 
this present time the subject of litigation. The question is whether they 
should not be restored to Ireland, their natural home. See an account 
of them by Mr. Arthur Evans, f.s.a., in Archasologia, lv., 391, 408 ; and 
another by Mr. Robert Cochrane, f.s.a., m.r.i.a., in Journ. Roy. Soc. 
Antiqq , Ireland, 1902, p. 211. (For more on this, see Appendix.) 

Cniamcnt : composed from the Book of Kells. 



Section i. Fences. . 

tVER since that remote time when legend and 
history begin to give us glimpses of the 
occupations of the inhabitants of this country, 
we find them engaged in agriculture and 
pasturage. For both of these purposes open 
land was necessary ; and accordingly, the 
clearing of plains from wood is recorded in the reigns of 
many of the early kings as a public service worthy of 
special notice. But there was always more pasturage 
than tillage. 

Farm Fences. — In very remote times, when the popula- 
tion was small and the land was mostly common property 
(as pointed out in vol. I., p. 184, supra), there was little 
need for fences, and the country was mostly open, so 
far as it was free from forest and bog. But in cours? of 
time, as tillage gradually increased, and private property 
in land became more general, it was more and more 
necessary to fence off the portions belonging to different 
individuals. Fences are referred to in our oldest literature : 
and how important they were considered appears from the 
number of regulations regarding them in the Brehon Law. 
The general terms for a fence are ime, fdl, felmae, and aile. 
When two or more persons came into possession of 
adjacent farms, it became their duty to fence off their 
portions, if not fenced off already. As each fence between 



two farms would be common property, arrangements for 
joint action were laid down in the law, so that each man 
should execute his own part of the work. In making the 
fences they had to be up to time. Three days were 
allowed for marking out the land : in five the fencing was 
to be commenced : in ten days the fence should be com- 
pleted, except the blackthorn crest at top, which was to 
be finished in a month. 

Four kinds of farm-fences are specified in the Law : — 
First, a trench (Irish dais, pron. clash) with the earth piled 
up on one side as a high embankment (called mur, or fert, 
or clad, pron. cly), a kind of fence still used all through Ire- 
land : Second, a six-foot stone wall of dry masonry, which 
is still very general in stony districts in the west and 
south : the Third was formed of logs laid horizontally and 
securely fastened : the Fourth consisted of pointed stakes 
standing six feet above the ground, and six or eight inches 
asunder, bound securely by three bands of interwoven 
osiers, and having a blackthorn crest on top. The top of 
each sharp stake should be blunted by three blows of a 
mallet. No man was directly compelled by law to make 
his fences of any particular height or pattern, or to have 
them put up by a certain time. But there was indirect 
compulsion ; for supposing a lawsuit to arise on a question 
of trespass or such like, the person owning — or part 
owning — the fences should be able to show that they 
were constructed as specified in the law, both as to make 
and time, otherwise the suit was pretty sure to go against 
him. If a fence was carelessly constructed so that some 
stake; were too sharp-pointed at top, or that sharp spikes 
projected from the sides, the owner was liable for damages 
in case cattle got injured.* 

* All the preceding regulations will be found in Br. Laws, iv. 71 to 77, 
113, 115, and Introd., cxxi. cxxxiii : and vol. in. 291. Some are also 
noticed in the general literature : for instance, the thorn-crest is mentioned 
in Mac Conglinne, 86 ia , 


Territorial Boundaries. — Fences such as these were too 
slight and temporary to serve as boundary marks between 
large districts. Various landmarks of a more enduring 
kind were assigned for them, some natural, some put down 
artificially. Suppose a dispute arose as to the exact 
limits of two adjacent territories, whose boundary had 
been marked out in times past, the Law (iv. 143) 
enumerate and recognises twelve different marks, by 
one or more of which the boundary might be recovered 
and defined. Among these are : — a " stone-mark," i.e. a 
large pillar-stone ; an " ancient tree " of any kind, or the 
stump and roots of an old oak, after the tree had fallen 
and disappeared ; a " deer-mark," namely, the hair-marks 
left by deer or cattle on the trees of a wood, or the hair- 
marked footpath made by them along a plain ; a " stock 
mark," i.e. stakes in the earth, or the ruin of a mill, or an 
old bridge under water ; a " water-mark," i.e. a river, lake, 
or well ; an " eye-mark," i.e. a straight line fixed by the 
eye between any two of the preceding which had been 
ascertained for a certainty, but which lay some distance 
asunder ; a " defect-mark," i.e. a place or line along which 
there was no cultivable land, such as a declivity, a sedge, a 
stony vale, or the track of a disused road ; a " way-mark," 
i.e. a king's road, or a carriage-road, or a cow-road (see 
PP- 393 to 395» below) ; a " mound-mark," i.e. a [great] 
mound or ditch or foss " or any mound whatever," such 
as that round the trunk of a tree. 

Pillar Stones and Ramparts. — That Pillar-stones were 
regarded as an important means of marking boundaries 
is shown by their frequent mention in the records. We 
are told in one law-tract that when certain tribe chiefs had 
taken possession of a district, they " erected boundaries or 
placed pillar-stones there " ; and in another place that after 
land has been enclosed a hole is made in the ground on 
the boundary, into which is put the "chief's standing-stone,* 

* Br. Laws, iv 7, , 7; 9, 9; 19, t, ,». 


in order that his share there may be known." It is stated 
in Cormac's Glossary (p. 84, " Gall ") that adjacent settlers 
are not considered as neighbours till their properties are 
provided with pillar-stone boundaries. The custom was 
so general that a legendary origin is assigned to it in the 
Coir Antnann, which says that a certain chief named Failbe 
[Falvy] " was the first person by whom of old in Erin a 
pillar-stone was erected to be set as a boundary " : whence 
he was called Failbe Fdl-Choirthech, i.e. 'of the pillar-stone 
boundaries.'* We have seen that a stone set up to mark 

IV.. 2S4. 

Pillar-stone, about 10 feet high, now called Cloch-fada-na-gcarn, the 'long stone of the cams,' in 
the centre of a rath beside Carnfree, the inauguration place of the O'Conor kings, near Tulsk, in 
Roscommon. (From Kilk. Arch.-eol. Journ. for 1870, p. 250.) 

a boundary was sometimes called a "stone of worship " : 
corresponding with the pillar- stone god Terminus wor- 
shipped by the Romans (see vol. I., p. 277). 

Boundary pillar-stones are found standing all over the 
country. But pillar-stones were erected for other purposes, 
of which the most usual was as a monument over a grave 
(for which, see chap, xxxi., sect. 5, infra), a practice that 
prevailed in Christian as well as in pagan times. Battles 
were often commemorated by pillar-stones as well as by 
earns and mounds. Sometimes pillar-stones were set up 

*Ir. Texte, in. 293. For other notices of pillar-stones as boundaries 
s-ee Kilk. Archaeol. Journ., 1867, pp. 5, 6 : and 1899, 221, 12. 




on raths, of which a fine example may be seen beside 
Kilkee in Clare : another is illustrated in fig. 284. It has 
been already mentioned that pillar-stones were sometimes 
erected as idols. Many of the standing-stones still 
remaining have a hole through them from which they 

are commonly called 
" holed-stones," but the 
use of these is a mystery. 
Pillar-stones are called 
by several Irish names: — 
coir the [curha] ; coirtlie- 
cloch (which is a duplica- 
tion, for clock is a 'stone') ; 
gall; gallan; and legann. 
As to the term gall, of 
which gallan is a diminu- 
tive, Cormac's Glossary 
tells us that pillar-stones 
were called "gall" 
because they were first 
erected in Ireland by the 
Gauls ; and as a matter 
of fact we have in Irish 
legendary literature 
accounts of a colony of Gauls coming in very early times 
to Ireland. As to many or most of the pillar-stones now 
remaining in the country, it is often hard or impossible to 
tell, in individual cases, for which of the above-mentioned 
purposes they were erected* 

•It is scarcely necessary to remind the reader that the custom of erecting 
unhewn pillar-stones for various purposes prevailed among most ancient 
nations: and such stones are found still standing all through Europe and 
Asia. Holed-stones are also very general, but, as in Ireland, their original 
purpose is unknown. " The standing-stones or menhirs " [of the world 
generally] — says Sir John Lubbock (p. 107) — " were no doubt generally 
erected in memory of some particular event, the majority being in fact tomb- 
stones of prehistoric times." See also Wood-Martin, Pagan Ireland, 
p. 307 : and Joyce's Irish Names of Places, vol. 1., pp. 95, 342. Cromlechs, 

Fig. 285. 

' Holed-stone," near Doagli, County Antrim. (From 
Kilk. Archieol. Journ. for 1887-8, p. 78.) 


Many of the great mounds or ramparts also still exist : 
and there is generally a popular legend that they were 
rooted by an enormous enchanted black pig. One of the 
largest of all is that in the valley of the Newry river, which 
separated the sub-kingdoms of Oriell and Ulidia, and 
which will be found fully described in the Ulster Journal 
of Archaeology — new series — for 1897. Great artificial 
dividing dykes are found in every part of the world, some 
historic like the Roman wall in Britain, and some pre- 
historic. Offa's Dyke dividing England from Wales is a 
grand example : but the most stupendous artificial dyke 
in the world is the great wall of China. 

2. Land, Crops, and Tillage. 

Classification of Land. — The Brehon Law (iv. 277) 
specifies three main divisions of " superior arable land," 
recording in this respect an Irish custom in general use in 
former times : — First : " arable land which takes precedence 
of all lands," i.e. land of the very best kind, which pro- 
duces " everything good " — corn, and milk, and flax, and 
glaisin, and roid (see, for these dyeing plants, pp. 358, 359, 
infra), and sweet herbs, and requiring no manure. There 
are no " sticking plants," i.e. no briars nor burdocks (which 
stick to one's clothes), so that if a horse should graze on it 
none of these sticking plants will be found on its mane or 
tail. Rich land of this kind is often called " land of three 
roots " (tir [or talamh] tri meccon) : " the richest soil, which, 
" according to the Irish notion of the present day, is always 
" known by the presence of three weeds, remarkable for 
" their large roots, namely the thistle, the ragwort, and the 
" wild carrot."* Second : hilly arable land which is well 
watered ; groves and copses of ash-trees grow here and 

dykes, pillar-stones, &c, are treated of in Lubbock's Prehist. Times, 
chap. v. For full information on holed-stones, with many illustrations, 
see Wakeman's Handbook of Ir. Antiqq., 3rd ed., by Mr. Cooke, p. 14. 
* Petrie, Round Towers, 218, note f. 


there ; and the parts of it that are tilled produce good 
crops. Third : labour-requiring arable land ; what is called 
axe-land, which requires much labour, but which, when 
well worked, produces good crops freely. 

Besides these three divisions of good land, the text 
gives three divisions of " weak land," or arable land of an 
inferior kind, viz. : — Land in which fern grows : upland 
with much heath and furze : and lastly, black land with 
bog on the surface, not absolutely beyond tillage. After 
this enumeration the values are set forth in milch cows. 

Manure (Irish ottrach) is very often mentioned in the 
Laws, showing the importance attached to it. A dung- 
heap is called in Cormac's Glossary crum-duma, which 
O'Donovan translates ' maggot-mound,' from cruim, a 
maggot, and duma, a mound. The manure mentioned 
in the Brehon Law was chiefly stable-manure : and the 
law-tract (iv. 277) mentions also the application of shells 
(sltg, a shell : pi. slige) to land to improve it.* This last 
law-tract (p. 279) , following old custom, enumerates eleven 
different things that add to the value of land, and estimates 
in seds or cows the amount added by each, or at least the 
amount to be taken as a basis of calculation. Of these the 
most important are : — a wood properly fenced in : a mine 
of copper or iron : the site of an old mill [with millrace 
and other accessories, rendering easy the erection of a new 
mill] : a road [opening up communication] : situation by 
the sea, by a river, or by a cooling pond for cattle. 

Digging for Water. — Various passages both in the 
Brehon Laws and in general Irish literature show that 
the ancient Irish understood the art of obtaining water 
by digging deeply into the ground. It must have been a 
pretty common practice moreover, for the annalists assign 
a legendary origin for it, a thing they never did except 

* The use of shells as a land-improver, is well-known : it will be 
found noticed in the Ulster Journ. Archaeol., iv. 271 : and Boate, Nat. 
Hist., p. 161, mentions it as common in his time. 


where the custom was general. The Four Masters say, 
under a.m. 3991 : " It was by this king (Fiacha Finailches) 
" that the earth was first dug in Ireland in order that water 
" might be in wells." The Coir Anmann (p. 395) assigns 
the discovery to a different person : — " Findoll Caisirni, 
" which epithet means cisternae or ' earth-rending ' : for he 
" was the first person by whom of old the earth was dug 
" to make a pit in which water was found at every time." 
The Greeks similarly assigned the origin of their custom 
of digging for water to their old hero Danaus, king of the 

The Coir Anmann (p. 381) states that the Fir Domnann 
(a tribe of the Fir Bolgs), during their slavery in Greece, 
were condemned by the Greeks to dig deeply into the 
earth to obtain water. There were in Ireland experts who 
pretended to discover by a sort of divination the proper 
places to dig for water. In the story of the siege of 
Knocklong we read that when the Munster army were 
perishing with thirst, their king called in the aid of the 
famous druid Mogh Ruith (for whom see vol. I., pr. 231, 
supra) , who hurled his spear high into the air, and directed 
his disciple Canvore to dig at the spot where it fell. He 
did so : and the water burst forth in a copious stream, 
which relieved the army. That same fine well exists to 
this day, and is universally known by the name of Tober 
Canvore, Canvore's well.* This practice is alluded to in 
a more unquestionable authority, the Brehon Laws 
(iv. 209, 9 ), where the gloss on the law of the rights of 
water has an expression implying that a stream of water 
was sometimes obtained by digging for it. 

Crops. — Most of the native crops now in use were then 
known and cultivated : chief among them being corn of 
various kinds, f Corn in general was denoted by the words 

* See Irish Names of Places, 1. 103. I found the name familiarly 
used : but the people — at least those I spoke to — knew nothing of the 

+ Kitchen-garden vegetables have been already noticed (p. 148, supra). 


arbar [arrar or arroor], and ith [ih] ; besides which there 
was a special name for each kind. In the " Vision of Mac 
Conglinne " (p. 98) eight different kinds of grain are 
enumerated and named ; but some of these were mere 
varieties. These eight occur also in a 14th century Welsh 
poem. We know for a certainty that wheat has been 
cultivated in this country from the most remote ages : for 
we find it constantly mentioned in our ancient literature : 
of which an interesting illustration will be found in the 
record of the death of the two princes in Mailoran's mill, 
p. 333, below. The most common native Irish word for 
wheat is cruithnechi [crunneght], which in Cormac's 
Glossary is derived from cruith [cruh], blood-coloured 
or red, and necht, clean : the first part of this derivation 
is probably correct, but necht is a mere termination. The 
etymology, however, sufficiently proves the interesting fact, 
that the wheat cultivated in the time of the venerable King- 
bishop Cormac — 1000 years ago — was the very same as 
the Irish wheat of the present day ; for every farmer 
knows that the old Irish wheat — now fast dying out — 
is distinguished by its red colour. 

It is worthy of remark that in several other languages, 
wheat — as Pictet shows (Les Origenis, I. 261) — has been 
named from its colour, not indeed from its redness as in 
Ireland, but from its whiteness as compared with other 
kinds of corn. As one instance, may be mentioned the 
English word wheat, which Pictet shows is only another 
form of white. Three other native Irish words for wheat 
are dagh, mann, and tuirenn. 

The observations made about the early cultivation of 
wheat apply equally to oats (Irish coirce, pron. curk-ya) ; 
numerous references to its cultivation and use are found 
in our most ancient literature. In modern times, before 
the potato became very general, oats formed one of the 
principal articles of food of the people, as it did of old 
(p. 141, supra). Barley (Irish edrna [orna]) and rye (Irish 


scgal [pron. shaggal : Lat. secale]) were cultivated, and 
formed an important part of the food supplies. 

Corn was cut with a sickle or reaping-hook, anciently 
called serr or searr [sharr], which in Cormac's Glossary 
(p. 149) is derived from Latin serra. Mac Firbis explains 
serr by carrdn, which is the present Irish word for a reaping- 
hook. Many specimens of reaping-hooks have been found 
in Ireland, some of bronze and some of iron, which may be 
seen in the National Museum in Dublin. They are all 
small, and cutting with them must have been slow work. 
Those of bronze are very ancient — probably beyond the 

Fig. 286. Fig. 287. 

Ancient Irish bronze reaping-hooks. Fig. 286 is of beautiful workmanship, 6!»' inches long. It 
was fitted with a handle, which was fastened in the socket with a rivet. Fig. 287 is of much the 
same construction ; a little imperfect at the top : 7 inches long. Both now in the National Museum, 
Dublin. (From Wilde's Catalogue, p. 527.) 

reach of history. The iron ones are hardly so old ; but 
still they have the look of great antiquity. Meadow-grass 
was cut with a scythe anciently and still called speal [spal] : 
but an ancient manuscript explains serr (sickle) by speal, 
a scythe* ; which may perhaps be taken to indicate that 
anciently there was little or no difference between a scythe 
and a sickle. 

Corn was reaped as it is at present by cutting the 
stalks off at the bottom. But the fact that a cliab buana, 
' reaping-basket,' is mentioned in some of the tales,f would 
lead us to think that in very remote ages — in the bronze 
period — the reaping was sometimes done — as it was, and 

*Corm. Gloss., I49. t Da Derga, pp. 198-9. 


is, elsewhere — by cutting off only the tops with the grain, 
which were brought away in the basket : leaving the straw 
to be dealt with separately. If this supposition is correct, 
it explains the smallness of the reaping-hooks represented 

The corn, while in sheaves, was stacked in a haggard, 
which was called ithlann, corn-floor or corn-yard. The 
word always applied to a corn-rick is cruach, which in 
Cormac's Glossary (p. 44) is derived from a verb meaning 
' to sew,' because "it is sewed all round." From this we 
learn that the people stacked their corn carefully after 
reaping, and covered the rick with thatch which they 
fastened by twining or interweaving with ropes of some 
kind, probably hay-ropes, or what we now call sugans. 
People do the same still in many parts of Ireland. 

Corn was threshed with a flail (suist), often on the floor 
of the kiln-house,* but more commonly on a regularly 
prepared threshing-floor near the kiln. In one corner was 
a little pit or hollow in the floor into which the grain was 
swept as it was threshed out, and which Adamnan (131) 
mentions by the Latin name fossula, ' little pit.' A pair of 
threshers sometimes stood face to face, sometimes side by 
side : as we see at the present day. The Book of Aicill 
has a series of rules for estimating the compensation for 
injuries to bystanders by an accidental stroke or by 
the head of the flail flying off,f as I have often seen the 
buailtedn or ' striking-stick ' fly off with the latter-day 
threshers when the gad or tying-withe broke. To get 
rid of chaff (cdith : pron. caw), the women winnowed 
the corn by hand, using a winnowing-sheet called caetig 
or cditeach.% 

* Br. Laws, m. 221. Adamnan (p. 131) calls the threshing-floor by 
the usual Latin name excussorium. 

f Br. Laws, in. 221 : se? also v. 159. 

% See Todd, Book of Hymns, 17, note 54 : cditeach, in O'Reilly : de- 
rived from the same root as cdith 


Farm Implements. — Most of the common implements 
employed in farm-work at the present day were used by 
the ancient Irish, though no doubt they were somewhat 
different in make. The sickle, scythe, and flail have been 
already noticed. The use of the plough was universal. 
The old word for it was arathar [arraher], which in 
Cormac's Glossary (p. 7) is derived from I .at. aratro, to 
plough, as he derives the Irish dr, tillage, from Lat. aro : 
but all these Irish and Latin words are cognate, and are 
not derived one from another, but from an older original. 
The word arathar is now quite obsolete ; and the present 
name for a plough is cechta [kaighta], which is also an 
ancient word. Several of the parts of the plough are 

Fig. 288. 

A two-horse or two-ox yoke, of timber, 3 feet 9 inches long. Found in a bog in County Monaghan. 
(From Wilde's Catalogue, p. 243) 

mentioned in the old records. The coulter or ploughshare 
was called socc, which is the word still used. That it was 
made of iron we know from many passages — so many that 
it is unnecessary to refer to instances. 

The plough was generally drawn by oxen : but some- 
times by horses : — " St. Ciaran had fifty tame horses for 
tilling and ploughing the ground."* Nevertheless when 
we find ploughing mentioned in old Irish writings, it is 
nearly always oxen that are in question. The ploughman 
(airemk, pron. arrev) had to see that the several oxen were 
taken day after day in turn ; and if, under certain circum- 
stances, anyone worked an ox out of his turn without 
the knowledge of the owner, he was liable to be fined.f 

*Feilire, 61, bottom. fBr. Laws, III. 271, I7 . 


Cormac's Glossary (p. 43) mentions the cuing or yoke, and 
says it was so called " from the hold {congbaiJ) it takes of 
the oxen " ; and he notices also the essem, a rope or strap 
" uniting one yoke to the other, or to the ox, or to the 
oxen " (p. 64). The ploughman held each ox by a halter 
(Irish cennos, pron. kennos), and he also carried a sharp 
goad (Irish brot), " so that " — as the law tract expresses 
it — " the ox may be mastered."* I find no mention in old 
documents of " ploughing by the tail," which, in com- 
paratively recent times, was so prevalent when horses were 
employed, f 

For breaking clods of clay in a ploughed field farmers 
used a clod-mallet called forcca or farcha, which means a 
mallet of any kind : it had a wooden handle, the head no 
doubt being also made of wood. J They used a spade 
(rama) and a shovel (sluasat), both fixed on wooden 
handles — as noticed in the Brehon Laws (in. 205) — and 
both probably made of iron. Elsewhere in the laws 
(iv. 335) a shovel is called by another name samtach. 
In Cormac's Glossary (p. 78) the word for a spade is fee, 
which is still in use even among the English-speaking 
people of many parts of Ireland, who call a spade jack or 
feck. Rama and sluasad are also retained as living words 
for spade and shovel : but the former gets the diminutive 
form rdmhan, often shortened to ran, both pronounced 
rawn. A rake was used, which, as far as we can judge 
from the description of it given in Cormac's Glossary 
(p. 147), must have been much the same as that used at 
present. There it is stated that " it touches the ground " 
[when in use], and that " its handle is through a hole." 
It was, of course, made of wood. It is called in the 
Glossary rastal, which is the word still in use. 

* Br. Laws, 111. 269 ; iv. 304, 305, 306, 307. 
t See Irish Names of Places, 1. 237. 
X Br. Laws, iv., 335. 


3. Some Farm-Animals. 
Cows. — From the most remote ages, cows formed one 
of the principal articles of wealth of the inhabitants of 
this country ; they were in fact the standard of value, as 
money is at the present day ; and prices, wages, and 
marriage portions were estimated in cows by our ancestors 
(see chap, xxvii., p. 385, infra). As might be expected, 
therefore, they are constantly mentioned in ancient Irish 
literature of all kinds ; and they are made the subject of 
special and important consideration in the Brehon Laws. 
The most general Irish word for a cow is bo, not only at 
the present day, but in the oldest manuscripts : in one of 
the eighth-century mss. of Zeuss it glosses bos, with which 
it is also cognate. In Cormac's Glossary a passage is 
quoted from the Senchus Mor to illustrate the word ferb 
as another name for a cow.* The term buar was applied 
to kine in general, derived from bo. A bull is called in 
Irish tarbh [tarruv], a word which exists in cognate forms 
in many languages : in the three Celtic dialects — Old Irish, 
Welsh, and Cornish — it is found in the respective forms of 
tarb, taru, and tarow while the old Gaulish is tarvos ; and 
all these are little different from the Greek tauros, and 
Latin taurus. Damh [dauv], an ox, is evidently cognate 
with Latin dama, a deer. How it came to pass that the 
same word signifies in Irish an ox, and in Latin a deer, it 
is not easy to explain, f The chief use of the ox was as a 
draft and plough animal, for which see " Oxen " in Index. 

* Corm. Gloss., p. 71 ; the original passage of the Senchus M6r, quoted 
in the ninth or tenth century in the Glossary, may be seen in the present 
edition of the Senchus M6r (Br. Laws, 1. 64, 65), where " teora ferba fira," 
' three white cows,' are mentioned : one of the evidences of the antiquity 
of the published edition. 

f The transfer of a name from one species of animals or plants to 
another is a curious phenomenon, and not unfrequently met with. The 
Greek phegos signifies an oak, while the corresponding Latin, Gothic, and 
English terms — fagus, boka, and beech — are applied to the beech-tree ; and 
I might cite several other instances. See this question discussed in Max 
Muller's Lectures on the Science of Language, Second Series. 


The words dairt and colpa, meaning a heifer, will be 
found mentioned in connexion with grazing (pp. 282-3 
farther on) and with standards of value (chap, xxvii., 
p. 386) : and mart, an ox, at p. 127, supra. The usual Irish 
word for a calf is gamuin, modern Irish gamhan [gowan]. 
Cormac's Glossary (61) gives the old word dedel as also 
meaning a calf : and loig or laogh [lay] was still another 
name. The word gamhan must not be confounded with 
its derivative, gamhnach [gownagh], which, according to 
Cormac's Glossary, means ' a milking cow with a calf a 
year old ' ; but which in modern Irish is used to signify 
simply a stripper, i.e. a milk-giving cow in the second year 
after calving. White cows with red ears carried a fictitious 
and fancy value ; and we often find them mentioned both 
in the Tales and in the Lives of the Saints. They are 
also noticed in Cormac's Glossary (p. 72, under " Fir "). 

Pigs. — In point of value to the community, pigs came 
next to cows, and were of more importance to the general 
run of people than horses. They were kept by almost all, 
so that they were quite as plentiful and formed as valuable 
an industry in those days as at present. It has been 
already stated that pork was valued as food by all persons, 
from the highest to the lowest ; and the supply was fully 
equal to the demand. The usual Irish word for a pig was 
and is still, muc or mucc : a boar was called tore. A very 
young pig was a banb or banbh [bonniv], a word which is 
still known in the anglicised forms of bonniv or bonny, or 
with the diminutive, bonneen or bonniveen — words used in 
every part of Ireland for sucking-pigs. But an older 
word for a little sucking-pig was cumlachtach, as given 
in Cormac's Glossary (p. 39). There were many other 
names for pigs : O'Davoren enumerates eight : but they 
need not be given here.* 

It was cheap and easy enough to feed pigs in those 
days. Forests abounded everywhere, and the animals 

* See Stokes, Three Irish Glossaries, Preface, 1. and lxv. 


were simply turned out into the woods and fed on mast and 
whatever else they could pick up. Wealthy people — chiefs 
and even kings, as well as rich farmers — kept great herds, 
which cost little or nothing beyond the pay of a swine- 
herd (muccaid, pron. muckee) : and they gave no trouble, 
for, except in winter, they remained out day and night, 
needing no sties or pens of any kind, being sufficiently 
sheltered by the trees and underwood. But in winter 
they were kept in sties, called in Irish muc-fhoil [muckole], 
as already stated (p. 41, supra). The special time for 
fattening was autumn, when mast abounded ; a practice 
mentioned by Adamnan (p. 135), whose words imply that 
the fat pigs were slaughtered at the end of autumn : so 
that few had to be kept in sties during winter. The 
Brehon Law mentions pigs feeding on mast (Irish mes 
or meas : pron. mas), in wood-covered land : and an 
expression in another part of the Law implies that wood- 
land for hog-feeding was sometimes private property, and 
was of value.* But woodland was generally a part of the 
" commons " (1. 187, supra), where every member of the 
sept was free to send his pigs to feed.f 

When woodland was not convenient, or when for any 
other reason pigs had to be kept and fattened at home, 
they were fed on corn or sour milk, and on offal of 
various kinds : these were managed chiefly by women. 
A pig kept at home was called a " sty-pig " (Irish muc- 
crai : era, a sty). J In Cormac's Glossary (p. 27, " Bacur ") 
mention is made of " braiches round which pigs go " : and 
in another part of the same Glossary (105) is noticed a 
" lotar or trough in which are brachles." Braiches and 
brachles both signify the grains or refuse from brae or 
braich, malt (p. 117, supra) : and from these expressions 
we learn that the custom of feeding pigs on malt-grains, 
now so familiar near breweries> was also practised by the 

* Br. Laws, u. 367, bottom ; in. 39, bottom ; and iv. 103, s . 

I Br. Laws, iv. 257, last par. % Ibid., 11. 367, bottom ; 369, top. 


ancient Irish : for we have seen that brewing was then 
very common. 

The old Irish race of pigs were long-snouted, thin- 
spare, muscular, and active : and except when fat they 
could scour the country like hounds. There are many 
indications in old writings that they were often wicked 
and dangerous, ready to charge and attack when pro- 
voked ; and sometimes they inflicted fatal wounds. In 
the Book of Ailill are a number of regulations providing 
for damages for injuries inflicted by pigs, taking into 
careful account whether there was provocation or not. 
For instance it is stated that when an idler provokes a pig, 
in consequence of which it " charges out on him," and 
wounds him, the owner is exempt.* In the remote forests 
there were plenty of wild pigs : and we have many 
references to them in our literature. In the twelfth 
century Giraldus gives us this testimony : — " In no part 
" of the world are such vast herds of boars and wild pigs 
" to be found." (Top. Hib., I. xix.) 

Sheep were kept everywhere, as they were of the utmost 
importance, partly as food, and partly for their wool : and 
they are constantly mentioned in the Brehon Laws as 
well as in general Irish literature. There was in Ireland 
a many-horned variety, which however has been long 
extinct. f The common Irish word for a sheep was, and 
is, cdera, gen. cderech. 

Trespass by Animals. — When two or more tenants held 
farms next each other, the Law lays down minute regula- 
tions for the trespass of all kinds of domestic animals, and 
enumerates many circumstances to be taken into account 
in fixing fines. Among these is the presence or absence 
of " the caretaking which the Law requires."! H the 
animals had not been properly cared, full fines were 
exacted for trespass ; but if it could be proved that due 

* Br. Laws, in. 243, 245. t Wilde, Catalogue, 249. 

J Br. Laws, iv. 87, top line. 


care had been exercised, the fine was mitigated. The 
following were some of the precautions that should be 
taken. Hens should have a hood (cocholl), i.e. a heavy 
rag tied partly on the back and partly up the neck and 
head. In other cases their wings were clipped, and their 
feet were tied with spancels, i.e. simply bits of twine, or 
their claws were covered with rag boots. A goat should 
be hampered by having some kind of leather cover — called 
a brogue or shoe in the Law — tied on each leg : the legs 
of yearling calves should have a spancel (urcholl) : there 
should be a herdsman with cows, and a shepherd with 
sheep. At night all animals should be in their proper 
enclosures, pigs in a sty, horses in a stable (or properly 
fettered if left outside), and cows in a bawn or enclosure. 
Pigs should have a yoke (Irish srathar, pron. srahar) or 
tie on the back and legs. Pet pigs were, as we might 
expect, notorious for their mischievous propensities ; and 
they were very ingenious moreover, for they commonly 
found an opening through the fence into the neighbour's 
field, and in this manner showed the way to the whole 
herd, which were quick enough to follow their pioneer. 
Accordingly the Law lays down as much fine for a pet pig 
as for two other animals, for the first offence ; as much as 
for three for the second ; and as much as for four on the 
third occasion.* 

4. Herding, Grazing, Milking. 

Herding and Grazing. — The old word for a cowherd was 
bochaill or buachaill, which glosses bubulcus (' herdsman ') 
in one of Zeuss's eighth-century glosses (p. 183, 5 ). In 
Cormac's Glossary (p. 20) it is derived from bo, a cow, and 
cail, keeping : a ' keeper of cows ' : but in modern times 
the word buachail has come to signify ' boy ' simply without 
any reference to occupation. Another old name for a 
cowherd was boare or boaire [bo-ar-e : 3-syll.], literally 

* For all these see Br. Laws, iv. 87, 109, in, 117, 119. 


' cow-carer.' The boare of Ross, king of Ulster a.d. 248, 
was Bairche or Boirche ; and from him were named 
Beanna Boirche [Banna Borka], ' Boirche's Peaks,' now the 
Mourne Mountains in Down, because he herded the king's 
flocks there.* The account in the Dinnsenchus adds that 
when herding, his favourite look-out point was the summit 
of Slieve Slanga, now Slieve Donard, the highest of the 
range ; from which he could see southwards as far as the 
Dublin and Wicklow Mountains, and northwards as far as 
Cloch-a-stookan, or the Giant's Causeway. 

There were special keepers of cows, of sheep, of swine : 
swineherds have been treated of elsewhere. At the present 
day a shepherd is called aedhaire and treudaighe [aira, 
traidee]. As an aid to herding, bells were sometimes 
hung round the necks of cows and sheep. Animals thus 
furnished are said — in the gloss to the Senchus Mor — to 
be " privileged " (Irish uaisli : singular uasal : literally 
' noble '), which meant nothing more than that they were 
distinguished above the rest of the herd.f There was a 
fine for removing the bell. Such bells have continued in 
use till this day : and in the National Museum may be 
seen many specimens, some no doubt modern, but some 
very old. 

The nature and use of " commons " have been already 
explained (vol. I., p. 187, supra). The commons pasture 
was generally mountain-land, usually at some distance 
from the lowland homesteads ; and it was grazed in 
common and riot fenced in. Each head of a family 
belonging to the tribe or fine had the right to send his 
cattle on it, the number he was entitled to turn out being 
generally in proportion to the size of his farm. In regu- 
lating the right of grazing, animals were classified, a cow 
being taken as the unit. The legal classification was 
this : — two geese are equivalent to a sheep ; two sheep 
to one dairt, or one-year-old heifer ; two dairts to one 

* Trip. Life 423. f Br. Laws, 1. 127, 4; 143, middle. 


colpthach. or two-year-old heifer ; two colpthachs to one cow ; 
a cow and a colpthach equal to one ox. Suppose a man 
had a right to graze a certain number of cow; on the 
common : he might turn out the exact number of cows, 
or the equivalent of other animals, any way he pleased, 
so long as the total did not exceed the amount of his 

When several persons had grazing farms lying adjacent 
to each other, or when they grazed their cattle on a 
common, they often employed one herdsman to attend 
to all, who was paid by contributions from the several 
owners, each giving in proportion to the number of his 
cattle. This is what was called comingaire [4-syll.], 
i.e. ' common herding,' from ingaire or gaire, herding ; 
and under this name it is mentioned in the Brehon Law. 
The gloss on the Senchus Mor says that all those owning 
the cattle should be "in brotherhood with each other, "f 
and that each one is to be faithful [to the others] : by 
which is meant that a man, while looking specially after 
his own cattle, should, so far as he reasonably could, have 
an eye to those of his neighbours. In the sixteenth and 
seventeenth centuries it was usual for all the people of a 
village or townland, after putting down the crops in spring, 
to migrate to the uplands with their families and cattle, 
living there in temporary settlements during the summer, 
and returning to their homes in the beginning of autumn 
in time to gather in the crops. An upland settlement of 
this kind was called a buaile [booley] : and the custom 
was known as booleying by Anglo-Irish writers, several of 
whom have described it. Probably this custom descended 
from early times, for it is noticed in the gloss to the 
Senchus Mor, in the first volume of the Brehon Laws, 

P- I33-+ 

* Br. Laws, iv. 101, bottom. f Br. Laws, i. 143. 

% For more information about " booleying," see Irish Names of Places, 
vol. 1., p. 239. 


Remnants of the old regulations regarding the use of 
commons land survive in many parts of Ireland to the 
present day. There are still " commons " — generally 
mountain-land — attached to village communities, on which 
several families have a right to graze their cattle according 
to certain well-defined regulations ; and there are bogs 
where they have a right to cut peat or turf — a right of 
turbary, as they call it : and if an individual sells or other- 
wise disposes of his land, these rights gi with it. Grazing 
in common was lately found, by the Congested Districts 
Board, in full operation in Clare Island, and in re- 
arranging the land there they wisely left the old custom 
undisturbed. A curious instance existing near Swords, 
seven miles from Dublin, has been described by Judge 
Kane in a letter which has been published in the Journal 
of the Roy. Soc. of Antiqq., Irel., for 1890-91, pp. 81, 82. 
The arrangement for the classification of animals in regard 
to the right of grazing continues also in force in many 
parts of the country : the present unit being commonly the 
colpthach, i.e. a cow : now called by the people a " collop." 
I know one place in Limerick where the people still speak 
familiarly of a man owning so many collops of grass on 
the adjacent mountain-common. During the winter, when 
grass was scarce, cows were often fed on straw — and 
probably on hay — as at present.* 

Farm Life and Milking. — The people of Ireland, not the 
farming classes merely, but the general community, were 
early risers, and went early to bed : of which many 
examples might be cited. One of the geasa or pro- 
hibitions of the king of Ireland enjoined that while at 
Tara he should be always out of his bed at sunrise. The 
two daughters of King Laegaire while living in Cruachan, 
came out at sunrise to wash their hands, according to their 
custom. The bondmaid of Dubthach, St. Brigit's father, 
milked the cows at sunrise. From a statement in the 

* Br. Laws, in. 151, J? _ 


Book of Ailill it would seem that the active working-day 
in the houses of farmers began at sunrise and ended when 
the cows came to their stalls : and in the houses of chiefs 
it began when the horse-boy let out the horses in the 
morning, and ended at bed-time.* A picture of the usual 
custom of the farmer's homestead in the evening is seen 
in an incidental entry in the story of the Voyage of the 
O'Corras, where the three robbers proposed to postpone 
their evil work go d-tiosdis ba ocus innile an bhaile da 
n-drasaibh ocus da n-ionnadaibh bunaidh — " till the kine 
" and the cattle of the homestead should come to their 
" byres and their proper places, "f Women always did the 
milking, except of course in monasteries, where no women 
were employed, and the monks had to do all the work of 
the community. 

From the custom of milking early the word ambuarach 
has come to signify early in the morning : from buarach, 
a cow-spancel, and that again from bo, a cow : am-buarach, 
' in spancel-time.'J The buarach or spancel was made 
then as now of a stout rope of twisted hair, about two feet 
long, with a bit of wood — a sort of long-shaped knob — 
fixed at one end, and a loop at the other end into which 
the knob was thrust so as to fasten the spancel round the 
two hind legs of the cow. That they used a spancel — and 
a strong one too, with a big knob — in the old times is 
shown by a story in the Dinnsenchus. An able-bodied 
idle fellow, roaming about, met a girl herding her cows in 
the evening in a lonely place, and attempted violence. 
But he reckoned without his host : for she turned on him 
and knocked him down with a blow of the wooden end of 
her spancel, and then twisting the strong hair-rope tightly 
round his neck, choked him.§ 

* Br. Laws, in. 419. 

t Rev. Celt., xvi. 36 : Old Celtic Romances, 403, 404. 
X Rev. Celt., xiv. 428, „, 437. See the word buarach in this sense 
in LU, 61, b, top line. 

§ Rev. Celt., xiv. 31 : LL, '* Contents," 43, b ; and Text, 167, b, j6 . 

Sculpture i 

i a Capital : Priest's House, Glcndalough : Bcranger, 1779. 
(From Petrie's Round Towers.) 



Section i. Chief Materials. 

imber. — All the chief materials for the 
work of the various crafts were pro- 
duced at home. Of wood there was 
no stint : and there were mines of copper, 
iron, lead, and possibly of tin, which were 
worked with intelligence and success. 
We know that in early ages Ireland abounded in 
forests ; so that wood as a working material was plentiful 
everywhere. Even in the time of Giraldus Cambrensis — 
the end of the twelfth century — when clearances and 
cultivation had gone on for a thousand years, the greater 
part of the country was clothed with trees. He says : — 
" Ireland is well wooded and marshy. The [open] plains 
are of limited extent compared with the woods."* The 
common Irish word for a tree was, and is still, crann : a 
wood is coill or fid. The Brehon Code (iv. 147), in setting 
forth the law for illegally felling trees, divides them into 
four classes, with a special fine for each class : — 

1. " Chieftain " trees (airigh feada) : oak (dair) ; hazel 
(coll) ; holly (cuileann) ; yew (ibur) ; ash (uindius, more 

* Top. Hib., 1. iv 



commonly called fuinnse, fuinnsecnn, or fuinnsedg) ; pine 
(ochtach or giumhas) ; apple (aball). 

2. " Common " trees : alder (fernn) ; willow (sail) ; 
hawthorn (sceith) ; mountain-ash, or rowan-tree, or quicken 
tree (caerthann) ; birch (beithe) ; elm (learn) ; and another 
which is not known from its Irish name, idha. 

3. " Shrub " trees : blackthorn or sloebush (draidean or 
droigen) ; elder or boor tree (trom) ; white hazel (finncoll) ; 
aspen (crithach, lit. ' shaking ') ; arbutus (caithne) ; and 
two others not known from their names, feorus and 

4. Bramble trees : fern (raith) ; bog-myrtle (rait) ; 
furze (aiteand) ; briar (dris) ; heath (fraech) ; ivy (eideand) ; 
broom (gilcach) ; gooseberry (spin). 

The commentator on the Law proceeds to state the 
qualities or circumstances that give each of the " chieftain " 
trees its " nobility." The oak : its nobleness in size and 
appearance, and its meas or acorns (for feeding swine : 
Irish dearcan, an acorn). Curious, no mention is made of 
its bark, which was very valuable for tanning (see p. 367, 
below). The hazel: its nuts (see p. 155, supra), and its 
wattles, for building wicker-houses. The apple : its fruit 
and its bark — which was probably used for tanning. The 
yew : its noble structures : i.e. the value of its timber for 
ornamental furniture, household vessels, and building. 
This tree was produced in great abundance : so much so 
that Giraldus Cambrensis records an opinion that the 
poisonous juices and exhalations of the yew-trees seriously 
checked the increase of bees. The holly : because it was 
made into chariot-shafts ; and for another reason, but 
here the Irish statement is unintelligible to me. The 
ash : " supporting a king's thigh " : i.e. probably it was 
used in making the king's throne ; also " half the furniture 
of arms " ; that is, the handles of spears were made of 
it. The pine : because its wood was used in making 


Among the various materials mentioned in the Senchus 
Mor is whalebone, which is called the fabra or ' fringe- 
bones ' of a mil-mor or whale. The gloss says that it was 
used for making saddle-trees and the bottoms of sieves : 
and also occasionally for hoops of [small] vessels when 
suitable wood-hoops were not to be had.* 

Metals. — The metallic weapons and tools preserved in 
our museums are generally either of bronze (sometimes 
brass, occasionally copper) or iron. The bronze objects 
far outnumber those of iron, which is partly explained by 
the fact that iron rusts and wastes away much more 
quickly than bronze. It is generally recognised that the 
three materials — stone, bronze, iron — represent three suc- 
cessive stages of human progress : that is to say, stone in 
its use as a material for tools and weapons, is more ancient 
than bronze, and bronze than iron. But there was no 
sudden or well-marked change from one to another : they 
all overlap. Stone was used in a primitive stage when 
bronze was not known ; but it continued to be used long 
after the introduction of bronze. So bronze was used for 
some long period before iron was known ; but continued 
in use long after the discovery of iron. And more than 
that : all three were used together down into Christian 

That the ancient Irish were familiar with mines, and 
with the modes of smelting and of extracting metals of 
various kinds from the ore, is shown by the frequent 
notices of mines and mining both in the Laws and in the 
general literature. The Law (iv. 279) enumerates eleven 
things that add to the value of land, among which is a 
mine of copper or of iron. The Senchus Mor mentions a 
penalty for digging a silver mine without the permission of 
the ownerf ; from which we may infer that a mine was, or 
might be, the private property of the owner of the land : a 
fact which is still more clearly stated in the Book of 

* Br. Laws, i. 125, I0 ; 135, last paragraph. f Ibid., 1. 167, 171. 


Aicill.* An ancient Irish ms. tract of the Brehon Laws, 
quoted by Petrie (R. Towers, 219), gives the pay of the 
delver who digs copper ore. When Connall Cernach was 
fighting the men of Connaught, while retreating in his 
chariot, he came to a river. " There were miners washing 
ore " (batar mianaighe ac nige mianaigh) in the river above 
him : and he difficulty he experienced in finding the exact 
fording place and crossing the turbid and troubled water 
enabled his pursuers to overtake and kill him.f Here the 
washing of ore is mentioned as quite an ordinary occur- 
rence ; and in many others of the oldest Irish tracts the 
smelting of ore is frequently referred to as a matter very 
familiar. The hard breathing of champions fighting is 
compared to the bellows-blowing of smiths smelting ore 
{tuaircnech nan goband ic meinlegad miannaig : LL. 218, 
b, 6 bot.). The plain now called the barony of Fermoy 
in Cork must have been famous for its mines, for it 
was anciently known as Magh meine, the ' plain of 
minerals. 'I 

The truth of all this documentary testimony — and 
much more might be adduced — is fully confirmed by 
evidence under our own eyes. Sir Richard Griffith — in 
his Report to the Royal Dublin Society, 1828 — remarks 
that the numbers of ancient mine excavations still visible 
in every part of Ireland, prove that " an ardent spirit of 
mining adventure " must have pervaded the country at 
some remote period. He instances old copper mines at 
Mucruss near Killarney, and at Ballydehob in Cork, old 
coal mines at Ballycastle, Co. Antrim, and the lead mines 
of Milltown in Clare, the oldest mines perhaps in Ireland.^ 
In these last many rude tools were found, such as oaken 
shovels and iron picks of extraordinary size and weight. 
O'Halloran also, in the Introduction to his " History of 

* Br. Laws, in. 203. 

f Death of Conall Cernach in Zeitschr. fur Celt. Phil., 1. 108. 

% O'Curry, Man. & Cust., 1. 215. 


Ireland," mentions the antique mining shafts on the Wes- 
tropp estate in Limerick, near the Shannon : and very 
ancient copper mines have been found at Knockmahon 
in Waterford.* 

The usual Irish words for smelting metals were brtith 
and berbad [bruh, berva], both of which signify ' boiling.' 
A smelter was called bruithneoir [bruhnore], meaning, as 
O'Clery's Glossary expresses it, " a man [employed in] 
" boiling or melting [ag bearbhadh no ag leaghadh] gold or 
silver or metal." Bruth and caer were both applied to a 
mass of any kind of metal. Of the detailed smelting 
processes of the Irish we have very little knowledge. But 
we know that, whether these arts grew from within or were 
brought hither by the first immigrants, the Irish miners 
successfully extracted from their ores all the native metals 
then known, f 

In Ireland as elsewhere copper was known before 
iron. It was almost always used as bronze, which will 
be treated of at page 297, farther on. We have unques- 
tionable documentary evidence — such as the " Confession of 
St. Patrick " — that iron was in familiar use in Ireland in the 
fifth century of the Christian era : and as we learn from 
Tacitus that the Caledonians used iron swords in his time, 
it is certain that this metal was known in Ireland at least 
as early as the first century : probably much earlier. 
According to tradition the iron mines of Slieve-an-ierin^ 
east of Lough Allen in the County of Leitrim (Sliabh- 
an-iairn, the mountain of iron), were worked by Goibniu, 
the great Dedannan smith ; and it is now as celebrated 
for its iron ore as it was when it got the name, long ages 
ago. In the Book of Rights we find it repeatedly stated 
that masses of iron were sometimes given as tribute to 
kings. In the old tale of the Death of Goll and Garbh, 
from the book of Leinster, steel is mentioned under the 

* See Brash's Article on Ancient Mining in Ireland, Kilk. Archaeol- 
Journ., 1S70-1, p. 509. j See Wilde's remarks : Catalogue 350-7. 


name of criiaid [croo-ee], which means ' hard ' ; and hence 
came the name of Cuculainn's sword, criiadin, a diminutive 
form of criiaid. Among the pagan remains found in a 
earn at Loughcrew were many specimens of iron imple- 
ments, all, as might be expected, very much corroded by 
rust. One was " an iron punch five inches long with a 
" chisel-shaped point bearing evidence of the use of the 
" mallet at the other end."* This was probably used for 
punching the patterns on gold ornaments (vol. I., p. 566, 
supra) . 

The Irish word for iron is not very different from the 
English : — iarann, Old Irish form iam [both pronounced 
eeran], and the word exists in various forms in Welsh 
and in several of the northern languages ; such as Gothic, 
cisarn ; Old High German, isarn ; Anglo-Saxon, iren ; 
Welsh, heyrn. 

Sir Robert Kanef says that tinstone has been found 
only in the auriferous soil of Wicklow. But Smith, in 
his " History of Kerry " (p. 125), states that he 'found 
near the lake of Killarney an ore which contained tin : 
and, according to Sir Richard Griffith, tin occurs in 
combination with lead and zinc in Dalkey, near Dublin. 
There is a very ancient tradition recorded by Nennius 
as well as by native Irish writers, that one of the 
" Wonders of Ireland " was the four metallic circles 
surrounding Loch Lein or the Lake of Killarney, viz. 
a circle of tin, a circle of lead, a circle of iron, and a 
circle of copper! : which, so far as tin is concerned, is 
corroborated by Smith's experience. But whether tin 
was mined at home or imported from Cornwall — or 
both, as is more likely — it was constantly used in making 
bronze: and often without any mixture. "The ores of 
" lead seem to occur in more places than those of any 

* Fergusson, Rude Stone Monuments, 218. 

t Industrial Resources, p. 210. 

J Irish Nennius, 220. See Kinahan's Geol. Irel., p. 357. 


" other metal."* The mines were worked too, so that 
the metal was sufficiently abundant : the very old lead 
mines of Milltown have been noticed at p. 289. Zinc, 
which was chiefly used in making brass, was also found, 
commonly in connexion with lead. Gold and silver have 
been already treated of. 

2. Builders. 

From the most remote times there were in Ireland 
professional architects or builders, as there were smiths, 
poets, historians, physicians, and druids ; and we find them 
often mentioned in our earliest literature. Even the very 
names of the mythical builders of Tara, Emain, Ailech, 
and other royal residences have been preserved. 

There were two main branches of the builder's pro- 
fession : — stone-building and wood-building. An ollave 
builder (1. 442, supra) was supposed to be master of both, 
and, in addition to this, to be so far acquainted with many 
subordinate crafts as to be able to " superintend " them, 
as the Law (v. 95, 4 ) expresses it : in other words, to be a 
thorough judge as to whether the work was properly 
turned out by the several tradesmen, so as to be able to 
pass or reject as the works deserved : all which resembles 
what is expected from architects and builders of the 
present day. 

The most distinguished ollave builder of a district was 
taken into the direct service of the king, and received from 
him a yearly stipend of twenty-one cows, answering to a 
iixed salary of £250 or £300 of the present day : for 
which he was to oversee and have properly executed 
all the king's building and other structural works. In 
addition to this he was permitted to exercise his art foi 
the general public for pay : and as he had a great name, 
and had pknty of time on hands, he usually made a large 

• Kinahan, p. 348. 


income. In one of the Brehon Law tracts,* there is a 
curious classification of the works an ollave builder might 
undertake, with the payment fixed for each, as taken 
separately : nineteen classes in all, which are as follows : — 
The two most important — (i) Building in stone and 
{2) Building in wood — are named first, as he was to be 
thorough master of these : six cows each. (3) Ibroracht, 
i.e. ' yew- work ' : six cows. (4) Cook-house or kitchen- 
building : six cows. (5) Mill-building : six cows. Con- 
structing the three following — (6) [large] ships : (7) barcas 
or ordinary small ships ; and (8) currachs or wicker boats : 
four cows each. (9) Making wooden vessels, namely 
vats, tubs, keeves of oak, and small vessels : four cows. 
(10) Uamairecht, conjectured to be ' cellar-making ' (nam, 
>a cave) ; perhaps making the subterranean stone-house 
under a rath (see p. 56, supra) : two cows. Constructing 
the three following — (11) causeways ; (12) stone walls ; 
(13) clochans or stepping-stones across a river : two cows 
each. For the three following — (14) carvings in. wood 
(rinnaighecht , pron. rinneeght) ; (15) crosses ; (16) chariots : 
two cows each. For these three — (17) wickerwork houses ; 
(18) shields ; (19) bridges : two cows each. Builders of 
the inferior grades (below the ollave) had correspondingly 
lower fees. 

It will be observed that in most of the above there is 
an absence of distinct specification as to quantity or time, 
as to who supplied materials, or paid the workmen, &c. ; 
but, as in many others of the Brehon Law provisions, all 
this was regulated by custom, which was at the time so 
universally understood and recognised that it was not 
considered necessary to put it in writing. As illustrating 
the systematic way in which the Law attempted to pro- 
vide for all such matters, it is worthy of remark that the 
permanent stipend of twenty-one cows received by a builder 

* Br. Laws, v. 93, 95 : Petrie, Round Towers, 346 : O'Curry, Man. 
& Cust., 11. 52, et seq. 


from the king was calculated on the above charges, in this 
way : — Full fees allowed for the first two works (stone- 
building and wood-building) : six cows each ; and one- 
sixth fees for all the others combined, i.e. one-sixth of 54 : 
9 cows ; which with the first twelve make 21 cows. Some 
of the handicrafts mentioned in the above list will be 
noticed in the present chapter : others have been or will be 
dealt with in other parts of this book. 

By far the most celebrated of all the ancient architects 
of Ireland was the Gobban Saer, who flourished in the 
seventh century of our era, and who therefore comes well 
within historic times. The best accounts represent him as 
a native of Turvey near Malahide, north of Dublin : and 
he is mentioned in the Lives of many of the Irish Saints 
as having been employed by them to build churches, 
oratories, and houses, some of which still retain his name. 
This great builder fills a prominent place in all sorts of 
Irish literature from his own time downwards ; he is men- 
tioned in the eighth-century poem referred to in vol. L, 
p. 230, supra — almost contemporary with himself ; and t© 
this day the peasantry all over Ireland tell numerous 
stories about him.* 

3. Braziers and Founders. 

Dan [dawn] is a general word for any art, science, or 
trade : and aes-ddna [' men of art '] is applied to those 
skilled in such arts. In the commentary on the Senchus 
Mor (Br. Laws, 11. 119), smiths, carpenters, shield-makers, 
physicians, and poets, are called collectively aes-ddna : but 
most commonly the term aes-ddna meant poets. Some- 
times an artisan was termed simply dan : but there were 
of course different epithets to distinguish the various 
callings. It was however usual to restrict the applica- 

* Almost everything that is known of his authentic history has been 
brought together by Dr. Petrie in his Round Towers, 385-7. Several 
popular stories about him will be found in the Dublin Penny Journal. 


tion of dan to a poem or poetry : whence a poet was often 
called fer-ddna, ' man of poetry.' 

The word goba, gen. gobann [gow, gowan], is applied 
to a worker in iron — a smith : cerd or cerrd [caird], to a 
worker in brass, gold, and silver — a brazier, goldsmith, or 
silversmith : saer to a carpenter, builder, or mason — a 
worker in timber or stone. Sometimes a bronze or brass- 
worker was called umhaidhe [oovee], from* uma (p. 297, 
infra). These are the usual applications : but as the arts 
and trades sometimes overlap, so the words are often 
applied in somewhat more extended senses : for example, 
Culann, the mighty smith of the Red Branch Knights, is 
called a cerd in the Book of Leinster.* Still they are 
generally distinguished, especially in Christian times : and 
we find goba and cerd sharply defined in a passage of 
the Tripartite Life (266, 267), specifying the duties of 
St. Patrick's household, where we are told that his three 
smiths (gobainn : pi.) made bells for him (which at that 
time was smithwork, as they were made of hammered 
iron), while his three braziers (cerdae : pi.) made chalices 
and other brazen and bronze vessels for the altar. The 
three classes of artists are also well distinguished in a 
passage in the " Battle of Moyrath " (p. 103), in which the 
skill of cerdae, gobainn, and saeir (all three words plural) is 
praised. The three mythical artisans of the Dedannans, 
the brothers of Diancecht the physician (1. 261, supra), 
were Goibniu (gen. Goibnenn) the goba or smith, who made 
their spearheads and swords : Creidne the cerd, who supplied 
rivets for the spears, hilts for the swords, and bosses and 
rims for the shields : and Luchtine the saer or carpenter, 
who made their wooden and wicker shields and spear- 
handles, f A goba and a cerd are distinguished also in the 
Brehon Law rule (ill. 193), that goods found in a kiln, a 

* LL, 63, a, 22; b, , 7 . 

J Corm. Gloss., 123, under " Nescoit." See also Man. & Cust., 1. 
246, 248, 249. 




kitchen, a forge, or a mill, left in charge of the owner, 
if they were unconnected with the proper business of 
the place, were forfeited : and the particular instance of 
this rule given is: — If gold, silver, or bronze was found 
in the forge of a goba (blacksmith), it was forfeited : 
because these materials had no connexion with the 
business of a smith, but belonged to that of a cerd. 

The word ctrdd glosses acrarius 
in Zeuss (page 60, 43) : and in 
the form of caird — which exactly 

Tig. ?8c,. 

Fig. 290. 

Bra/iers' or Goldsmiths' Anvils. Fig. 289 is the natural size, and is much worn : the little shallow 
holes were for riveting. (From Wilde's Catalogue, p. 523.) Fig. 290 is 3 inches high, and 1J4 inch 
thick. Two of its corners form right angles : one is rounded : one bevelled : so as to suit the 
different shapes required. (From Kilk. Archxol. Journ,, 1885-86, p. 538.) 

represents the sound — it has held its place as a living 
word in Scotland, even among speakers of English, but 
it is applied to a tinker : — 

" Her charms had struck a sturdy caird, 
As weel as poor gut-scraper." 


Aerarius, which, according to the glossator of twelve 
hundred years ago, is equivalent to cerdd, signifies literally 
a ' worker in brass ' ; and, curiously enough, this corre- 


sponds exactly with the description the caird gives of 
himself in Burns's poem : — 

" My bonnie lass, I work in brass, 
A tinkler is my station." 

The work of a cerd proper has been dealt with in the 
chapter on art. 

We have already seen that the ancient Irish were very 
skilful in metallic art. Metallic compounds were carefully 
and successfully studied, copper commonly forming one 
of the ingredients. The most general alloy was bronze, 
formed of copper and tin : but brass, a compound of 
copper and zinc, was also used. The Irish name for 
copper was uma, gen. umai, whence the Irish word umaide 
or umhaidhe [oo-vee], one of the names of a brazier — a 
bronze- or brass-worker ; for this word uma is used also 
to denote both bronze and brass. Thus, according to the 
Tripartite Life, the chieftain Dare gave St. Patrick a 
caldron of uma, which it is pretty certain was made, -not 
of pure copper, but of bronze ; for all the caldrons still 
preserved are of that alloy. There were several special 
terms for different alloys, each no doubt designating a 
compound of certain metals in definite proportions ; but 
the exact compounds referred to by some of these terms 
are unknown to us. 

There were two chief kinds of bronze, red and white, 
or rather reddish and whitish. The red bronze was called 
derg-uma {derg, red) or cred-uma (for I take it that these 
two words mean the same thing) and sometimes cred, 
simply ; and the white was called finn-uma (finn, white) 
or findruine [fin-dnna], two terms that also seem to me 
to be identical. Findruine was much more expensive 
than creduma, and was kept for the finer kinds of work. 
Assuming that the ancient Irish pinginn or penny repre- 
sented in those times a value equivalent to that now 
represented by 6s. 8d. of our money — which may be 


f 6s. 8d. 

represented J 

1 3s. 4d. 

in present 1 

I y. *d. 

value by 

1 is. 8d. 


taken as approximately true — a statement in an ancient 
authority quoted by Petrie (" Round Towers," 219) enables 
us to assign value on a similar basis to one ounce of each 
of the following metallic materials : — 

1 oz. of finn-uma, findruine, or white bronze 
r oz. of derg-uma, creduma, or red bronze 
1 oz. of zinc ...... 

1 oz. of lead ...... 

The difference in value between the two kinds of 
bronze is recognised in the tales : as when Queen Maive 
estimates the comparative merits of the three heroes : — 
" The difference between creduma and findruine is between 
" Loegaire Buadach and Conall Cernach ; and the difference 
" between findruine and red-gold is between Conall Cernach 
" and Cuculainn." Accordingly she proceeds to give effect 
to her judgment by presenting the three heroes with three 
goblets of values according to merit.* The red bronze may 
be seen in the spear-heads and caldrons in the National 
Museum, and the findruine or white bronze in the orna- 
mental shrines, and other ancient works of art. Many 
of the spear-heads and other bronze articles belong to a 
period some centuries before the Christian era. 

Metal-casting is very often referred to in general terms 
in our literature, showing how familiar it was : and through 
these incidental references we get now and then a glimpse 
at the artists' tools and appliances. The workmen used 
charcoal for their fires, that made from birch-wood, as we 
have seen (vol. 1., p. 565, supra), giving the greatest heat 
then attainable, sufficient — with the help of a flux — to 
melt all ordinary metals. They used a ladle (Irish liach) 
to pour out the melted metal ; and it had to be used 
carefully, for the Book of Aicill (Br. Laws, in. 213) has a 
series of provisions for accidents, mentioning damages for 
injuries to persons or animals during casting, and also 

* Fled Brier., p. 75-79, and 93-95- 


during the process of mining. All this indicates how 
generally metal-founding was practised. 

A moulding-compass used by founders was called 
luaithrinu [loo-rin], compounded of luath [looa], ashes, 
and rinn, a point. In Cormac's Glossary this 
word is used characteristically in describing 
the whirlpool of CoireBrecain, 
where he says the waters 
whirl round " in the likeness 
of moulding-compasses" 
[fo cosviailius luaitlirinde)* 
showing how familiar the im- 
plement was in the ninth and 
tenth centuries. The exact 
use of the moulding-compass 
and the origin of its name, are 
instructively illustrated in a 
legendary story quoted from 
an ancient manuscript by 
O'Curry.f Mac Enge, a shield- 
maker, was employed to make 
shields for the Ultonians, and 
had exhausted all his patterns 
(for each chief had a special 
design for himself: see vol. I. 
p. 125, supra) when Cuculainn 
came to him for a shield, and 
demanded a design different 
from all the others. While 
the artist was puzzling his brain trying to invent a new 
device, a man having in his hand a small fork with two 
sharp prongs came up to him and said : " Spread ashes 
(luath) on the floor of your workshop " : which he did. 
Then the man planted one prong of the fork in the ashes 

#Corm. Gloss., 41 : Th'ee Irish Glossaries, 13. 
t Man. & Gust., r. 329. 

Fig. 291. 

Ornamental inlaid hook, 
natural size. Possibly for 
suspending a sword. The 
scroll-work indicates that 
it belongs to Christian 
times (voi. I., p. 551). Now 
preserved in the National 
Museum. (From Wilde's 
Catalogue, p. 572.) 

See next page. 

Fig. 292. 

Spear-head, now in 
National Museum, 
where many equally 
or more beautiful are 
preserved. Others 
of very graceful 
form, and of admi- 
rable workmanship, 
are figured in voi. l» 
pp. 107, no, in, sh- 
pra. (From Wilde's 
Catalogue, p. 499) 
See next page. 




Fig. 293. 

Vm. 294. 

and with the other described circular devices for the hero's 
shield. Accordingly luaithrindi or luaithrind, ' ashes- 
engraver,' was thenceforward the name of this sort of fork 
or compass. 

The exquisite skill of the ancient Irish braziers is best 

proved by the articles 
they made, of which 
hundreds are preserved 
in our Museum. Two 
illustrations are given on 
last page (figs. 291 and 
292); a beautiful speci- 
men of enamelled metal- 
work is described at page 
258, supra, and shown in 
fig. 280 ; and others will 
be found in various parts 
of this book. As to the 
hook represented in' fig. 
291, Wilde describes it 
as " one of the t most 
"beautiful specimens of 
"inlaying bronze with 
"silver and some dark 
" metal (after the fashion 
" of the ancient niello) 
" which has yet been 
" discovered in Ireland." 
The gracefully - shaped 
spear-heads, which, in point of artistic excellence, are fully 
equal to any of those found in Greece, Rome, or Egypt, 
were cast in moulds : and we have not only the spear- 
heads themselves but many of the moulds, usually of stone, 
proving — if proof was needed — that all these articles were 
of native manufacture. In one glass case in the National 
Museum there are more than forty moulds for celts, spear- 

Fig. 295. 

Stone Moulds. Figs. 293 and 294 in National Museum. 
DuMin: fig. 295 in the Belfast Museum. (From Wilde's 
Catalogue, pp. 91 and 392.) 


heads, arrow-heads*, &c. : some looking as fresh as if they 
had been in use yesterday. Probably clay- and sand- 
moulds were used ; but these would not be preserved.* The 
old cairds were equally accomplished in making articles of 
hammered bronze, of which the most characteristic and 
important are the beautifully-formed caldrons — many of 
exquisite workmanship — made of a series of bronze plates, 
hammered into shape and riveted together. Of these 
numerous specimens may be seen in the National Museum. 
One will be found figured at page 125, supra (fig. 215), and 
another in vol. I., p. 21, fig. 13. In both of them the heads 
of the rivets project outwards so as to form ornamental 
conical studs, a kind of ornament used in other metal- 
work, as in the bronze trumpets and gold gorgets. But 
all caldrons had not these cone-headed rivets. A hand- 
some dish, hammered into shape with great skill from a 
single flat piece of bronze, is shown on fig. 196, p. 71, 

4. The Blacksmith and his Forge. 

In a state of society when war was regarded as the 
most noble of all professions, and before the invention of 
gunpowder, those who manufactured swords and spears 
were naturally looked upon as very important personages. 
In Ireland they were held in great estimation ; and in the 
historical and legendary tales, we find smiths entertaining 
kings, princes, and chiefs, and entertained by them in turn. 
We know that Vulcan was a Grecian god ; and the ancient 
Irish had their smith-god, the Dedannan Goibniu, who 
figures in many of the old romances. It sometimes was 
considered an additional distinction in a chief or warrior 
to be a good metal-worker. Fergus mac Roy, to show 
his fitness for the duty of rearing the infant Cuculainn, 
enumerates his accomplishments, and among other things 
says, " I am a good craftsman. "f Smiths, like the men 

* See Wilde, in Catalogue, 392. t Miss Hull, Cuch. Saga, 19. 


of other arts and professions, were of different grades : 
an ollave-goba or prim-goba (' prime-smith ') or flaith- 
goba (' chief-smith ') being of the highest rank. We find 
these distinctions often mentioned in both secular and 
ecclesiastical writings,* showing that they were real, and 
universally understood and acknowledged. 

Cerdcha or cerddchae originally meant a workshop in 
general, derived, according to Cormac's Glossary (p. 46, 
" Ca ") from cerd, an artisan, and ca,. a house : " artisan- 
house.' In " Three Fragments of Irish Annals " (p. 223), 
it is applied to the workshop of a fuller of cloth : and in 
Zeuss (60, 44 ) it glosses the Latin officina, a workshop of 
any kind. But its most usual application was to a forge : 
and it is still so applied, and pronounced cartha (the first 
syll. long, as in car). A forge was in old times regarded as 
one of the important centres of a district. If, for instance, 
horses whose owners were not known were impounded 
for trespass, notice had to be sent to the dun or fortress of 
the nearest lord, to the principal church, to the fort of the 
brehon of the place, and to the forge of the smith (Br. 
Laws, iv. 107) ; and in like manner notice of a waif should 
be sent to seven leading persons, among them the chief 
smith of the district {ibid., III. 273). For forges were 
places well frequented, as they are at the present day, 
partly by those who came to get work done, and partly 
by idlers. And sometimes individuals took a nap with 
comfort and laziness, as we know from this provision of 
the Book of Aicill : — that if any one who had business 
at the forge fell asleep while waiting for his turn, it was 
the duty of the smith or the bellows-blower to awaken 
him when dangerous showers of sparks were flying 
about, otherwise they were liable (with some limitations) 
for any injury that befell him (Br. Laws, in. 191, and 
note 2). 

* For instance, Br. Laws, in., p. 273, 22i Oss. Soc, iv. 299, lS; and 
Stokes, Lives of SS., 235. 


The anvil [inneoin : pron. innone) was placed on a 
block or stock (cepp : pron. kepp) : in cepp i mbdi ind 
inneoin : " the block on which the anvil is set " (LL. 
35, 0> 5)- The anvil must have been shaped something 
like those in use now ; with a long projecting snout on 
the side : for in a passage in Cormac's Glossary (p. 135) 
describing an ugly-looking giant, it is said : " His nose is 
larger than the urbuinde [oorbinne] or ' anvil-snout ' of a 
smith." The anvil was large and heavy, as we may infer 
from the following story. On one occasion King Ochy 
Moyvane, passing by a large forge, saw his five sons 
standing inside : and wishing to test their courage, he 
quietly set fire to the building and shouted to them to 
save the smith's property. Four of them took out small 
and portable things ; but Niall seized the heavy and 
valuable articles and removed them one by one, among 
them the bellows and the anvil-and-block, while the house 
was blazing round him.* This young prince subsequently 
succeeded his father, and is well known as Niall of the 
Nine Hostages. Yet the anvil of those times could not 
have been as massive and firm as the present ponderous 
anvil : for in the Book of Aicill provision is made to meet 
the case of the sledge (Irish ord) breaking or injuring the 
anvil, or the sledge itself breaking on the anvil, either 
through the carelessness of the sledger, or because the 
smith held the red-hot iron in an awkward positionf : all 
which would indicate that neither anvil nor sledge had the 
solidity or weight of those now in use. 

If the anvil was not well secured on the cepp or block, 
it was liable to slip off during working : or the sledge 
might slip off the anvil if struck awkwardly ; or the 
head of the sledge might fly off the handle if fastened 
insecurely ; or the whole sledge might slip from the hand 
of an awkward sledger ; or two sledges might come into 

* O'Curry, Man. & Cust., 1. 147 : LL, 35, a. See for another version 
Rev. Celt , xxiv. 195. f Br. Laws, ill. 191. 


collision : in any one of these cases injury to persons 
might result, for which the Law (in. 189) made provision 
for compensation by the person in fault. So far as I am 
aware no ancient blacksmith's anvil is to be seen in any 
of our museums ; but small braziers' anvils made of bronze 
have been found, two of which are figured at p. 296, supra. 
Another small anvil like these was found in a crannoge in 
Ulster. Ancient anvils, especially large ones, are rare in 
all parts of the British Islands. 

The smith held the red-hot iron in a tennchair [tinne- 
her], pincers or tongs. In the " Voyage of Maildune," 
as the boat approached an island inhabited by gigantic 
blacksmiths, the adventurers heard the thundering sound 
of smiths' hammers striking a red-hot mass of iron on the 
anvil : and as soon as the smiths saw the boat one burly 
fellow rushed out with a great piece of glowing iron in the 
tongs (tenchoir) and flung it after the curragh* : which, 
however, it missed. A similar incident befell St. Brendan 
as related in his Voyagef : and both remind us of Ulysses' 
escape from the Cyclops. While the smith held the 
glowing iron on the anvil, another person struck it with the 
ord or sledge ; and sometimes two persons were sledging 
at the same time (Br. Law, ill. 189). It is to be presumed 
that the smith used a hand-hammer like those of the 
present day. 

Making and fixing of rivets (seman or semman, a rivet} 
was part of the work of either smith or brazier, but the 
brazier usually put them in spear-heads and swords. In 
some of the swords and spear-heads in the museum, the 
rivets still remain. 

A water-trough was kept in the forge, commonly called 
umar, and sometimes telchuma ; but this last word is also 
used to denote a barrel or puncheon. The smith kept a 
supply of wood-charcoal in bags, called cual craing, or 

* Rev. Celt. x. 53 : Old Celtic Romances, 145. 
t Brendaniana, 161. 


cual craind, i.e. " coal of wood."* I do not know if coal 
from the mine was used : but the distinctive term cual 
craing would seem to imply that it was : and besides, as 
already remarked (p. 289, supra), very ancient coal mines 
have been found near Ballycastle. The smith wore an 
apron commonly of buckskin, like those smiths wear 

The last of the smith's appliances to be noticed is the 
bellows. The Irish name for a smith's bellows is builcc or 
builgg [bullig], which is merely the plural form of bolg, a 
bag, like the English bellows ; in the Book of Leinster 
the plural article is in one place brought in, na builgg, * the 
bags ' ; all indicating that, in Ireland as in other countries, 
the primitive bellows consisted of at least two bags, which 
of course were made of leather. Why two bags were used 
is obvious — in order to keep up a continuous blast ; each 
being kept blowing in turn while the other was filling. 
This word builcc the Irish continued to employ for their 
bellows, even in its most improved form, just as we now 
call the instruments we have in use " bellows," though this 
word originally meant ' bags,' like the Irish builcc. The 
following passages relating to the use of the forge-bellows 
will give us some idea of its construction. In the story of 
the " Courtship of Emer " we are told that when Cuculainn 
and the other heroes went to be trained by Domnall, the 
great Scotch instructor in military and athletic exercises, 
he set them to practise, in the first instance, on a bellows, 
and on a spear, as a sort of preliminary exercise to attune 
their muscles properly for learning — what they came to 
learn — the special and more difficult battle-feats : — " they 
were taught by him " — says the old text — " one thing on 
" the flagstone of the small hole, namely, to blow bellows 
" (foseted cethar bolcc) : they had to work on it till the soles 
" of their feet were all but black or livid : and [they were 

* Telchuma and cual (or cuail) craing. LL, 35, a, « . see also vol. 1., 
p. 565, supra. 


" taught] another thing on a spear, on which they were set 
" to climb np."* 

An independent and probably older authority is part 
of an elegy on a smith by his wife quoted in Cormac's 
Glossary (which itself belongs to the ninth or tenth century), 
in which occurs the following passage : — " The red flame 
" of his furnace mounted up to the roof : sweet were the 
" murmurs that his bellows (a di bolg) used to chant to the 
" hole of his furnace. "f 

In the Book of AicillJ the rule is laid down that if 
sparks from a smith's fire injured a bystander under certain 
circumstances, the bellows-blower was liable for damages 
if he had blown with unnecessary violence, so as to scatter 
showers of dangerous sparks : but if he had done so by the 
direction of the smith, then both were liable in equal 

These passages will enable us in a measure to recon- 
struct the old Irish smith's bellows, and exhibit the mode 
of working it. From the Brehon Law extract last quoted 
we see that in every forge there was a special bellows- 
blower, who blew strongly or gently as occasion required, 
sometimes directed by the smith. From the passage in 
the " Courtship of Emer," where the heroes are set to blow, 
we may infer that the bellows were worked with the naked 
feet, that it took some time to learn how to do so, and that 
the bellows was large and laborious to work, since it taxed 
the strength of mighty heroes. That it was large and 
heavy we know also from the story of Niall at p. 303. 

* The original Irish of that part of this passage relating to the bellows, 
as printed by Kuno Meyer, is this : — Forceta leiss aill for lice detcain [edon] 
foseted cetharbolec : noclistis fuiri iarom napdar dnba na glassa a fond 
(Rev. Celt., xi., pp. 444, 445) Another version of the same passage, 
slightly but not materially, different, will be found in the Stowe MS., 
D, 4, 2, fol. 82, b, col. 2, line 8, in the Roy. Ir. Academy, Dublin. 

f The original Irish of the above passage, as printed by Stokes, is 
this : — For bir ifraig dercc anis ; babinde nochantais dord friderc aneis 
afli bolg (Three Ir. Glossaries, 32 : Corm. Gloss. 124). 

% Br. Laws, in. 191. 


The passage in Cormac's Glossary speaks of the sweet 
murmurs of the bellows blowing through the " hole of 
the furnace " ; while that from the " Courtship of Emer " 
gives us a somewhat closer view by the expression " the 
flagstone of the small hole " [lice, a flagstone : derc, a 
hole : diminutive dercan, with its genitive dercain, a small 
hole). All this means that the smith's hearth or furnace 
was constructed of flags, in one of which was a small hole 
through which the pipe directed the air-current from the 
bellows into the fire. It is doubtful whether this smali 
hole was in the under flag on which the fire was placed, 
or in the bottom of one of the side-flags : but it is an 
interesting fact that at the 
present day, in some parts 
of Ireland, the fire in ordi- 
nary dwelling - houses is 
often blown — as I have 
seen done — with a small 
fan-bellows up through an 
aperture in the hearthstone 
by means of a pipe run- 
ning from the blowing in- 
strument under the floor 
to the aperture. 

The name given to the bellows in Cormac's Glossary 
— di bolg, " two bags " — indicates that the bellows in view 
here had two separate chambers lying side by side. Each 
of these must have consisted of an upper and an under 
board with sides of leather : and in the under board of 
each was a simple clapper-valve as in our present kitchen- 
bellows. From each chamber extended a pipe, the two 
pipes uniting into one which was inserted into the hole 
in the flagstone. The two chambers were placed close 
to each other, and there must have been a short crossbeam 
or lever (aa in fig. 296) turning on a centre pivot, with its 
two ends loosely fastened to the two backward projections 

Fig. 296. 

Conjectural plan of double or two-chambered 
force-bellows. The bellows-blower stood with his 
feet on Bis, and his face towards the fire. AA, the 
cross-beam or lever, turning on its centre-fulcrum. 
CC, clapper-valves in bottom boards. The rest of 
the diagram explains itself. 




of the upper boards. The bellows-blower stood on top, 
one foot on each board (at bb), and pressed the two down 
alternately. As each was pressed down, and its chamber 
emptied through the pipe, the other was drawn up by its 
own end of the cross-beam, and the chamber was filled 
through the clapper-valve at bottom : and thus the 
chambers were compressed and expanded in turn so as 
to keep up a continuous blast. There was a cross-bar 
fixed firmly above the bellows for the blower to grasp 
with his hands, so as to steady him and enable him to 
thrust downwards with his feet when a strong blast was 
required, like a modern bicyclist when mounting a hill. 
But there was another and a better sort of bellows, 

having four chambers, 
as we see by the name 
employed in the 
" Courtship of Emer " 
— cethar-builcc, ' four 
bags ' (of which the 
ceihar-bolcc of the 
above extract is the 
genitive plural). This 
was probably con- 
structed and worked 
in something like the following manner as partly shown 
in the plan : it is indeed impossible to conceive how 
four chambers could otherwise be brought into play. The 
fire was blown from the two chambers bb by pressure 
of heavy weights like the present smith's bellows. The 
two chambers aa were worked by the bellows-blower, 
who stood on them with his face towards the fire. When 
one of the chambers aa was pressed down, it filled the 
chamber b in front of it through a pipe communicating 
with an opening in the bottom board of b (with clapper- 
valve) : and the other chamber B was similarly filled in 
its turn. The function of the chambers aa was to keep 

Fig. 297. 

Cordectural plan of quadruple or four-chambered force- 
bellows. The two pipes through which the chambers AA 
kept the chambers BB 611ed are shown : also the four clapper- 
valves (by little dotted circles) In the four under boards. 


the two bb filled : the function of bb was to blow the 
fire. This cethar-builcc, or four-chambered bellows, gave 
a more uniform blast than the two-chambered one. But 
it was much harder to work : and this doubtless was the 
reason why old Domnall selected it for the heroes, in 
order to make sure that, as they needed hard exercise, 
they should have enough of it. 

A bellows — no doubt a very large one, or more 
probably several worked together — was also used in 
smelting, as we know from the following comparison in 
Cormac's Glossary, a part of his description of the Spirit 
of Poetry disguised as a monstrous giant : — " Like the 
" blowing of a bellows smelting ore (oc berbad mianaig), 
" was the drawing in and the puffing forth of his breath : 
" sledge-hammers would not strike from a glowing mass 
" [of iron] such a shower of fire as his lips struck forth." 
The comparison of the hard breathing of champions or 
animals fighting, to the blowing of a smith's bellows, 
is very common : — Rabulgsetar a n-6li ocus a srona 
mar bulgu goband i certchai : " their cheeks and their 
noses puffed out like the bellows of a smith in a 

It may be as well to finish this subject here with what 
there is to say about the domestic bellows. This was 
totally different in make and mode of using from the 
forge-bellows, as well as from our present common kitchen- 
bellows. The Senchus Morf mentions a bellows among 
the domestic utensils of a chief's house ; but the name 
used is not builcc but trefet, i.e. ' blower.' The gloss upon 
this (Br. LL., I. 145) explains it thus — bringing in two other 
names : — " Trefet of a chief's house means the teite, namely, 
" that which keeps turning round, and through it the fire is 
" blown through the leather : or [as another derivation] 
" it [trefet] refers to the strong fet or pipe through which 
" the fire is blown in each chief's house ; i.e. the seitiri 

* LL, 104, a, ,, I Br. Laws, i. 126, ? . 127, 7 . 


" or blower." So far the gloss. Seitiri, as we know, is 
derived from sett, to blow : and the idea put forward in 
this alternative derivation in the gloss is that trefet is 
shortened from tre-feit, which means ' through the pipe,' 
from tre, through, and fet or fead, a pipe. From this 
description we must conclude that the bellows used in 
private houses was one of those made to blow by revolving 
fans inside. This is further indicated by the fact that in 
the law-tract it is not called builg (' bags '), though three 
other names are applied to it — trefet, teite, and siitiri. 
This form of bellows is still occasionally met with, but 
the body is now made of lacquered tin instead of wood 
and leather. Moreover, among the English-speaking Irish 
people it is not called a " bellows " but a " blower," which 
is the exact equivalent of the old Irish term seitire, or in 
its modern form seidire [shaidera], and which is indeed 
the very term used by O'Donovan in his translation of 
the Senchus Mor (Br. Laws, i. 127, 7 ). 

5. Carpenters, Masons, and other Craftsmen. 

Carpenters. — We have seen how carefully handicrafts 
were classified by the ancient Irish, as set forth at page 293, 
supra. Some of these were sufficiently important and 
engrossing to give exclusive employment to separate 
tradesmen : but it is probable that in case of others the 
same craftsman worked at two or more of them as occasion 
arose. Woodworkers of whatever kind do not figure near 
so prominently in the ancient literature as smiths and 
braziers : yet they must have been more numerous, for 
there was more work to be done in wood than in metals. 
One important source of employment for carpenters was 
the building of houses, which in old times were nearly 
always of wood. A carpenter who devoted himself to 
house-building was called ailtire, from an old word alt, 
meaning a house : and this branch of the business was 


called ailtirecht* Accordingly O'Davorenf defines an 
ailtire as saor denma tighi, ' a house-building carpenter.' 

It has been already stated that the yew-tree was 
formerly very abundant. Its wood was highly valued 
and used in making a great variety of articles : so that 
working in yew was regarded as one of the most important 
of trades. It required great skill and much training and 
practice : for yew is about the hardest and most difficult 
to work of all our native timber : and the cutting tools 
must have been particularly fine in quality. Yew-work 
was called ibraracht, modern form iubhraracht [yooraraght] , 
from ibar or iubhar, the yew-tree. Various domestic 
vessels were made from it (p. 69, supra), and it was used 
for doorposts and lintels and other prominent parts of 
houses, as well as for the posts, bars, and legs of beds 
and couches, always carved. In the most ancient of the 
tales we often find mention of houses ornamented with 
" carvings (aurscartad) of red yew " J : and even so late as 
the first half of the thirteenth century this custom is 
recorded in the following words written by the Ulster poet 
Mac Conmidhe in his poetical description of the cathedral 
of Armagh : — " Upon the arches of this white- walled church 
are clusters of rosy grapes carved from ancient yew."§ 
So high was the estimation in which these ornamental 
carvings in yew were held that the Brehon Law has a 
special provision for their protection, prescribing fines for 
scratching or otherwise disfiguring the posts or lintels of 
doors, the heads or posts of beds, or the ornamental parts 
of any other furniture. || It is probable that bows for 
archers were made of yew as well as of other wood ; but I 
have not met with any passage mentioning this. 

Among other tradesmen, there were the dualaidhe 
[doolee] or painter (from dual, a brush) ; the rinnaidhe 

* Br. Laws, v. 106, 7i and note ; and 107, 9> I4 . 

f Three Ir. Gloss., 54 : Br. Laws, vi., Glossary, " Ailtire." 

t O'Curry, Man. & Cust., II. 57. § Ibid., 58. || Ibid , 57. 




[rinnee] or metal engraver (from rinn, a sharp point, a 
sharp-pointed instrument) ; and the erscoraidhe [erscoree] 
or wood-carver.* Carvers were in much request and 
exercised their art in the highest perfection — as we have 
seen — on yew-wood. 

Various Tools. — Besides other tools mentioned else- 
where in connexion with certain special arts and crafts, 
the following, chiefly used by wood-workers, may be dealt 
with here. They are often noticed in Irish literature, but 
more frequently in the Brehon Laws 
than elsewhere. The 
old Irish wood- and 
metal - workers seem 
indeed to have used 
quite as many tools 
as those of the pre- 
sent day. 

There were two 
names for a saw, 
turesc and rodhb 
[rove], of which turesc 
is still used. Some- 
times it was called serr; but this term was more commonly 
applied to a scythe or a sickle : the point of resemblance 
between saw and sickle being the teeth on the edge. 
Sawing (with a rodhb) is mentioned in the Agallamh 
(p. hi) as a specially noisy work: and the derivation 
of turesc in Cormac's Glossary (p. 161) makes plain the 
mode of working : from tain's, across, because — says the 
Glossary — " it cuts everything across." In the crannoge 
of Cloonfinlough in Roscommon were found deer- horns 
neatly sawn in preparation for further manufacture. 

There were — as at the present day — several kinds of 
axes and hatchets variously shaped, and used in different 
sorts of work, as may be seen by the number of names 

* O'Cuny, Man & Cust., n. 209, 21Q, 

Fig- 298. Fig. 299. 

Stone hatchets, of very hard close-grained stone. Fig. 298 
is very large and heavy, being 8^ inches long. Fig. 299 is 
S% inches long, beautifully made, and highly polished. 
<From Wilde's Catalogue, p. 80.) 


for them, and the manner in which they are often dis- 
tinguished. The common hatchet used in the workshop 
was called tuagh [tooa], which seems to be a general name 
for a hatchet or axe of any kind : it was applied not only 
to the hatchet used by tradesmen, but also to a battle-axe. 
In all forms of axe, the metallic head was fixed on the 
handle, the same as now, by wedging the wood through 
the era or opening in the iron or bronze. The head, too, 
of the carpenter's axe, if not securely fixed, was liable to 
fly off ; and if this occurred through carelessness, the Law 
(in. 175) laid down a rule regarding compensation when a 
bystander was injured. Great numbers of bronze axes 
are preserved in the National Museum, Dublin. The 
carpenter's hatchet was probably like some of those 
figured in vol. I., p. 119, supra. Two primitive stone 
hatchets belonging to prehistoric times, are shown on last 
page : the originals are in the National Museum, Dublin. 
The Crith Gabhlach, in enumerating the various articles 
that a brewy or keeper of a house of public hospitality 
should have always ready, mentions three kinds of axes :— . 
a fidchrann [feecran], a fidba [feeva], and a Mail [beeal].* 
The Mail was used in felling and clearing wood. Bishop 
Olcan, we read, went looking for a place in which to settle, 
with his " Mail on his shoulder "f ; of course to clear a 
space from trees and bushes. Fid, the first syllable in 
both fidchrann and fidba, means wood, and chrann or 
crann in the former means a tree or a wooden handle. 
A fidba, or, as it is sometimes called, fodb [fove], was 
something like our bill-hook : we find it mentioned in 
the Crith Gabhlach as used in making wooden fences. J 
Again in the Book of Aicill, a decayed king is quaintly 
said to retain only " the kingship of the three handles, the 
" handle of his flail, the handle of his Mail, and the handle 

* Br. Laws, iv., 310, I2> I3 . t Tri P- L i fe , 136, 30 . 

J Br. Laws, iv. 315, I9i 20 . See also Senchus Mor, in vol. 1. 124, I4 _ IS . 
and Gloss, 141, top lines. 


" of his fidba." In the Dinnsenchus* it is said that a man 
named Raigne cut down a wood with his bacc and his 
spade, showing that the bacc was a felling axe. It was 
something like the present hedge-cutter's bill-hook. The 
word b/rc or bacc means a ' bend ' (in this case a hooked 
blade fixed on a handle) ; and the gloss on the Brehon 
Law mentions a bac as used for cutting ivy.f 

A tdl [tawl] or adze — i.e. an axe having the edge 
across or at right angles to the line of the handle — was 
used for special sorts of work ; as, for instance,. in making 

wooden shields ; and of course, 
in cooperage. It was an exceed- 
ingly common tool, as it is con- 
stantly mentioned in all sorts ot 
records. More than one histo- 
rical personage had the epithet 
Mac Tdil (' son of an adze ') 
affixed to his name, to denote 
F|G - 3 °°- that his father was one oi 

Bronze adze: in National Museum: 
4H inches wide along the ed K e. (From thOSe WOOd - WOrkerS who USed 

Wilde's Catalogue, p. 523.) 

the adze. 
An awl, by whatsoever tradesman used, was called 
menad or meanadh [manna], which is still the Irish word 
all through Ireland : but in Munster it takes the form 
meanatha [mannaha]. The fanciful derivation of menad, 
given in Cormac's Glossary (p. 108), is very suggestive : — 
from mlu [meen], small, and dith [a], sharp, as if the 
word was contracted from min-dith, ' small-sharp.' The 
old Irish carpenters used an auger and called it tardthar 
[tarawher], a name which is still in use. In Cormac's 
Glossary the word is * fancifully analysed as if contracted 
from dair-uath-air [dar-oo-ar], meaning ' the oak hates it 
(dair, oak : tiath, hatred) : " because," as the Glossary 
adds — " of its cuttingness, for it cuts through the oak.' 

* Folklore, in. 480. f Br. Laws, v. 488, n ; 489, 25. 


The Crith Gabhlach, enumerating the tools that ought 
to be in the house of every brewy, includes a tarathar. 

Irish carpenters and others used compasses which they 
named gabulrind [gowlrin], a word given in Cormac's 
Glossary* as the equivalent of the Latin circinus (a 'pair 
of compasses '). The Irish term is quite descriptive, being 
compounded of the two words, gabal, a fork, and rind or 
rinn, a point : that is to say a fork with two points. 
Among the pagan relics found under 
a earn at Loughcrew are many combs 
engraved with circles by a compass, 
and also a bit of iron having all the 
appearance of being the leg of a 
compass.f The large circles on 
some of the flat golden gorgets 
(p. 234, supra) were obviously 
made with a compass: all going 
to confirm the truthfulness of 
the records. 

Fig. 301. Fig. 302. Fig. 303. Fig. 304 • 

Figs. 301, 302, ahd 303 are small primitive stone hammers. (From Wilde's Catalogue, p. 79.) 
Fig. 304 is a bronze hammer, found in Sligo : 6 inches long : well worn from work. (From Kilk. 
ArchEeol. Journ. for 1885-6, p. 538.) 

The mallet used by carpenters, fence-makers, and other 
workmen, was generally called farcha or forcha. In the 
year 512 (FM, A.D. 503), Lewy, king of Ireland, was killed 
by lightning at a place thence called Achad-farcha, a name 

* Corm. Gloss., p. 30: Irish text in Three Ir. Gloss., p. 9, 
t Fergusson, Rude Stone Mornurcjits. p. 218. 


which is commonly translated the ' Field of lightning,' but 
which primarily means the ' Field of the mallet.' Indeed 
the very words used by the Four Masters and other 
annalists are (in English) " having been struck by a mallet 
of lightning " (forcha-teinntighe) : the idea being the same 
as that of the Scandinavians, who armed their god Thor 
with a lightning hammer. A sledge was called ord : an 
ordinary hammer was lamh-ord (' hand-sledge ') : but some- 
times cas-ord, now generally made casur [cossoor]. The 
cas in this, which means ' twisted ' or ' bended,' probably 
refers to the " claw," so that a casord or casur would be a 
' claw-hammer.' This is in some measure borne out by 
the fact that the word mailin was used to designate 
another kind of hammer, no doubt one without a claw : 
for mailin means bald or bare : a " bare or clawless little 

Carpenters used a rungenn or runcan, a plane : a 
slightly different form of the name is found in the Brehon 
Law, where it is stated that the posts of the doors and 
beds of certain classes of houses were finished off with 
a rungcin [rungkeen], which O'Curry understood as a 
moulding-plane.* In the Story of Tain bo Fraich (138), 
139, bot.) in the Book of Leinster, one of the houses of the 
palace of Cruachan is described as having decorations of 
red yew " with variegated planing " (Jo m-brecht-runcain). 

Workers in wood used a sort of press called cantair, 
either for straightening wood or forcing it into certain 
shapes — after being softened probably by water or steam, f 
In Stokes's Irish Glosses on Latin Declension cantair is 
the word used to explain the Latin troclia : and O'Reilly 
gives cantaoir as a name for any sort of press. The ancient 
Irish builders used a crane of some kind for lifting heavy 
articles, as is proved by the following sentence in Cormac's 
Glossary describing a very repulsive-looking giant :— 

* Man. & Cust., II. 29, bottom. 

I O'Curry, in Stokes's Irish Glosses on Lat. Decl., p. 60, No. 239. 


" cuirre ina corr aurocbala a dhd gruad, " rounder than a 
lifting-crane his two cheeks."* Here the Irish word cor- 
responding to " crane " is corr, which is still the name of any 
bird of the crane kind : and it is applied in this passage 
to the machine, exactly like the English word crane, on 
account of the long beak. The comparison of the giant's 
cheeks to the lifting-crane refers to the rounded or bulging 
shape of the body of the machine. 

The lathe and other turning-wheels were well known 
and employed for a variety of purposes. The Brehon 
Law (v. 107) when setting forth the privileges of various 
classes of craftsmen has tornoire or turners among them, 
explaining that these are the men " who do tornaireckt or 
turning." A much older authority, an eighth-century 
Irish glossator, in his remarks on Ps. II. 9, explains a 
potter's wheel as " a round wheel {roth cruind) on which 
the cerda or potters make the vessels. "f Mr. Johnson, in 
his observations on ancient Irish gold-work, states that he 
found the cups of one gold fibula marked with three eon- 
centric circles so true as to " have all the appearance of 
being done on a lathe."J Once the lathe was known it 
would of course be used on wood : and in the crannoge 
of Cloonfinlough in Roscommon were found, among many 
workshop remains, a quantity of shavings exactly such as 
would be left by a turner. § The Irish word for a lathe 
is deil [dell], which is used by Keating|| ; and at the pre- 
sent day, speakers, whether using the Irish or English 
language, call a lathe a dell. But I have not found the 
word in any very old documents. 

Chisels of a variety of shapes and sizes were used by 
wood-workers : of which the following illustrations will 
give a very good idea : the originals — which are all of 

* Corm. Gloss., 135, bottom : Irish text in Three Ir. Gloss., p. 36, ^ 
t Stokes and Strachan, Thesaur., 1. 23 : see also p. 79, supra. 
X Proc. Roy. Acad., 1893-6, p. 782. § Ibid., vol. v. p. 211. 

|| See "Deil" in Glossary of Atkinson's Three Shafts. 




bronze — are preserved in the National Museum. It has 
been suggested by Sir John Lubbock that many of the 
smaller and thinner bronze celts were used as chisels. 
The Four Masters use the word fonsura for a chisel.* A 
large number of bronze gouges are preserved in the 
National Museum ; but I have not found any special 
Irish name for a gouge. Among the collection of bronzr 
tools found at Dooros-Heath in King's County (p. 320, 
infra) are three gouges with the regularly curved edges, 
well adapted for excavating and paring wooden bowls and 

Fig. 305. 

Fio. 306. 

Fir.. 307 

Fig. 309. 

Figures 305, 306, 307, anil 308. bronze chisels : figure 309, a bronze gouge. All in National Museum. 
(From Wilde's Catalogue, p. 521 J 

gobletsf : and about the same time another was found in 
Wexford.J The bronze of these and of all the other cutting 
instruments in the King's County collection is excessively 
hard. It may be observed that bronze can be made almost 
or altogether as hard as steel by hammering. 

Sharpening. — For sharpening edged tools and weapons, 
the people used a whetstone, which is called in Cormac's 
Glossary (p. \2)cotud, literally meaning ' hard,' and defined 
"a//£ or stone on which iron tools or weapons are ground": 
but it is often called lee, which is the general name for a 

•O'Donovan, Suppl., 647. t Pioc. Roy. Ir. Acad., IV. 240. 

J Proc. Roy. Ir. Acad., iv. 369. 


flat stone, just as we now sometimes call a whetstone 
" a stone " for shortness. The whetstone is very often 
mentioned in the Brehon Laws (as in V. 485, line 7 from 
bottom). But they had also a circular grindstone which 
was turned on an axis like those now in use. The 
grindstone was called liom-broii [leev-vrone], ' sharpening 
millstone,' and also lic-limad [lic-leeva], 'stone of 
grinding' — corresponding exactly with the English name 
" grinding- stone " : and it was turned round by means of 
a cranked handle. The crank was called ruiti\ which 

Fig. 310 

Specimen of dry or mortarless masonry : portion of the wait of Caller- 
more, near Kilnaboy, in Clare. The stones are in their natural state — 
unhaminered. (From Mr. Westropp's Article on Prehistoric Stone Forts 
of Northern Clare, Kilk. Archaeol. Journ. for 1896, p. 367. To illustrate 
observations at p. 323, infra, 

is defined in connexion with the grinding-stone in a gloss 
on a sentence of the Book of Aicill (Br. Laws, III. 295) 
in a manner that leaves no room for doubt as to what 
it was : — " Ruitech, i.e. the thing which runs well from 
"him and to him [i.e. from and to the person turning it], 
" namely the crooked stick." A grindstone was one of 
the numerous articles which a brewy was bound to have 
in his house (Br. Laws, IV. 311). Mr. Wakeman records 
that in 1872 some whetstones and two circular grindstones 
were found in a crannoge in Fermanagh, the larger one 
eight inches across.* 

* Kilk. ArclirCol. Journ , 1872-3, p. 320.. 




Remains of Ancient Workshops. — It is worthy of remark 
that the remains of ancient workshops or factories belong- 
ing to several trades have been discovered from time to 
time in different parts of Ireland. About the year 1820 a 
brazier's workshop was turned up in a place called Dooros- 
Heath, in the parish of Eglish near Birr in King's County* 
where great quantities of gold-coloured bronze articles 


Fig. 311. 

Clochan-na-Carraige, the ' Stone house of the Rock,' on the Great Island of Aran, Galway Bay. 
Pagan circular stone house : round outside : inside it is quadrangular, and 19 feet long by 7^ feet 
broad, and 8 feet high. Walls of dry masonry, converging by overlapping till closed at top by a 
single stone. Two apertures in roof served for windows and chimney. (From Petrie's Round 
Towers, p. 130. To illustrate what is said at p. 323, infra). For a Christian house of similar 
construction, see p. 323, infra. 

were found — bells, spearheads, celts, trumpets, gouges, and 
soforth : also whetstones, flat, convex, and concave. That 
this was a workshop is shown by the fact that many of the 
articles were unfinished or only half made, while some 
were mended : and there was one lump of unworked 
bronze — mere material * The remains of a glass factory 

* See Mr. Thomas Cooke's intelligent article on this find: Proc. Roy 
Ir. Acad., iv. 423 : see also same vol., p. 239. 


will be found mentioned at p. 33, supra ; and an old work- 
shop of a family of goldsmiths near Cullen in Tipperary 
is described in vol. 1., p. 556, supra. In parts of Ulster 
where flints are common, flint workshops are sometimes 
turned up, with vast numbers of finished and half-finished 
flint articles.* Ancient Gaulish workshops of various 
crafts have in like manner been lately found in France. f 

Masons and their Work. — A knowledge of the use of 
lime-mortar and of the arch was introduced by St. Patrick 
and his foreign missionaries. Before his time the Irish 
built their stone structures of dry masonry : and not 
knowing how to construct an arch they brought their walls 
to converge in a curve — like the ancient Greeks and 
other nations of antiquity — by the gradual overlapping 
of the flat-lying stones. Numerous specimens of their 
handiwork in this department of ancient art still remain, 
especially in the south and west, in the beehive-shaped 
houses and stone cahers, which show much skill in fitting 
the stones to one another so as to form very close joints. 
Even after the introduction of Christianity the old pagan 
fashion of building was retained in the erection of many 
of the ecclesiastical structures : and stone oratories belong- 
ing to those primitive ages are still to be seen in various 
parts of the country, built without mortar and converging 
upwards by the overlapping of the stones. The outer wall 
of the cashel enclosing the little hermit-monastic establish- 
ment on Inishmurray is of dry masonry and in all respects 
like the pagan cashels. (See fig. 313 farther on.) 

Although the Irish did not employ lime (Irish ael) in 
making mortar till the fifth century, it was used as a 
whitener in pagan times (p. 63, supra). They made lime 
by burning limestone or sea-shells in a lime-kiln much 
as is done at the present day : but I find no notice 
of a kiln for this purpose till far into Christian times — 

* Kilk. Archaeol. Journ., 1883-4, p. 120. 

f De Jubainville, La Civil, des Celtes, p. 130. 




yet still before the Anglo-Norman invasion- viz. in 
1 145, wnen " a lime-kiln, which was sixty feet every way, 
" was erected opposite Emain Macha by Gillamacliag, suc- 
" cessor of Patrick, and Patrick's clergy in general."* The 

Fig. 312. 

Round Tower of Dcvenish Island, in Lough Emc : 8s feet high. To illustrate what 
is said next page as to beauty of outline and general shape. (From Petrie's Round 
Towers, p. 360.) Round towers are figured also in chap, x., vol. I., supra. 

Annals record the erection in 1163 of another lime-kiln — 
which they call tene-aoil (literally " fire of lime " — pron. 
tenneel), seventy feet square, by the abbot of the Columban 
monastery of Derry — Flaherty O'Brolchain — and his clergy- 
But the erection of these great structures indicates long- 

* Reeves, Churches of Armagh, p. 38. 


existing previous knowledge of lime-kilns and of the art 
of constructing them. There can be no doubt that sea- 
shells were used for making lime in the old times : this 
was Petrie's opinion* : and we know that they were turned 
to this use in the time of Elizabeth : Docwra, in his 
" Narration," says : — " Cockleshells to make a lyme wee 
"discovered infinite plenty of, in a little island in the 
" mouth of the harbour [at Derry] as wee came in." 

Fig. 313. 

Stone house on Church Island, Valentia, Kerry. Example of a Christian oratory, built of 
uncemented stones, with walls converging-, after the old pagan fashion (see figure 311, supra). 
Interior dimensions, about 19 feet by n feet. Near it, on the same little island, is a circular pagan 
clochan, or stone house. (From Journ. Antiqq. Irel. for 1900, pp. 152, 155.) To illustrate observations 
at p. 321, supra. 

Numerous structures erected in Christian times, but 
before the invasion, with lime-mortar, still remain all over 
the country, chiefly primitive churches and round towers. 
It is only necessary to point to the round towers to show 
the admirable skill and the delicate perception of grace- 
fulness of outline possessed by the ancient Irish builders. 
A similar remark might be made regarding many of the 
ancient churches, especially those called Romanesque, for 
which that part of Petrie's Round Towers relating to 
churches may be consulted. 

* Stokes's Life of Petrie, p. i6j. 


Blessing the Work. — In old times it was a custom for 
workmen, on completing any work and delivering it over 
finished, to give it their blessing. This blessing was called 
abarta, " and if it was omitted, the workman was subject 
" to a fine, to be deducted from his hire, equal to a seventh 
" part of [the cost of] his feeding." These are O'Donovan's 
words, which are merely an expansion of the explanation 
of abarta, given in Cormac's Glossary (p. 9). The same 
rule is laid down in the Senchus Mor, from which the 
explanation in the Glossary was borrowed (see Br. Laws, 
1. 133). This custom is mentioned in the " Small Primer," 
where bendacht (' benediction ') is used instead of abarta* 
It would appear also that the first person who saw the 
work after it was finished was bound to give it a blessing 
on pain of fine : and it was specially incumbent on women 
to bless the work of other women. The custom has 
descended to this very day : for the peasantry on coming 
up to people engaged in work of any kind always say 
" God bless your work," or its equivalent in Irish, go 
m-beannuighe Dia air bhur n-obair. 

6. Protection of Crafts and Social Position of 

Artificers of all kinds held a good position in society 
and were taken care of by the Brehon Law. Among the 
higher classes of craftsmen a builder of an oratory or of 
ships was on the same level — in respect to honour-price 
and dire-fine — with an aire-desa, the lowest rank of noble : 
that is to say he was entitled to the same compensation 
for any injury inflicted on him in person, honour, or 
reputation. In like manner a chariot-maker and a wooden- 
house-builder, and some others, ranked with the tanist, 
or intended successor to a bo-aire chief. And similar 
provisions are set forth in the law for craftsmen of a 

• Br. Laws, v. 98, „ , 99, ,,. 


lower grade.* Elsewhere it is stated that the artist who 
made the articles of adornment of precious metals for the 
person or household of a king was entitled to compensation 
for injury to person or property equal to half the amount 
payable to the king himself for a like injury.f But the 
most striking illustration of the estimation in which handi- 
craft workers — especially artists of all kinds — were held, 
occurs in the Brehon Law (v. 383), where, prescribing the 
fine for the retention or loss of an embroideress's needle, 

Fig. 314. 

Beautiful window of Castledermot Abbey. (From Miss Stokes's High Crosses of Castledetmot and 
Durrow, p. 7.) To illustrate the statements about the skill ol Irish masons, at p. 323, supra. 

the text winds up with this statement : — " For every 
"woman who is an embroideress deserves more profit 
" than even queens." These are a few examples of the 
provisions found in many parts of the Brehon Law for 
the protection of craftsmen and the recognition of their 
proper position. 

As illustrating this phase of society we sometimes find 
people of very high rank engaging in handicrafts. One of 
St. Patrick's three smiths was Fortchern, son of Laegaire, 

* Br. Laws, V. 103-109 : see also Petrie's Tara, £08, note 8. 
t O'Curry, Man. & Cust., ir. 205. 




king of Ireland. Beoan, the father of St. Mochoemoc, 
and another Beoan, father of St. Ciaran of Clonmacnoise, 
though both of royal descent, were famous carpenters* 
But, on the other hand, a king was never allowed to 

Fig. 3x5. 

Doorway of Kalian Church, King's County : dating from about the middle of the 
eighth century. Specimen of skilled mason-work to illustrate what is said at p. 323, 
supra. (From Petrie's Round Towers, p. 246.) 

engage in manual labour of any kind (vol. I., p. 60, supra). 
Many of the ancient Irish Saints were skilled artists. In 
the time of St. Brigit there was a noted school of metal- 
workers near her convent, over which presided St. Conleth, 
first bishop of Kildare, who was himself a most skilful 

* Cambr. E versus, II. 173. 


artist.* St. Daig or Dega of Iniskeen in Louth was a 
famous artificer. He was chief artist to St. Ciaran of 
Saigir, sixth century, and he was a man of many parts, 
being a cdird or brazier, a goba or smith, and besides, a 
choice scribe. In the Martyrology of Donegal it is stated 
that " he made 150 bells, 150 crosiers : and also [leather] 
cases or covers for sixty Gospel Books," i.e. books con- 
taining the Four Gospels. Elsewhere we find it seated 
that he made " covers or cases for books, some plain, but 
others covered with gold, silver, and precious gems."f 

In the muinnter, or familia, or household of St. Patrick, 
there were several artists, all of them ecclesiastics, who made 
church furniture for him. " His three smiths (gabaind) 
expert at shaping," were Macecht, who made Patrick's 
famous bell called Finn-faidhech or ' sweet-sounding ' ; 
Laebhan ; and Fortchern, who was son of King Laegaire. 
His three brasiers (cerda ; or, as Evinus calls them in Latin, 
tres fabri cerarii, ' three copper-smiths ') were Aesbuite, 
Tairill, and Tasach. In the Tripartite Life it is stated 
that " the holy bishop Assicus {i.e. Aesbuite) was Patrick's 
" coppersmith {faber ereus), and he made altars and quad- 
" rangular tables, and quadrangular book-covers (leber- 
" chometa : lit. * book-preservers ') in honour of Patrick."! 
We have already seen how highly scribes and book-illumi- 
nators were held in esteem. It is to be observed that 
nearly all the artists selected by St. Patrick for his house- 
hold were natives, though there were many foreigners in 
his train, some of whom he appointed to other functions : 
a confirmation of what has been already observed, that he 
found, on his arrival, arts and crafts in an advanced stage 
of cultivation. 

In common with most civilised people of old time the 
Irish attempted to fix by law the wages of workmen and 

* Todd, St. Patrick, 26. 

f Stokes, F6ilire, 131 : O'Cl. Cal., 223 : Adamnan, 115, note c : Petrie, 
Round TowttS, 202. { Trip. Life, 97 ; FM, a.d. 448. 



[PART itl 

artists : . the rates are laid down in the law, but, as might 
be expected, they vary a little as given in different 
authorities. The Senchus M6r states that for making a 
lann oir, i.e. a gold head-band or necklet of gold, the 

Fig. 316. 

West front of stone-roofed church at Killaloe, the burial-place of Murkertagh O'Brien, 
king of Munster, and afterwards king of Ireland (died A.D. 1119). An example of skilled 
mason-work. See observations at p. 323, supra. (From Pctrie's Round Towers, p. 278.) 

caird or artist should be paid in silver one-ninth of the 
value of the finished article* : and for making a gold ring 
on:-twelfth of its value in silver. A legal commentator, 
quoted, by 0'Curry,f says : — " The law tells us that the 

* Br. Laws, n. 415. 

f Man. & Cust., II. 205. 


" weight of the lann-oir in silver was paid to the cerd for 
" making it " : one of the many proofs — if proofs were 
needed — that these articles were made by native artists. 
In another part of the Senchus Mor (Br. Laws, I. 133) it is 
stated that the price for making any article is one-tenth 
of its value with food and drink. The Glossator of the 
Heptads has this remark : — " The payment of all handi- 
" craft, namely, the tenth of everything made, is the price 
" of making " (Br. Laws, v. 215) : and in this part of the 
Laws there are many statements to the same effect. The 
rule of the tenth was very general, and it was observed in 
many crafts down to recent times. A little more than a 
century ago the bakers of Dingle in Kerry charged one- 
tenth of the value of bread for baking. 

No individual tradesman was permitted to practise till 
his work had been in the first place examined at a meeting 
of chiefs and specially-qualified ollaves, held either at 
Croghan or at Emain, where a number of craftsmen candi- 
dates always presented themselves. But besides this there 
was another precautionary regulation. In each district 
there was a head-craftsman of each trade, designated sai- 
re-cerd [see-re-caird], i.e. ' sage in handcraft.' He presided 
over all those of his own craft in the district : and a work- 
man who had passed the test of the examiners at Croghan 
or Emain had further to obtain the approval and sanction 
of his own head-craftsman before he was permitted to 
follow his trade in the district.* It will be seen from all 
this that precautions were adopted to secure competency 
in handicrafts similar to those now adopted in the pro- 

Young persons learned trades by apprenticeship, and 
commonly resided during the term in the houses of their 
masters. They generalfy gave a fee : but sometimes they 
were taught free — or as the law-tract expresses it — " for 

* Keating, 419, from old authorities. 


God's sake." When an apprentice paid a fee, the master 
was responsible for his misdeeds : otherwise not. The 
apprentice was bound to do all sorts of menial work — 
digging, reaping, feeding pigs, &c. — for his master, during 



Section i. History. 

very early Irish tradition, transmitted through 
ancient manuscripts, assigns the erection 
of the first watermill in Ireland to the 
illustrious King Cormac mac Art (reigned 
a.d. 254 to 277). The story is that he 
had a beautiful cumal or bondmaid whose business 
it was to grind corn with a quern. In order to relieve 
her from this heavy drudgery, the king sent across 
" the sea " for a saer-muilinn — a ' mill-wright ' — who 
constructed a mill on the stream of Nith, flowing 
from a well named Nemnach (' sparkling ') beside Tara. 
This account is given by Cuan O'Lochain, chief poet of 
Ireland, who died a.d. 1024, in a poem on Tara preserved 
in several of our ancient manuscripts. It is given in 
Petrie's Tara, p. 143 : the passage relating to the mill 
will be found at p. 147. O'Lochain's poem was copied 
into the present existing manuscripts from much older 
books. The well Nemnach still exists, though not now 
known by its old name : it was identified more than 
sixty years ago by Petrie and O'Donovan (see Plan of 

* Br. Laws, iv. 237, and note 1. 


Tara, p. 81, supra). It is remarkable that this ancient 
written record is corroborated by a vivid oral local 
tradition of the present day, which gives some details 
not in the written account, one of which is that King 
Cormac obtained the mill-wright from the king of Scot- 
land. But here the modern tradition is probably wrong, 
as it appears that watermills had not been introduced 
into Britain by the Romans so early as the third century. 
According to the same oral tradition the name of the 
imported mill-wright was Mac Lama. It is an interes'ing 
fact that there has been a mill on the spot time out of 
mind in possession of one family named Mac Lama, 
having descended from father to son ; but in modern 
times they have translated their name to Hand (Irish 
lam, a hand). It has been always called the mill of 
Lismullin (the ' fort of the mill ') : and the place, which 
is a mile north-east from Tara, retains the name Lismullin 
to this day.* 

Whatever amount of truth or fiction may be in the 
tradition of King Cormac's mill, we have ample evidence 
that f om a period soon after the advent of St. Patrick, 
watermills were in very general use all through Ireland, 
and were an important factor in daily life, both in the 
monasteries and among the people in general. Each 
muilenn or mill was managed by a skilled muilleoir 
[millore] or miller. Mills and millers are mentioned in 
the oldest Irish literature ; and monastic mills are mixed 
up with the Lives of many of the early Irish saints. In 
the Tripartite Life (p. 211, 6) there is a passage in which 
St. Patrick is made to prophesy of certain streams in the 
present County Waterford that there never would be mills 
on them. In the Lives of very many of the early saints, 
we find it recorded, among their other acts, that they built 
mills at their monasteries. Cogitosus's Life of St. Brigit, 
written in the tenth century, mentions a millstone (molaris 

* This account has been taken from Petrie's Tara, p. 162 et seq. 


lapis), showing that there was a mill in connexion with 
her nunnery in Kildare. St. Brigit died A.D. 523. The 
mill built at Fore, in the present County Westmeath, by 
St. Fechin, in the seventh century, which is recorded in 
his Life, is noticed by Giraldus Cambrensis ; a mill has 
been kept up there from the saint's time to the present 
day ; and it is still called St. Fechin's mill. There was 
a mill at the monastery of Mailman at Tallaght which 
is frequently mentioned in old Irish writings.* The annals 
record the burning of St. Ciaran's mill at Clonmacnoise, 
A.D. 954. In the Story of the Boroma in the Book of 
Leinster, we are told that certain persons who went in 
pursuit of St. Moiling (seventh century) found him at a 
place called Fornocht laying out the site of a mill.f 
Cormac's Glossary (p. 109) — written in the ninth or tenth 
century — speaks of the mill-shaft and of the millstones, 
which, it says, were larger than the stones of a quern : 
and the passage goes on to say that the mill was driven 
and the corn ground by water-power. Elsewhere (p. 41) 
in the same Glossary the motion of the great whirlpool 
of Coire-Brecain is compared to the whirling of mill- 
paddles : all showing how common mills were in his 
time. Mills and millers are also often mentioned in the 
oldest of the tales : as for instance in the " Feis Emna " 
(or " Feast of Emain "), in the " Voyage of Maildune," in 
the " Fled Bricrenn," in " Da Derga's Hostel," and in the 
" Courtship of Emer " — in so many indeed that references 
are needless : and in one passage a warrior's spear is 
compared to a mol muilind, the ' wheel-shaft of a mill." 
Many of the tales, in which mills are spoken of as objects 
very familiar, are quite pagan in character, and originated 
according to the best authorities, in the seventh or eighth 

A most interesting notice of an ancient Irish mill 
occurs in connexion with an undoubted historical event, 

* F6ilire, p. 8, bottom. f O'Grady, Silva Gad., 423. 


the death, a.d. 651, of Donogh and Conall, the two sons 
of Blathmac (one of the joint kings of Ireland — 656 to 
664), who were slain by the Leinstermen at " the mill of 
Mailoran the son of Dima Cron." This event, which 
created a great sensation at the time, is recorded in the 
Annals of Tigernach, as well as in those of Ulster, of 
the Four Masters, and of Clonmacnoise, and it is com- 
memorated in the form of a short story in an Irish MS. 
in the Bodleian Library, which has been published and 
translated by Kuno Meyer in " Hibernia Minora " : but 
the storyteller's version differs from the annalists' record 
in some important details, though non-essential for our 
purposes. On a certain occasion the princes were pursued 
by Mailoran and his party, who determined to kill them. 
They succeeded in wounding them, near the mill, on which 
the brothers took refuge among the works, beside the tnol 
or shaft : but the pursuers forced the woman who had 
charge of the sluice to let the water run, so that the mill 
was set going, and the young men were crushed to death 
in the works. A contemporary poet composed a poem 
on this event, in which he apostrophises the mill in the 
following strikingly vivid stanza : — 

" O mill, what hast thou ground ? Precious thy wheat ! 
It is not oats thou hast ground, but the offspring of Kervall 

[i.e. the princes]. 
The grain which the mill has ground is not oats but blood-red 

wheat ; 
With the scions of the great tree (Kervall, their ancestor), Mailor- 

an's mill was fed." 

Mageogheghan, in his translation of the Annals of 
Clonmacnoise, says that " Donogh and Connell were killed 
" by the Lynstermen near Mollingare, in the mill of Oran 
" [or Mailoran] called Mollenoran." This mill was situated 
on the little river that runs from Lough Owel to Lough 
Iron, near the point where the river is now crossed by a 


bridge ; and the place still retains the name of Mullenoran. 
It is curious that a mill existed there from the time of the 
death of the princes — and no one can tell how long before 
—down to the end of the eighteenth century ; and there 
are some old people still living there whose grandfathers 
saw it in full work.* 

Tigernach and other annalists record that a celebrated 
pillar-stone called Lia Ailbe, which stood at Magh Ailbe, 
now Moynalvy in Meath, fell down in the year 999 (998, 
FM) : and that from this lia, Malachi the Great, king of 
Ireland (a.d. 980-1002), made four [pairs of] millstones. 
When St. Columkille dwelt at Clonard under St. Finnen 
(d. 549), they ground their corn with a quern, which the 
students worked in turn. But it seems plain that after 
Columkille settled in Iona, he had a watermill erected. 
Adamnan speaks of the grain, of the kiln, and of grinding 
the corn : and though he does not tell us expressly what 
sort the mill was, he uses an indirect expression that points 
to a watermill. Speaking of an incident in the life of 
Columkille, he says it occurred at a spot " where a cross 
" was afterwards erected and fixed in a millstone, which 
" may be seen to this day "f [i.e. about a.d. 697 : a century 
after its erection). Innes suggests that this millstone was 
a quern. But it must have been a large and heavy mill- 
stone belonging to a watermill to give sufficient support 
to a stone cross — a conspicuous long-standing memorial. 
Add to all these early notices that a mulenn or mill is 
mentioned in the St. Gall glosses of Zeuss (p. 778, 2 o) — 
seventh or eighth century — at which time the name 
mulenn, which is used in the Irish passage copied by 
Zeuss, and which was borrowed from Latin, had become 
well naturalised in the Irish language. We may then take 
it for certain that watermills — howsoever derived — were in 

* See O'Donovan in FM, at a.d. 647. The above poem is in the FM, 
i., p. 263 : and it is also quoted by the annotator of the F6ilire, p. 88. 
t " Crux molari infixa lapidi," Adamnan, m. xxiii. (p. 231). 


use in Ireland from the earliest ages of Christianity : but 
there is as yet no sufficient evidence to prove that they 
were known in pagan times. 

2. The " Eight Parts " of a Mill. 

The Brehon Laws took careful cognisance of mills, 
descending to minute particulars, in order to determine 
how far the law of distress applied to them, as well as to 
fix the amounts of fines and compensations in case of 
accidental damage or injury to persons. In the Senchus 
Mor* there is a ver}^ interesting enumeration of the " Eight 
Parts " of a mill, viz. — i. The water : 2. The upper mill- 
stone : 3. The shaft : 4. The supporting-stone : 5. The 
shaftstone : 6. The wheel : 7. The axis : 8. The cup or 
hopper. It will be useful to make a few observations on 
all these, in accordance with the explanations given in the 
commentaries and glosses,! an d with various passages in 
other Irish writings. 

First : The en or water consisted of three parts : — 
The spring (topur or tobar) : the mill-race (tuinidhe : pron. 
tunnee), from the spring to the mill-pond : and lastly the 
mill-pond itself. We see from this that in those times the 
water for a mill was brought from the head source along a 
channel or mill-race, much the same as at present, till it 
flowed into a pond, natural or artificial, where it was 
stored till wanted, when the sluice was raised and the 
wheel set going. The mill-pond was as familiar an object 
as the mill, and we find it very often noticed, sometimes 
by the name linn (which means a pool of any kind), and 
sometimes by the special name toiden or taidhin. St. 
Moiling is mentioned as being on a certain occasion in 
his toiden, % where he often stayed, standing in the water 
merely to mortify himself. 

* Br. Laws, 1. 125. \ Ibid., 141. 

\ Sjlva Gad. : Irish text, p. 377, ,. 


According to the Brehon Law, anyone constructing a 
mill could bring the necessary supply of water through the 
intervening farms belonging to his neighbours, acquiring 
the ground needed for the mill-race by compulsory pur- 
chase, and paying the compensation fixed by law : a 
provision which anticipated by centuries the modern 
statutes by which persons are compelled to sell any 
portion of their lands required for certain public works, 
such as railways. " Every co-tenant " — says the law- 
tract on the ' Right of Water ' — " is bound to permit the 
" other co-tenants to conduct the water [required for a 
" mill] across his land "* : " and this " — says the gloss on 
the text — " is the second instance in the Berla F&ne 
" speech where the Law commands a person to sell his 
" land though he should not like to do so."f But certain 
lands — as the tract goes on to say — were exempted from 
compulsory purchase, such as a fair-green, the land 
belonging to a church, the land round a king's residence : 
water could not be brought through these under any 
circumstances. In some exceptional cases, where the 
passage of the water would benefit instead of injuring the 
owner, the land had to be given without compensation. 
The owner of the land, when compelled to sell, might take 
direct payment, or he might choose, as compensation, to 
have a share in the mill — i.e. the use of it for one or more 
of the rotation days (p. 345 below). 

Second : the upper millstone, which is called liae and 
clock in the law text : but the general name for a millstone 
was, and is still, bro, gen. brdn [brone], or cloch-mhuilinn. 

Third : the mol [mull] or shaft ; that is, the shaft or 
axis of the mill-wheel. Mol is still the living word for a 

Fourth : the supporting stone, or lower millstone ; 
called indeoin or inneoin [innone]. 

* Br. Laws, iv. 213. f Ibid., 215. 


Fifth : the herinthiu or shaftstone, which is described 
in the gloss as the little stone which is under the head of 
the mol or shaft, and on which the mol turns. 

Sixth : the paddle-wheel or mill-wheel, which is called 
oircel [urkel]. The gloss, in explaining this word, says : — 
" over its eel or paddle the water flows." Here the whole 
wheel is called oircel, and the single paddle eel : but in a 
passage in the " Fled Bricrenn " (p. 67) the paddle is called 
oircel. In this last passage Queen Maive, speaking of 
Cuculainn's impending attack on her army, says " he will 
" grind us to mould and gravel . . . like as a mill of ten 
" paddles grinds very hard malt." This is instructive as 
giving us an idea of the number of paddles, and as inti- 
mating that a mill-wheel with ten paddles was considered 
a moderately powerful one. The present name for a mill- 
wheel is roth [ruh], which properly signifies any wheel.* 

The seventh part was the axis [of the revolving 
millstone] which is called milaire [millere]. This is 
explained in the gloss, " the burden of the mol or shaft, 
i.e. the gamul." Milaire is now the usual word for the 
pivot on which the millstone turns. 

The eighth " part " was the cup or hopper, so called 
from the Irish verb cup, to drop ; " because it cups or drops 
" the corn out of itself into the upper millstone, i.e. the 
" tual, i.e. the perforated iron." 

From the above description (especially the expression 
" over its eel or paddle the water flows ") we see that the 
water-wheel here under consideration was an overshot one, 
and revolved round a horizontal mol, shaft, or axis ; that 
the millstones lay flat ; that the upper or revolving one 
moved on a perpendicular milaire or axis : and that the 
motion of the shaft was communicated to this axis by 

The writer of the Battle of Moyrath (p. 257), describing 
two mighty heroes grasping each other and whirling 
* Corm. Gloss., p. 143 



rapidly round in mortal struggle, says : — " They might be 
compared to the huge wheel of a mill at rapid grinding." 
From this and many other such passages in the tales, as 
well as from the manner in which mills are often mentioned 
in the Senchus M6r and Cormac's Glossary, and especially 
from the story of the destruction of the two princes in the 
works of Mailoran's mill, it may be inferred that some at 
least of the old Irish mills were fairly large and powerful. 
The law-tract (on the " Right of Water " : Brehon Laws, 
iv., p. 219) affords an idea of the cost of what may be 
considered as an average-sized mill, which is set down as 
a cumcd, i.e. three cows, equal to £40 or £45 of our money. 
This is the expense of the mill alone, and does not include 
the cost of the building. 

3. Small Mills. 

But a small light mill of much simpler construction 
was also in use, portions of which are represented in 
fig. 317. In this little mill the shaft stood vertically, 
and the wheel horizontally at the lower end of it.* The 
pivot or gudgeon at the bottom of the shaft worked in 
a hole in stone or iron, fixed firmly beneath. The two 
little millstones — which were not larger than querns — were 
placed horizontally on the top of the shaft, of which the 
lower one was fixed moveless, by means of the surrounding 
frame, and the axle (or a round iron bar, a continuation of 
the axle upwards) passed through a hole in its centre in 
which it turned freely without disturbing the stone. The 
top of the axle or bar was fixed firmly in the upper stone, 
which turned with it. A simple contrivance for slightly 
altering the distance between the two stones enabled the 

* Much of the description that follows is abridged from the accounts 
given by many eye-witnesses, as they saw little mills of this kind working 
in Ireland and Scotland within the last 250 years : as quoted by Mr. 
Robert Mac Adam in an interesting article by him on " Ancient Water- 
mills " in Ulster Journ. Archseol., vol. iv., p. 6. 




operator to grind coarse or fine. There was an opening 
near the centre of the upper stone with a hopper or open 
box fixed over it (often called the Ml or mouth), through 
which the grain was supplied : and the flour or meal, as 
it escaped at the edges of the 
stones, was received in a cloth of 
some kind. 

The water was directed through 
a spout in a powerful stream 
against the little spoons or 
paddles, and turned the wheel 
round very quickly, ioo revolu- 
tions or more in a minute ; the 
wheel whirling with it the axis 
and the upper millstone. All this 
corresponds with a passage in 
the " Montgomery mss.," written 
in the seventeenth century, quo- 
ted by Mr. Mac Adam in his 
article. This passage also indi- 
cates that these mills were very 
common in Ulster : — 


FIG. 517. 
Mill shaft and wheel, found near Bally- 
money, in Antrim. Length of shaft, 6 
feet 6 inches : diameter of paddle-wheel, 
3 feet 3 inches. (From Ulster Journal ot 
Archaeology, IV., p. 6.) 

[From a bog near Newtownards in 
Down] " issue many rills and streams 
. . . ; and on them each townland 
almost had a little miln for grinding 
oats, dryed in potts or singed and 
leazed in y e straw, which was y e old 
Irish custom, the mealle whereof called 
greddane was very cours. The milns 

are called Danish or ladle milnes ; the axeltree stood upright, 
and y e small stones or querns (such as are turned with hands) on 
ye top thereof : the water wheel was fixed at y e lower end of y e 
axeltree, and did run horizontally among y e water ; a small force 
driving it." 

These little mills were common in other parts of Ire- 
land also, and fifty or sixty years ago they were in full 
work all over Connaught, and probably also in Munster. 


The Irish-speaking people, to distinguish them from mills 
of a larger and better kind with vertical wheels, gave 
them the expressive name of muileann ton re talamh, 
" molendinum podex ad terram " (' mill with backside 
to the ground '). The Connaught people when speaking 
English called them " gig-mills."* These descriptive details 
regarding small mills are given here from late authorities ; 
first, because there is hardly any early literature that 
enters into particulars regarding their construction and 
mode of working ; and secondly, because it is pretty 
certain they descend from ancient times, like many other 
Irish institutions. 

Little mills of this kind, which did not call for much 
skill, and were very inexpensive, no doubt existed from 
the earliest period, as well as the larger and more expensive 
ones. They are not Danish, as they are called in the above 
extract : for the Danes, who did not begin to arrive till 
the ninth century, had nothing to do with the introduc- 
tion of mills, which, as we have seen were known and 
worked in Ireland long before their time. The popular 
tradition attributing them to the Danes, referred to by 
Mr. Mac Adam (p. 14 of his Essay), counts for nothing; 
for we know that similar popular traditions attribute all 
the raths and lisses, as well indeed as most other important 
works, to that people ; so that the erroneous name " Danish 
raths " — like " Danish milnes " — was quite prevalent until 
very lately (see p. 65, supra). .f 

The truth is, as Professor O'Reilly has shown in his 
instructive article on " Ancient Horizontal Water-mills, 
Native [Irish] and Foreign " (Proc. R. I. Acad., 14 April, 
1902), the little mills of the pattern here described have 
been found in use, not only in Norway, Sweden, and the 

* Ulster Journ. of Archaeol., v. 91. 

t This tradition or opinion of the Danish origin of the little Irish 
mills is adopted without further inquiry by Messrs. Bennett and Elton 
in their History of Corn-Milling. 


British Isles, including Ireland, but also in France, Spain, 
Italy, Roumania, Greece, the Holy Land, Asia Minor, 
and even Western China. Where the knowledge of them 
originally came from it seems now impossible to tell. We 
are. only concerned here to assert that howsoever or when- 
soever they got introduced into Ireland, they were not 
brought hither by the Norse invaders. 

Ancient mill-sites and the remains of old mills have 
been found in various parts of Ireland buried deep in bog 
or clay, always beside a stream, many presenting appear- 
ances of very remote antiquity. Some are small horizontal- 
wheel mills like those just described ; some are the remains 
of larger mills with vertical wheels. In most of those sites 
millstones have been found, of various sizes up to three 
feet in diameter : and there is often a long narrow oaken 
trough or shoot — generally hollowed out from a single 
tree-trunk — for conveying the water to the wheel. Parts 
of the framework surrounding the mill, with the flooring, 
also remain in some of these old sites, mortised together, 
but never fastened by nails : the woodwork of all generally 
of oak. Sometimes a large cistern is found : one, for 
instance, 15 feet by 7, and 20 inches deep ; from which 
the immediate water-supply was led by the shoot to the 
little wheel : another is described in Stokes's " Life of 
Petrie," p. 126.* 

4. Drying and Grinding. 

Preparatory to grinding, the corn had to be dried in a 
kiln, which was, and is, called in Irish dith, gen. dtha [aw, 
aw-ha]. The oven containing the fire was called sorn or 
som-na-hdtha, ' oven of the kiln.' It was heated by fire- 

* For these old mills see the article on ' Ancient Irish Water-mills " 
by Mr. Prim, in Kilk. Archaeol. Journ., 1849-51, p. 154 : see also the 
vols, for 1860-61, p. 347 ; and 1899, pp. 221 and 223, 8 . and Professor 
O'Reilly's Notes on Horizontal Water-Mills, Proc. Roy. Ir. Acad., 1902. 
All in addition to Mac Adam's Article in Ulster Journ. Arch., vol. iv. 


wood ; and it required some skill and experience to 
manage, for, if overheated, the kiln might take fire or the 
corn be scorched. On one occasion when St. Ciaran, dur- 
ing his residence in St. Enda's monastery in Aran, was 
drying corn, the kiln caught fire. There was a specially 
experienced man in charge of the kiln. St. Aengus the 
Culdee, when resident in Tallaght disguised as a working 
man, had charge of the kiln for some time. A usual plan 
was to put the grain in a sort of twig-basket or sieve which 
was held over the fire, while a man kept stirring it up, till 
the whole basketful was sufficiently dried. The basket 
was called in Irish laetn ; and Latin writers of the Lives 
of Irish Saints refer to it as rota de virgis contexta, ' a rota 
or round sieve woven of twigs.' Adamnan (p. 88, 2 ) calls 
the drying kiln by the Latin name canaba.* 

A more primitive way of drying, which was practised 
down to recent times, was by burning or roasting the corn 
in the ear. A woman — sitting down at her work — took a 
handful of unthreshed corn in the left hand and a short 
stick in the right : she then set fire to the ears, which 
blazed up ; and watching the right moment, when the 
outer husk or chaff was burned off, but before the fire had 
time to reach the grain, she struck off the burning top with 
the stick. Most country-women could do this work with 
more or less skill ; but it would seem that certain women 
followed it as a sort of trade ; and constant practice made 
them dexterous, so that they separated the grain very 
quickly. Corn burned off in this manner was called 
loisgredn [lusgraun], i.e. ' something burned,' from loisg, 
to burn : and the practice must be an old one, for many 
places in Ireland are still called by names derived from 
this word — such as Loskeran near Ardmore in Waterford, 
probably commemorating the fact that at some former 
time a professional corn-drier lived there, f In Ulster 

* See Adamnan, p. 88, note c, and p. 362. 
f See Irish Names of Places, 1. 238. 

chAp. xxv] corn mills 343 

and Scotland scorched corn is called graddan, as stated 
in the Montgomery MS. above (p. 339), from the Gaelic 
gread, to burn (with which the English gridiron and 
griddle are connected) ; and the Scotch and Ulster 
peasantry greatly preferred graddan bread (which has a 
slightly burnt taste) to that made from kiln-dried corn. 
Martin (p. 204) says that in his time, 1703, corn could 
be dressed, winnowed, ground, and baked, in one hour 
after reaping (see also Carmichael's " Carmina Gadelica," 
1. 254). 

The ground corn came from the mill in the form of 
whole-meal. If different qualities of bread or of porridge 
were required, this meal was sifted in a criathar [criher] or 
sieve, which, as well as the baking, was always done by 
women, as already remarked. 

The Brehon Law provided for personal injuries in 
mills, caused by culpable negligence. In case any one 
was injured in a kiln during the process of drying, four 
persons are mentioned in the Book of Aicill, one or more 
of whom might be liable for damages : — The man who 
splits the firewood, the man who kindles the fire, the man 
who puts on the firewood (i.e. tends the fire), and the man 
who dries the corn.* 

When the upper millstone was badly set, it was liable, 
in its rapid revolution, to break from its fastening and slip 
off the lower one : and so to injure persons looking on 
or engaged in the work. The Book of Aicill lays down 
rules for compensation in such cases, and mentions three 
persons, of whom one, two, or all three might be liable 
according to the apportionment of the blame : — the owner 
of the mill, the mill-wright who constructed it, and the 
person engaged in grinding, f Sometimes accidents 
happened from the too great force of the water : and 
here again the proper assignment of liability is provided 

* Br. Laws, 111. 265. f Ibid., in. 281 283 


5. Common Properly in Mills. 

A mill was a usual appanage to a ballybetagh or 
ancient Irish townland, and went with it on sale or other 
transfer, as is proved by records of many ancient grants 
and purchases of land. Written into the spare blank pages 
of the Book of Kells are several such grants, some in Irish, 
some in Latin : and in the Registry of Clonmacnoise are 
similar documents. One of those in the Book of Kells 
records that in the middle of the eleventh century, the 
munter or family of Kells made a grant of Ballyheerin 
with its mill and with all its land, and Ballycoogan with all 
its land and with its mill, to God and St. Columkille, 
meaning that they were granted to St. Columkille's 
monastery at Kells.* In the Charter of Newry in which 
King Murkertach O'Loghlin granted several townlands 
to the monastery there, about the year 1101, this expres- 
sion (in Latin) occurs : — " These lands, with their mills 
" [molendinis], I have confirmed of my own proper gift 
" to the said monks, "f Several other such grants of town- 
lands of about the same period, f in which mills are included, 
might be mentioned. All the mills in question here were 
large ones with vertical wheels. 

The mill belonging to a ballybetagh or townland was 
often owned by several families in common, all of whom 
had a right to the use of it, according to the amount of 
their several shares. " Whenever a mill was to be erected 
for the use of neighbours " — writes Dr. O'Donovan in an 
instructive note on the Brehon Law tract on the Right 
of Water§ — " It was left to the option of the persons 
" concerned (who were generally the inhabitants of the 
" three nearest lands) whether they would all join in con- 
" structing the works and conducting the water thereunto, 

* Irish Arch. Misc., 1846, p. 129. f Dub. Pen. Journ., I. p. 102. 

X For which see Irish Arch. Misc., 1846, pp. 127-160. 
§ Br. Laws, iv. 220, 221 

Chap, xxv] corn mills 345 

" or let all be done by one man, who was to pay his 
" neighbours for conducting the water through their lands. 
" If the neighbours had assisted in forming the mill-pond, 
" mill-race, and other works, they were entitled to certain 
" days' grinding at the mill." In order to assign the 
number of days belonging to each partner, there was a 
regular rotation extending over three weeks, i.e. eighteen 
working days* ; and the usual arrangement was that the 
several owners or claimants had the use of the mill on 
certain days of those eighteen, according to their several 
claims ; at the end of which the rotation began again and 
went on in the same order. An outsider could get his corn 
ground by purchasing for a sufficient time the right of one 
or more of the owners. When one rich man constructed 
the mill with the consent of the neighbours, he paid all 
expenses, both of purchase and work, and then the mill 
belonged to him. In this case he ground his neighbours' 
corn for payment, which usually consisted of a certain 
proportion of the corn or flour : commonly a tenth (see 
p. 329, supra). 

Sometimes a man who had a share in a mill had a kiln 
of his own, and dried his corn at home ; and occasionally 
a kiln, as distinct from a mill, was owned by several people 
in common. A brugaid or brewy always had a mill and 
kiln on his premises — a thing we might expect, from the 
quantity of provisions he needed. In connexion with most 
monasteries was a mill for the use of the community.! 

6. Querns and Grain-Rubbers. 

A grinding machine much more primitive and ancient 
than the water-mill was the quern or hand-mill. It was 
called in Irish bro, gen. bron [brone] : and often cloch-bhron 
[cloch-vrone] : clock, a stone : but both these terms were 
also applied to a millstone. An older term was meile 

* Br. Laws, iv. 215-219 : see also Introd., clxi ; p. 305, 2S . and vol. 1 
217, 227 t See Br. Laws, iv. 309, a6 . ani 305, 2S . 315, Ifi 




[melle : O'Clery's Gloss.], evidently cognate with English 
mill and the corresponding terms in other languages. 
Querns were of various forms : sometimes the grinding 
surfaces were fiat : sometimes the under surface was con- 
vex and the upper concave : sometimes the reverse — pot- 
shaped. In all cases the upper stone worked on an axis 
or strong peg fixed in the lower one, and was turned 
round by one or by two handles. The corn was supplied 

at the axis-opening 
in the centre of the 
upper stone, and 
according as it was 
ground between the 
two stones flowed 
out at the edge. 
Sometimes it was 
worked by one per- 
son, sometimes by 
two, who pushed the 
handles from one to 
the other. In ancient 
times it was — in 
Ireland — considered 
the special work of 
women, and especi- 
ally of the cumal or bondmaid, to grind at the quern : 
and so generally was this understood that in Cormac's 
Glossary (p. 42) a cumal is explained : " a woman that is 
" grinding at a quern ; for this was the business of bonds- 
" women before [water-] mills were made." Querns were 
used down to our own day in Ireland and Scotland ; and 
they may still be found at work in some remote localities, 
especially to grind malt secretly for making pottheen or 
illicit whiskey. 

The almost universal use of querns is proved by their 
frequent mention in the Brehon Laws and other ancient 

Fig. 318. 

Upper stone of a quern : 18 inches in diameter : ornamented 
with sculptured cross. In National Museum. (From Wilde's 
Catalogue, p. 107.) 


Irish literature, as well as by the number of them now 
found in bogs, in or near ancient residences, and especially 
crannoges. Some of these are very primitive and rude, 
showing their great antiquity. 

In comparatively modern times mill-owners who ground 
the corn of the people of the neighbourhood for pay looked 
on the use of querns with great dislike, as taking away 
custom. Quern-grinding by the poorer people to avoid 
the expense of the mill was regarded as a sort of poaching; 
and where the mill belonged to the landlord he usually 
gave orders to his miller to break all the querns he could 
find ; so that the people had to hide them much as they 
hide an illicit still nowadays.* 
In Scotland laws were made in 
the thirteenth century to com- 
pel the poor people to abandon 
querns for water-mills, all in the 
interests of landlords and other 
rich persons. It was the same 
in England: in 1556 the local r , t ' '. . . 

o J J Complete pot-shaped quern : 9 inches 

lord in one of the western !" dian £" -, 1 " Tf National *! useum ' 

(From Wilde s Catalogue, p. 108.) 

counties issued an order that no 

tenants should keep querns " because they ought to grind 
at their lord's mill."f But these laws were quite ineffec- 
tive, for the people still kept their querns. Pennant and 
McCulloch found them in general use in the Scottish 
Highlands and islands at a recent period. 

When two women worked the quern, they sat facing 
each other, and passed the handle, or both handles, quickly 
from hand to hand. They ground oats always in the husk 
and afterwards silted it. Before grinding— in the absence 
of a mill-kiln — the corn was very often dried in an iron pot 
over a fire, and was kept constantly stirred round to pre- 

*Dub. Pen. Joum., iv. 295, where there is an interesting and instructive 
article on this subject : see another good article in Kilk. Archseol. Joum., 
1858-9, p. 352. f Roberts, Social Hist., p. 323. 




vent scorching. Quern-grinding was tedious work : for it 
took about an hour for two women to grind 10 lb. of meal. 
In Scotland, oatmeal or a preparation from it, is called 
broes or brose, which is probably a plural form from bro, 
the Gaelic name of the quern. It is hardly necessary to 
say that the quern or handmill was in use among all the 
ancient peoples of Europe, Asia, and Africa : and that it 
is still extensively employed where water-mills have not 
found their way. 

The most ancient grinding-machine of all, and most 
difficult and laborious to work, was the grain-rubber, about 
which sufficient information will be derived from the 
illustration. Several of these primitive grinding-machines 
may be seen in the National Museum : they are still used 
among primitive peoples all over the world. 

Grain-rubber: oval-shaped : 16 inches lone. (From Wilde's 
Catalogue, p. 104.) 


Sculpture on a Capital: Priest's House, Glendalough: Beranger, 1779k 
(From Fetric t Round Towers.) 

Ornament on leather case of Book of Armagh. (From Petric's Round Towers, 



Section i. Wool and Woollen Fabrics. 

hearing. — Clothing, as may be anticipated, gave 
rise to many industries, in Ireland as in all 
other civilised countries. Of these, the most 
important was that connected with wool. The 
Irish name of wool was olann or oland, which 
is still in use. The wool was taken from the 
sheep with a shears, which, from the manner 
in which it is. mentioned, must have been much like those 
used at present. The usual old Irish name is demess, 
which is explained in Cormac's Glossary (p. 55) in a 
manner that clearly indicates the make of the article 
itself. He says that it was so called from mess, an edge ; 
and that demess signifies ' two edges ' ; for — he goes on to 
say — it has two knives, and the knives have two handles. 
This shows that the old Irish wool-shears was like the 
present hedge-shears. The demess is mentioned in the 
story of the Feast of Bricriu (33 and 162), which throws 
back the knowledge of the instrument to a still earlier 
date. In the Brehon Law (iv. 310, 12) it is called dias, 
which means ' a pair,' that is to say, a pair of blades. 
A small hand-scissors was also in use, and known by 
the same name. We read in the Tripartite Life (p. 103) 



that St. Patrick tonsured the druid Caplait : " and Patrick 
put the deimess round his hair." This old name is still 
used for a shears, in the modern form deimheas, which 
is pronounced djeeass. About the year 1849 an ancient 
iron shears was found in a tumulus at a place called 
Seskin in the County Kilkenny.* The process of shearing 
was called lomrad, from lorn, bare. 

Preparation lor Spinning. — The shearing appears to have 
been done by men : but after this the whole work up to 
the finished cloth was regarded as specially pertaining to 
women : except fulling, which was often or mostly men's 
work. After being sorted, the wool was greased (belad, 
greasing), adding to the natural oiliness, which rendered 
it more easy to remove the grease altogether in the next 
process — scouring. After scouring it was teased or mixed 
(cutnusc or bocad, i.e. ' mixing '). It was next combed or 
carded twice, first roughly, and a second time more care- 
fully and finely. The carding [chad, pron. keera : from 
cir, a comb) was done by hand : the woman sitting down 
while at work, and using a pair of cards, much the same 
probably as those in use for hand-carding now. A quantity 
of wool lay at her feet in a sort of bag called a pes-bolg 
(which the gloss dervies from pes, a foot : ' foot-bag '), from 
which she drew handfuls as needed. The second carding 
turned out the wool in the form of soft little loes, locks or 
rolls (a l-loaib, ' in locks ' : lo, a lock of wool) fit for spinning, 
just as wool-carders do at the present day.j 

Spinning. — In those times spinning was done, in Ire- 
land as elsewhere, by the distaff and spindle ; for the 
spinning-wheel was not invented till the fifteenth or six- 
teenth century. The wool or flax in preparation for 
spinning was wound and fastened loosely on a rock or 
distaff called in Irish cuigeal [quiggail]. From the distaff 
the material was drawn off gradually, with the help of the 

* Kilk. Archaeol. Journ., i. 9. 

t Br. Laws, n. 369, 371, 417, 419 : O'Curry, Man. & Cust., 11. 115. 


left hand, by the spindle or spinning-stick, which was held 
in the right hand and manipulated dexterously so as to 
twist the material into thread, and wind it on the spindle 
according as spun. When one spindle was full, the operator 
began with another. The spindle used for flax was called 
in Irish fertas : that for spinning wool was called snimaire 
[sneemara or sneevara], lit. ' spinner,' from snim, modern 
Irish sniomh, spin.* But there seems to be some con- 
fusion in the gloss in the use of these words — fertas and 
snimaire ; and at any rate, the distinction is now forgotten. 
That a part at least of the process of spinning was often 
performed by bondmaids appears from the derivation given 
in Cormac's Glossary (p. 14) for abras, a word which was 
applied to yarn of any kind, or to the material for making 
thread, rolled on the rock or distaff. He derives it from 
abra, a bondmaid, and feis, hand-produce : abra-feis (con- 
tracted to abras), because it is — as the Glossary goes on to 
say — " the hand-produce of a bondmaid." 

The abras or thread ready for weaving was rolled "up 
in balls (Irish certle, equivalent to Latin glomus, a clew or 
thread-ball : Z, 68, s), on which it was wound from the 
spindles according as these got filled. The following 
quotation from the Law gloss makes matters clear : — 
" Abras, i.e. the material finished [as thread or yarn] and 
" wanting only to be woven, i.e. the white balls [na certle 
" gela], i.e. white thread."f The fact that the thread, 
" wanting only to be woven," i.e. ready for weaving, was 
white, points to the wool in its natural colour, and is a 
confirmation of the statement made farther on, that woollen 
material was dyed in the piece. 

Weaving. — The thread was woven into cloth in a hand- 
loom, nearly always by women : and like all the rest of the 
cloth-making process, it was a cottage industry. The 
complete weaving machinery or loom had two beams : 

* Br. Baws, 1. 152, I0 , „. 153, I4 . 

1 Ibid., I. 152, „; 153, «. 




the larger one called gar main (and sometimes gae-mathri), 
and the other lu-garmain or 'smaller beam' (/#, small); 
which O'Curry believes to have been the front beam on 
which the warp was rolled up to be woven, and from which 
it was unrolled as the weaving went on. Accordingly he 

Fig. 322. 

Fin. 323. 

Fig. 324. 

Fig. 323 

Specimens of ancient Irish weaving. In 1780 the body of a woman, covered with antique articles 
of clothing, all wool, was found, buried in hard gravel, under 4% feet of bog. Figs. 321 and 322 
represent portions of two of these. Fig. 323 la part of a long web. made of goat's hair, not exactly 
woven, but tied or knotted together, as sl,-..u in the illustration. The hairs are across, or at right- 
angles to the length. Found in Cavan, under 14 feet of bog. With this was found a fine plaited or 
woven woollen band, portion of which is shown in fig. 324. Fig. 325 is portion of a coarse woollen 
cloth, of which there is a whole suit in the National Museum. It appears from these and other 
specimens that certain loom-adjustments of the warp, commonly supposed to be of modern inven- 
tion, were known to the ancient Irish weavers. (From rroc. Roy. Ir. Acad., IX. 103, 104; and 
Wilde's Catalogue, 295 and 325.) 

calls it the " rolling beam." The principal beam must 
have been large : for we find it recorded that a certain 
widow cooked a calf in her house for St. Brigit with a fire 
made from her garmain, as she had no other fuel ; and the 
massive spear of a hero, like that of Goliath, is sometimes 
compared^— in Irish tales — to a weaver's beam. In the 

chap. Xxvt] Trades connected with clothing 353 

Annotations of the Feilire of Aengus* mention is made of 
the " nin of a garmain, i.e. the ' fork ' or ' mouth ' on the 
head of a weaver's beam," which refers to some peculiarity 
of construction. 

What were called the " swords " (claidim), or weaving- 
rods (slata figi), were long laths used during the process of 
weaving, which were nearly or altogether as long as the 
beam : for in the Bruden Da Derga it is stated that the 
three great swords of three champions were each longer 
than a claidem n-garmnae, ' the sword of a weaver's beam.'f 
These swords or laths are what O'Curry calls " heddles," a 
word used in this application in his own early days in 
Clare. J The warp was called dluth [dluh] : and the weft 
or woof innech.§ While the woman was weaving she used 
a feith-geir [feh-gair], " which put a smooth face upon her 
weaving " : and which is represented by the sleeking-stick 
or " rubbing-bone " still used by hand- weavers. || 

The piece of woven cloth had usually a border or 
fringe (corrthar, pron. curher), which was sometimes woven 
with the whole piece and formed part of it : and some- 
times separately and afterwards sewed on. In this last 
case it was woven with a short light claidem or lath, alto- 
gether apart from the loom, something like the crochet, or 
netting or meshing work of modern times : and weaving 
ornamental borders or long scarfs in this manner was 
practised by ladies of the higher ranks as they practised 
embroidery. We read in the Tain that once when Queen 
Maive was in her chariot, a strange lady suddenly appeared 
sitting beside her : and " what the woman was doing was, 
" weaving a border (corrthar) with a claidem or lath of 
" findruine [findrinne] or white bronze."^ 

* Feilire, p. 66, and note a ; and " Ninach " in Glossary. 
f LU, 95, a, 33 : and Man. & Cust., n. 148, note 221 
X Man & Cust., n. 116. 

§ Corm. Gloss., 95 : the weft is also called eanglaim in O'Cl. Gloss. 
|| Man. & Cust., 11. 116. 

1J Man. & Cust., 11. no, note 71 LL, 55, b, 32. See also corrthair in 



Fulling. — A fuller of cloth was called ciormhaire 
[keervara], literally a comber (from cior, a comb) ; or 
fucaire [fookera], or ucaire, from fucad or ucad [fooka, 
ooka], to full, and there were persons who practised this 
as a distinct trade. In the Irish Annals it is recorded that 
Cerbhall [Kerval] king of Leinster was on one occasion 
riding a spirited steed through Kildare ; when, passing the 
shop of a fuller, it happened that a man was sent out to 
blow the fuller's congna or horn : and the horse shied and 
started at the sudden sound, so that the king was wounded 
by his own javelin, of which he died (a.d. 909).* This 
incident tells a plain story. The fuller began his opera- 
tions on each occasion whenever his materials were 
prepared, fulling a large quantity at a time, and he fulled 
his neighbour's cloth as well as his own— for pay of course. 
When he was ready to begin, he sent out his man to blow 
the horn at the door, as a signal for the people to bring 
in their cloth. The custom of tradesmen blowing a horn 
for such purposes continued to a period almost within our 
own memory, of which an example, in case of a different 
trade, may be seen in the beginning of the story of " The 
Whiteboy " in the Dublin and London Magazine, vol. 
for 1826, pages ys, 74. 

2. Flax and its Preparation. 

The preparation of flax is described in old Irish 

authorities, especially in the Brehon Law, though not in 

such detail as that of wool. One of the names of this 

plant is still preserved in a great number of the Euro- 

Corm. Gloss., 44. Besides the special references given here, see for the 
whole of these details of weaving, Br. Laws, I., pp. 150-153 : and O'Curry, 
Man. & Cust., n. 116, 117. In the Highlands of Scotland they still pre- 
serve most of the ancient methods of manufacturing cloth, from the wool 
up, including dyeing, as may be seen by the brief, but very interesting, 
description of the processes as carried on by the Highland women within 
the last forty years given in Carmichael's Carmina Gadelica, 1. 298, 306, 
308, 310. 

* Three Fragments, 223, notes c and d. We now know that congna 
means an antler or horn. 


pean languages, the forms slightly varying, but all 
derived from the root lift. The Greek word is linon ; 
Latin, linum ; English, linen and linseed ; A. -Sax., lin ; 
Russ., lend ; &c. This shows that it was cultivated by 
the western Aryan people since before the time of their 
separation into the various nationalities of Europe. 

The Celtic tribes who first set foot on our shores, 
brought a knowledge of the plant and its cultivation 
with them ; and corresponding to all the names given 
above, is the Irish lin [leen], which is still the word in 
universal use for flax. Besides the evidence of philology, 
our own records show that linen was manufactured in 
Ireland from the earliest historic times. It was a very 
common article of dress, and was worked up and dyed 
in a great variety of forms and colours, and exported 
besides in large quantities to foreign nations. So that 
the manufacture for which Ulster is famous at the present 
day, is merely an energetic development of an industry 
whose history is lost in the twilight of antiquity. 

The flax, after pulling, was tied up in sheaves and 
dried. It was then steeped, as at present, to rot the 
woody fibre ; and after remaining a sufficient time in the 
water it was taken up and spread out to dry. After a 
still further and final drying, over a fire it was beaten 
with a smachtin or flax mallet, to break up the brittle 
woody covering of the flax fibre. In order to remove 
this, the operation of scutching (flescad) came next, which 
was done with a scutching-stick called flesc or flesc-lin.* 
In the commentary on the Senchus Morf this process 
gets another name — ailgubad. More than half a century 
ago, scutching was called cloving by the English-speaking 
people of the south of Ireland, where flax-growing and 
linen-weaving still lingered on from the days of old : and 
the forked scutching-stick or flesc-lin — which was always 

* Br. Laws, i. 152, I5; 153, 2I . Man. & Cust., n. 116. 
f Br. Laws, 11. 368, line 3 from bottom. 


worked by women — was called a " cloving-tongs " : the 
word " clove " being merely the Irish clomh or clobha 
[clove, clova], one of the names for a tongs. But the 
whole industry — which I saw in full work — is now dead 
and gone. After the flescad or scutching came what was 
called in English in modern times, " hackling," to divide 
the fibres into finer filaments, which brought away tangled 
masses of tow. This was done by drawing it with the 
hand over the points of a number of strong steel needles 
fixed closely in a little frame : a work done by certain 
persons as a special trade. This hackling process is not 
mentioned in the old account in the law from which the 
description and names of all the other processes given here 
are taken : but the flax must have been subjected to it. 
Next came spinning into thread with a distaff and a 
spindle, or in later times with a spinning-wheel, which 
is still found at work in the homes of the peasantry in 
some parts of Ireland. The thread was made up in 
hanks or skeins, boiled in home-made potash, and spread 
in the sun to bleach on a grassy spot called a tuar or 
bleach-green. Lastly, it was wound up into balls or clews 
(certle) and woven into calico or linen.* 

3. Dyeing. 
Dyestuffs and dyeing in general. — The beautiful illumi- 
nation of the Book of Kells, the Book of Mac Durnan, and 
numerous other old manuscripts, proves that the ancient 
Irish were very skilful in colours : and it will be shown here 
that the art of dyeing was well understood. The dyestuffs 
were not imported : they were all produced at home : and 
so important were they considered that among the blessings 
believed to fall on the country during the reign of a just 
sovereign, the Book of Leinster and other ancient authorities 
enumerate " abundance of dyestuffs."f 

* For all this about flax, see Br. Laws, 1. 150-153 : and O'Curry, Man. 
& Cust., 11. 121, where a good abstract of the processes is given. 

J Rennes Dind. in Rev. Celt., xvi. 281 : O'Curry, MS. Mat., 528, top. 


In this Book of Leinster passage the word used for 
dyestuffs (in general) is ruaman (gen. ruamna) : and hence 
the Four Masters (vol. I., 42, 5, 6) use the verbal form 
ruamnad [roomna] for dyeing of any colour — though, as 
we shall see, ruam primarily means red : " whence comes 
ruamnaig, reddening, or blushing, and ruanaid, red." The 
word ruam, as it is written in one copy of Cormac's 
Glossary, or main as it appears in another, is stated in 
the Glossary* to be " an herb (luss) that gives colour to 
the face until it is red " (derg) : and in the Coir Anmann 
we are told that a person named Diarmait Ruanaid, i.e. 
' Dermot the Red,' was so called from " ruan," which " is 
a plant (lus) that produces colour on the face."f In an 
old tract in the Book of Ballymote the tree or bush 
called rois — which is understood to be the elder-bush — 
is designated " the reddening of faces." Ruam is the 
alder-tree, more commonly called fearn or fearnog : and 
as this plant is used in dyeing a reddish brown, it may 
be concluded that the words ruam, ruaim, ruan, ruain, 
which we find in the authorities, are all different forms of 
the name for the alder-tree. 

It has been already stated that the Irish people were 
fond of bright colours : and they wore in fact clothes of 
all the chief colours then known. But only in a few 
cases have descriptions of the processes of producing the 
dyestuffs and of imparting the colours come down to us, 
and even those we have are often not very precise or 
clear. The people understood how to produce various 
shades by the mixture of different colours, and were 
acquainted with the use of mordants for fixing them. 
One of these mordants, alum, is a native product, and 
was probably known in very early times. J Dyeing was 
what we now call a cottage industry, i.e. the work was 
always carried on in the house : as I saw it carried on 

* Corm., p. 144 : Three Ir. Gloss., 39. f I r - Texte, III. 345, 347. 

X Sullivan, Introduction, 402. 


in the homes of Minister more than half a century ago. 
In the cultivation of the dye-plants, men might take a 
part : but the rest of the process was considered the 
special work of women, so that men seldom assisted. 
In the actual dyeing of the cloth, even the very presence 
of man or boy was considered unlucky, and liable to mar 
the process, as is shown by the legend of St. Ciaran given 
at p. 360 below. It appears from the same story, as well 
as from what is said about white balls of yarn at p. 351, 
supra, that cloth was dyed in the piece, the wool being left 
of the natural colour till after weaving and fulling. But 
woollen cloth was often worn without being dyed at 
all — just with the shade it brought from the back of 
the sheep. 

Ground Colour. — There were two main stages in the 
process of dyeing. The first was imparting a ground or 
foundation colour of reddish-brown, which was done by 
steeping and boiling the cloth with the twigs of the ruam 
or alder. " Ruadh " (red : pron. roo) — says O'Clery's 
Glossary — " i.e. ruamann, the first dye or tinge, or the 
" stuff that gives it and prepares for the second or last . . . 
" no colour [can be given] without ruamann." In later 
times this preliminary colouring was called in English 
riming, from ruaim. After this the cloth was ready for 
the second stage — imparting the final colour : which was 
done by boiling it with the special dyestuff. 

Black. — The dyestuff for black was a sediment or 
deposit of an intense black found at the bottom of pools 
in bogs, called dubh-poill, i.e. ' black-stuff of the poll, hole 
or pool.' It always contained more or less iron, which 
helped in the dyeing. Boiled with this, the cloth acquired 
a dull black colour : but if some twigs or chips of oak were 
added, the colour produced was a glossy jet black, very 
fixed and permanent. 

Crimson. — A crimson or bright-red colour was imparted 
by a plant anciently called rud or roid, which required 


good land, and was cultivated in beds like table vege- 
tables, requiring great care.* It was probably a species of 
the plant called bedstraw. In the Senchus Mor provision 
is made for dividing the home-stock of this dye-plant, or 
rather of the prepared dyestuff, in proper proportions 
between husband and wife in case of separation,! which 
shows that it was of much value (see p. II, supra). The 
several stages of preparation are indicated by distinct 
terms. First, the plant as gathered from the beds : second 
stage, trilsens : third stage, scriplins% : fourth and last 
stage, the dyestuff, which was a sort of meal or coarse 
flour of a reddish colour.§ Some sort of crimson was also 
produced from lichen as mentioned below. 

Blue. — To dye the cloth blue, after it had been rimed, 
it was boiled with a dyestuff obtained from woad, called 
in Irish glaisin [glasheen] : the Irish word evidently a 
descendant of the Gaulish name of this plant — glastum. 
Pliny records that the ancient Britons used the glastum to 
dye their bodies blue. The name glaisin, which has long 
fallen out of use, was also applied to the prepared dye- 
stuff. The glaisin was cultivated in beds, and was a very 
valuable crop, requiring great care and watching during 
growth.|| In vol. I., p. 216, supra, has been mentioned a 
celebrated lawsuit brought about by sheep eating a crop 
of glaisin. 

That the dye of glaisin was blue is indicated by the 
name, which is a diminutive of glas. This word glas was 
however applied to several shades of colour, as for instance 
to the green of fields and to bluish-grey coloured eyes. 
But it was also applied to pure blue, as is shown by many 
ancient passages, as for instance the Voyage of Bran, 

* Br. Laws, iv. 277, I0 . f Br. Laws, 11. 421. 

% See Scriplin and Trillsen in Atkinson's Glossary to Br. Laws. From 
the authorities he quotes one may conjecture that the trillsen was a 
little wisp of the dried plants : and the scriplin a larger bundle in a 
further stage of preparation. 

§ Br. Laws, 11. 421 : see also MS. Mat., 528, 29 . || Br. Laws, 11. 371, bot. 


where in one place (1. 9. is) the word glas is applied to 
" the hue of heaven." But as to glaisin, the colour it 
imparted is placed beyond dispute by a legend in the Life 
of St. Ciaran in the Book of Lismore. On a certain day, 
when he was a boy, his mother was about to dye some 
cloth with glaisin : — " Then his mother said to him — ' Out 
with thee now, Ciaran ' : for " — continues the old Irish 
narrative — " they did not deem it right or lucky to have 
" men [or boys] in the same house in which the cloth was 
" dyed." Ciaran walked out, saying in a childish pout as 
he went : — "I wish that there may be a dark grey stripe 
in it." Accordingly when the cloth was taken out finished, 
every piece had a dark grey stripe [which spoiled it]. 
Again the glaisin was prepared and the cloth was boiled : 
and this time — on account of some other words spoken by 
Ciaran — it came out whitish. A third time the glaisin 
was prepared : and the boy's mother said to him : — " Now 
" Ciaran do not spoil the glaisin any more, but give it a 
" blessing." He did so : and this time the cloth came out 
dyed a beautiful intense blue (gorm).* Here the word 
applied to the colour produced by glaisin is gorm, which 
means pure blue. The legend attributes to a miracle 
what must have been a usual occurrence : failure by some 
mismanagement, followed by success after more careful 

In the preparation of dyestuffs from glaisin there were 
four distinct stages as in case of the roid plant. First, the 
plant as gathered from the bed : second, a stage called cro 
or cru : third, a second stage of cro : fourth, the fully- 
prepared dyestuff, which was in lumps or cakes. But 
what the first and second cro states were we do not know. 
Here also, as in the case of roid, the law provides for the 
proper division of the glaisin between husband and wife 
if they should separate, f 

* Stokes, Lives of SS., 266. 

f For all this see Br. Laws, 11. 419 


Purple was called in Irish corcur, which answers to the 
Latin -purpura by the usual change from p to c. Purple 
cloaks, purple flowers, and purple colour in general, are 
very often mentioned in Irish writings, such as the 
Tripartite Life, the Book of Rights, the Tales, &c. ; 
showing how familiar this colour was. Purple dyestuff 
was obtained from a species of lichen, and also from a 
cockle-fish. In one of the pages of an ancient manuscript 
now in Turin, is a passage written by an Irish hand in the 
beginning of the ninth century, and published by Chevalier 
Nigra in his Irish Glosses, which proves that at that early 
time the Irish were acquainted with the art of dyeing 
purple by means of a lichen. The gloss which the Irish- 
man wrote in explanation of the Latin text is this : — 
Donaib caircib, edon, ar is di lus bis forsnaib caircib dogniiher 
in chorcur buide : which is in English " from the rocks, that 
"is to say, because it is from a plant which is found on 
" rocks the yellow purple is made." (In Sullivan's Introd., 
p. 643, the word buide of this gloss is misprinted bailie). 
There is even a more direct notice of rock-purple in a poem of 
the Agallamh (in LL) in praise of Aran Island in Galway bay, 
in which it is mentioned that purple-lichens (corcra) grow 
on the rocks there (Silva Gad. 109, i 9: Irish text, 102, 7 ). 

The knowledge of dyeing from rock lichen was never 
lost, but was continued from generation to generation 
down to recent times. When Martin visited the Scottish 
Western Islands in 1703, the people there dyed " a pretty 
" crimson colour with a scurf scraped off rocks and sub- 
" jected to proper preparation." Joseph Cooper Walker 
tells us that in his time — the beginning of the last 
century : — " The purple was obtained from the coarser 
" kind of arcell [or orchil] growing on rocks, which, being 
" steeped in urine, and made up into balls with lime, pro- 
" duced a beautiful purple. Considerable quantities thus 
" made up are frequently sold in the market at Dingle."* 
* Memoirs of the Irish Bards, 11. 264 


Walker also mentions, in the same passage, that " a fine 
" bright crimson dye was obtained from a finer kind of 
" lichen resembling a thin white scurf, which they scraped 
" from the rocks, dried, and reduced to powder, then infused 
" in urine for three weeks or a month." 

I cannot find any ancient Irish authority in which 
mention is made of purple being obtained from shellfish. 
But we may infer from several circumstances that this 
branch of the dyeing art was known to the ancient 
Irish. Mr. Franci Joseph Bigger, in the " Proc. Roy. Ir. 
Academy," vol. for 1893-96, p. 727, gives an interesting 
account of the remains of a prehistoric settlement in 
Connaught in which were whole heaps of a species of 
whelks called purpura lapillus — which we know are used 
to this day for dyeing purple in Ireland and elsewhere. 
He found all the shells broken uniformly at one particular 
point — just the point inside which was situated the elon- 
gated little sac containing the purple colouring matter : 
evidently with the object of extracting the precious little 
globule. In the time of Joseph C. Walker (about 1800) this 
method of dyeing purple — from " periwinkles and limpets " 
— was practised in the eastern Irish counties, as also on 
the opposite coast of Wales. He states (Ir. Bards, II. 265) 
that the shell was broken at a particular point at the back, 
very delicately, so as not to bruise the fish, and with a 
bodkin they picked out what he calls a " white vein," which 
yielded a few drops of the colouring liquor. This they did 
several times in succession at proper intervals, the fish 
renewing the liquor after each occasion. All this corre- 
sponds exactly with what Mr. Bigger found : so that the 
knowledge of this process has descended from prehistoric 
times to our own day. In like manner this process has 
been perpetuated from old times in Wales ; for we know 
that Bede (Eccl. Hist., 1. i.) records that in his day the 
Britons (or Welsh) produced a most beautiful purple 
colour from shellfish. The reader will scarcely need to 


be reminded that the celebrated Tyrian purple was 
produced in a similar way. 

The purple dyestuff, however obtained, was produced 
in very small quantities, so that it was extremely scarce ; 
and the colour was excessively expensive in Ireland as 
elsewhere : on the Continent in old times it was worth 
thirty or forty times it weight in gold. Partly for this 
reason, and partly for its beauty, purple was a favourite 
with kings and great chiefs, so that writers often designate 
it a royal or imperial colour. 

Saffron. — Until recent times linen was dyed saffron, 
probably with the crock or saffron plant (Lat. crocus), 
which was the simplest of all the dyeing operations. But 
I do not find this mentioned in any ancient authority. 

Popular Knowledge of Dyeing. — The Irish peasantry of 
the present day, as well as the Highland Scotch, possess 
considerable knowledge of the stuffs — chiefly obtained 
from herbs — used in imparting various colours, and are 
skilled in simple dyeing : knowledge and skill that have 
descended to them from old times. In Donegal they dye 
woollen cloth yellow with the tops of heather ; and light 
brown with peat soot : and in various parts of Ireland — 
as well as in Scotland — they use a sort of lichen called 
crotal, that grows on rocks, to impart a reddish brown. 
In the County Mayo a species of moss is used for dyeing 
stockings a reddish brown : and they also dye stockings 
black with the roots of the blackberry bush. 

4. Sewing and Embroidery. 

Needle and Thread. — The thread used for sewing was 
generally of wool : all the sewing on the various articles 
of dress found on the body of the woman mentioned at 
p. 352, supra, was done with woollen thread. In primitive 
ages fine filaments of gut were often used. The sewing- 
thread was kept in the form of a certle, clew, or ball, like 
that for weaving ; and women sewed with a needle 


furnished with a cro or eye as at present. From an early 
age needles were made of steel, but in primitive ages of 
bronze. In those days a steel or bronze needle was 
difficult to make ; and its value may be estimated by 
the fine imposed in the Brehon Law on a person with 
whom a needle was pledged, for withholding it when the 
owner demanded it back and tendered the loan. For a 
common needle it was a dairt or yearling calf ; for a needle 
used in the ornamental work on mantles, it was a colpthach 
or two-year-old heifer ; and for an embroidering needle, 
an unga or ounce of silver.* The word for a needle was 
sndthat [snaw-hat], which is still in use : it is derived in 
Cormac's Glossary (p. 150) from snath [snaw], a thread, 
and sit, a road or way, i.e. sndth-shet [snaw-hait], ' thread- 

=====a _ - road,' because the 

==========— thread passes 

F.C.,* through the cro or 

Two bronze needles, natural size. These, with a large eVC The first part 

number of others, are now in the National Museum, Dublin. . . 

(From Wildes Cataloeue, p. M 7.) 01 thlS deOVatlOn 

(from sndth) is correct, but the second is fanciful. Bronze 
needles are now often found, which, judging from both 
material and shape, must be of great antiquity, f 

Dressmaking. — Needlework was most commonly prac- 
tised in ordinary dressmaking. A dress in general, 
whether for man or woman, was denoted by etach or 
idach, and sometimes by dillat : and there were, as at 
present, professional dressmakers — always women — called 
etidach, in modern Irish eadaigheach [aideeach], a word 
derived from etach. The old Irish dressmakers were 
accomplished workers. The sewing on ancient articles 
of dress found from time to time is generally very 
neat and uniform, like that on the fur cape mentioned at 
p. 189, supra, which Mr. Mac Adam describes as " wonder- 
fully beautiful and regular." 

* Br. Laws, v. 381, 383 : see also O'Curry, Man & Cust, II. 112, 117. 
J Kilk, Archseol. Journ., vol. 1., p. 260 : Wilde, Catalogue, 546. 


When women were at needlework, or any such em- 
ployment, they kept their materials in a light wooden 
workbox called a cusal : cusals, says the law-tract, were 
" little crannoges or wooden boxes in which women kept 
their abras or working materials in old times."* 

Embroidery was also practised as a separate art or 
trade by women. The common word for an embroiderer 
was druinech : but another term sometimes used was 
greusaidhe [graissee], which however more usually meant 
a shoemaker. An embroiderer kept for her work, among 
other materials, thread of various colours, as well as silver 
thread, f and a special needle. The 
design or pattern to be embroi- 
dered was drawn and stamped 
beforehand by a designer on a 
piece of leather, which the em- 
broiderer placed lying before her 
and imitated with her needle : or F,G - 337 - 

. . _ . _ . Specimen of antique (prehistoric) Irish 

aS It IS expressed With perfect needlework. A deerskin dress. C(Aer- 

1 1 ,f 1 . 1 iiijf a human body, found, 10 feet deep, 

clearness by the glossator, who, in«bo*inGaiway, ima*. w»wwed 

„____-___ 4. i„_ __ j-u _ __, ,_ „r 11 together in this manner with fine gut. 

commenting on the name of the st * ches very regula r an u,™^ s e , 
article (" the pattern of her needle- !" c,ric ' s f des R cr f ion J" £"j" ™£ 

V ± Journ. for 182s, p. 433. (From Wilde s 

work ") given in the Senchus Mor, catalogue. P . 277) 
says : — " She [the embroiderer] can the more easily perform 
" her handiwork by having the leather pattern before her 
" with the picture of the needlework upon it."| This curious 
and interesting record indicates the refinement, carefulness, 
and artistic skill of the old Irish embroiderers. This art 
of stamping designs on leather, for other purposes as 
well as for embroidery, was carried to great perfection 
as we know from the beautiful specimens of book covers 
preserved in our museums (see vol. 1., pp. 32, 488). 

* Br. Laws, in O'Curry, Man' & Cust., 11. 117. 

f Br. Laws, I. 151, u . 153, 7 from bottom ; and v. 315, top : see also 
Man. & Cust., 11. 119, top. 

I Bt. Laws, 1. 151, „ . 153, 3 o . see also Man. & Cust., 11. 117. 


It was usual for the most eminent of the Irish saints 
to have one or more embroiderers in their households, 
whose chief employment was the making and ornamen- 
tation of church robes and vestments. St. Patrick kept 
three constantly at work, namely, Lupait (his sister) ; 
Cruimtheris, a lady of royal birth ; and Erca, the daughter 
of Dare, the chief who granted Armagh to St. Patrick. 
St. Columkille in like manner had a special embroiderer, 
namely, St. Ercnait or Coca, from whom is named Kilcock 
(Coca's Church) in the County Kildare. She is described 
in a note in the " Feilire of Aengus " as " the embroideress, 
" cutter, and sewer of clothes to St. Columkille and his 
" disciples."* 

Embroidery was practised in Ireland in pre-Christian 
times, and was a well-recognised art from the earliest 
period of legend : for we are told in the Dinnsenchus that 
Aengus the Firbolg, who gave name to Dun Aenguis on 
the great island of Aran, had a daughter Maistiu, who 
was embroideress to the famous Dedannan chief Aengus 
of the Bruga : " she was the first person that formed the 
" figure of a cross in Erin, [namely] on the breast border 
" of Aengus's tunic."f From her also was named the 
historic fort of Mullach-Maistenn or Mullamast near 
Athy in Kildare — the ' summit of Maistiu.' We know 
from many ancient authorities that Irish ladies of the 
highest rank practised needlework and embroidery as an 
accomplishment and recreation. For this purpose they 
spun ornamental thread ; and in the Brehon Laws the 
distaff is constantly spoken of as among the articles in 
the possession of ladies. In the " Feast of Bricriu " (p. 83), 
it is casually mentioned, and as a matter of course, that 
the wives of the great heroes had their needles at the 
feast, and brought them about with them, no doubt with 

* O'Curry, Man. & Cust., II. 123. St. Ercnait is commemorated in 
the Feilire on the 8th Jan. ; but the above note, quoted by O'Curry, is 
not in Stokes's Feilire under that date. f O'Curry, Man. & Cust., 11. 122. 


other articles, in the little bag mentioned below. When 
Cuculainn came to the house of Fergall Monach at Lusk 
to woo his daughter Emer, he found the young lady on 
the lawn before the house with her foster-sisters, whom 
she was instructing in needlework and embroidery.* 

Ladies' ornamental handbag. — Ladies carried a little 
ornamental handbag, or workbag, called iadach or tiag, 
or more usually ciorbholg [keerwolg], ' comb-bag,' like the 
modern reticule, which contained certain choice articles of 
daily use, and which was closed at the mouth by a string. 
The handbag of a queen or of a chieftain's wife contained, 
among other things, a minn or diadem of gold, a lann or 
thin band of gold (for the forehead or neck), a veil, a silk 
handkerchief, needles, and thread both woollen and silver 
for embroidery, f 

5. Tanning. 

The art of tanning leather was well understood in 
ancient Ireland. The name for a tanner was sudaire 
[soodera], which is still a living word. Oak bark was 
employed in tanning, and in connexion with this use was 
called coirtech [curtagh : Lat. cortex], as we find the word 
used in the Laws : whence comes the verb coirtighim, I tan. 
It is laid down in the Law (iv. 149) that the penalty for 
stripping as much bark from another person's oak-tree as 
would tan a cow-hide was a pair of women's shoes worth 
half a screpall, and for as much as would tan an ox-hide 
a pair of men's shoes worth a screpall. A distinction is 
also made as to the amount of the circumference of the 
tree that is stripped : and whether the bark had been 
taken off in the " killing months " or in the " non-killing 
months." In the Irish Life of St. Columkille it is stated 
that at Kells there was an oak-tree which was greatly 
revered, because the saint had at one time lived under it : 

* Courtship of Emer, 71 : Man. & Cust., 11. 122. 
f Man. & Cust., II. 113, 114, quoting Br. Laws. 


but it was blown down by a storm. "And a certain 
man " — says the narrative — " took some of its bark to tan 
" [leather for] his shoes : but when he put on the shoes, he 
" was smitten with leprosy from sole to crown " — in punish- 
ment for the desecration.* 

By the process of tanning, the hide was thickened and 
hardened, as will be seen from the passages quoted below. 
The tanned leather was of a reddish or reddish-brown 
colour, as we find by several old passages. In the " Voyage 
of Maildune," the thieving cook calls his boat curuch nua 
co n-derg codail, " a new curragh [covered] with red hide," 
i.e. " tanned hide " (codal, a hide ; derg, red). Teigue the 
son of Cian made a " large curragh which took to cover 
it forty ox-hides of hard bark-soaked red leather " (do 
dhoinn-lethar chruaid choirtigthe : donn, a reddish-brown). f 
Hence also in the Latin Life of St. Brendan, his vessel is 
described as covered with cow-hide rubricatis in cortice 
roborina, ' tanned in oak-bark,' where rubricatis, though 
signifying ' tanned,' literally means ' reddened. 'J 

6. Workers in Leather, and the articles they made. 

Tanned leather was used for various purposes, one of 
the principal being as material for shoes ; but we know 
that shoes were also made of untanned hide (see p. 216, 
supra). Curraghs or wicker-boats were often covered with 
leather (see below pages 423 and 424). A jacket of hard, 
tough, tanned leather was sometimes worn in battle as a 
protecting corselet : and in connexion with this use, one of 
the oldest references to leather — in the lay literature — 
occurs in the Book of the Dun Cow, in the story of the 
Demon Chariot of Cuculainn. The hero is described as 
placing around him " his champion battle-girdle outside 
"of [a jacket of] hard, tanned, smooth leather of the 
" shoulder of seven ox-hides of yearling heifers, so that it 

* Stokes, Lives of SS., 176. f Silva Gad., 386, a7; Irish Text, 343, 26 . 

% Navig. St. Brend. Card. Moran, 90 : O'Donohue, 119. 


" extended from the waist of his side to his armpit. It 
" [the jacket] was put about him to repel lances and 
" sword-points, and spikes, and spears, and darts : because 
" they used to fly off him the same as if they had been 
" shot against a rock."* The word lethar in the Irish of 
this extract, which is still the word for leather, is of 
course cognate with the English word. 

Bags made of leather, and often of undressed skins, 
were pretty generally used to hold liquids : a practice 
which is alluded to in Cormac's Glossary (p. 104) in the 
explanation of lesan as "a [leathern] bolg or bag in 
which ale is kept." Adamnan (p. 155) relates that one 
of Columba's disciples, preparing for a voyage from Iona 
to Ireland, took among other things, a leather milk-bag 
(utrem lactarium) to bring a supply of milk in his boat : 
but before using it he put it to steep for a night in the 
salt water at the strand to soften it, placing some large 
stones on it to prevent the tide current from carrying 
it away. In the Latin narrative of the " Voyage of 
St. Brendan " such bags are often mentioned. On one 
occasion he directs his crew to bring on board a number 
of skin-vessels (utres : sing, uter) filled with water. A 
leathern bottle was commonly called in Irish petit [pot]. 
Maildune and his companions, when leaving a certain 
island, put a quantity of ale into fiaits and brought them 
to their curraghf : and in an old Irish translation or para- 
phrase of 1 Kings xxv. 18, we read, " The women gave him 
five sheep, two hundred loaves, and two paits of wine."J 
But fait is also used to denote a pot of any kind. There 
was a sort of leather wallet or bag called a crioll, used 
like a modern travelling-bag to hold clothes and other 
soft articles.§ In Brocan's Hymn occurs the expression 
dobert dillat i crioll, " he put a garment in a crioll." 

* Kilk. Archaeol. Journ., 1870-1, p. 426: LU, 79, a, 36 . 

f LU, 23, b, l and a . J Sullivan, Introduction, 358 

§ Silva Gad., 75, s Irish text, 71, ^ 



The parts of every article made of leather were joined 
together by stitching with thongs. A maker of leather 
bags was called criollaidhe [creelee] (sometimes written 
cliaraidhe) from crioll, " a leather bag stitched with 
thongs" ("Man. & Cust.," II. 117). A leather-bottle 
maker was most commonly called a pattaire [pottera], i.e. 
a maker of paits or bottles : sometimes also called sutaire 
or sudaire* We have seen that a maker of cuarans or un- 
tanned-leather shoes was called cuardnaighe [coorawnee]. 
The usual name for a shoemaker was, and is still, greu- 
saidhe [graissee] : but an older name was cairem, gen. 
caireman, pi. cairemain.] O'Clery explains cairemhain 
by greusaighthe, shoemakers. But cairemain was applied 
to a maker of leather bottles, as in the Brehon Law, v. 
106, 17 ; where also in lines 20, 21, the word is explained 
as people " who properly sew the round bottles." The 
word sutaire or sudaire [3-syll.], cognate with Lat. sutor, 
was sometimes applied to a shoemaker : but it more usually 
signified a tanner. From the preceding it will be seen 
that the several terms applied to leather-workers of 
different kinds were a good deal interchanged one with 
another. Those tradesmen in leather-work who stitched 
with thongs, namely, the leather-bottle maker, the shoe- 
maker, and the leather-wallet maker, worked with a pair 
of thongs, forming a stitch with each alternately, the 
workman, while using the free end of one, holding the 
end of the other between his teeth : exactly like the 
Egyptian shoemakers as they are depicted in stone and 
brick records. All this we know from some details given 
incidentally in a passage of the Brehon Laws (v. 81, top, 
and line 18). 

The artistic uses of leather in making covers for books 
and embroidery patterns have been already mentioned. 

* Br. Laws, v. 80, top line. f Rev - Celt., III. 97. 

Sculpture on Chancel Arch, Monastery Church, Glendalough. 
(From Fetrie's Round Towers, 1845 ) 



Section i. Length and Area. 

)IKE other ancient peoples, the Irish fixed their 
standards of length-measures, for want of 
better, mostly, but not exclusively, with 
reference to parts of the human body. 
The troigid [tro-id] or foot was the length 
of a man's foot, which was counted equal 
to twelve ordlachs — thumb-measures or 
inches : ord or ordu, a thumb, now ordog : 
so that this troigid was practically the same as the 
present English foot. It was so constantly mentioned 
that it may be considered as the unit for all moderate 
measurements. Sometimes the space was measured out 
by the actual length of the person's foot : Conall, the 
son of Niall of the Nine Hostages, measures the site of 
a church for St. Patrick, " sixty of his [own] feet " in 

The following table of long measures, which is given 
in the Book of Aicill,f may be taken as the one in most 
general use. The grain, i.e. the length of a grain of 

* Trip. Life, 71. 

f Br. Laws, m. 335. 



wheat of average size, was the smallest measure used by 
the Irish : — 

3 grains, 

4 inches, . . 
3 palms, 

12 feet, 

12 rods or fertachs, 

12 forrachs in length by\ 

6 forrachs in width / 

I ordlach or inch, 
i bas, palm, or hand. 
i troighid or foot. 
i fertach or rod. 
i forrach. 
i tircumaile (i.e. 
' cumal-land '). 

According to this table a tir-cumaile [teer-cummala] 
was equal to a space 576 English yards long by 288 
broad : that is, about 34^ English acres. A cumal repre- 
sented three cows (p. 385, infra) ; and a tir-cumaile (land 
for a cumal) was as much land as was considered sufficient 
to graze three cows. This almost exactly agrees with the 
statement in vol. 1. p. 40 — from a different source — that 
a ballybetagh (which contained 3600 English acres) was 
allowed for 300 cows : one of the many illustrations of the 
general consistency and accuracy of the old Irish records. 
When English ideas and practices began to obtain a 
footing in Ireland, after the Anglo-Norman Invasion, 
various other measures of land were adopted, the most 
general of which was the acre. Land was commonly 
estimated in acres and ploughlands according to the 
following table* : — 

120 acres, i seisrech or ploughland. 

12 ploughlands, . . i baile, bally, or townland. 
30 bailes, i tuath or triucha. 

As all Ireland contained 184 tuaths (vol. I., p. 40), this 
gives the (old) Irish acreage of the whole country as 
184 x 30 x 12 x 120 = 7,948,800. There are, we know, 
20,815,460 English acres in all Ireland, which gives the 
old Irish acre a little more than r z\ of the present English 

* Comyn's Keating, 113. 


acre : exactly bearing out Keating's words : — " The acre 
" of the measure of the Gaels is twice or thrice greater than 
" the acre of the division of the Galls or English now."* 
All this is on the supposition that in the old estimate the 
whole surface was included, waste as well as arable land. 
There is some vagueness in all these calculations, which 
may account for the fact that within recent times the 
Irish acre is more than once and a half the size of an 
English acre. 

Various other length-measures were in use in ancient 
Ireland. A ceim [kaim] or step was 2| feet : " the lawful 
step is two feet and a half," says the Gloss on the Law 
(iv. 215), which gives the full pace (deis-ceim : pron. desh- 
kaim), 5 feet. But in another law-tract on the " Division 
of Land," the full pace is given as 6 feet, making the single 
step 3 feet. 

For small measures the has [boss] and the dorn [durn] 
were in constant use. The has or " palm " was the width of 
the hand at the roots of the fingers, which was fixed at 
4 inches. The dorn or ' fist,' with the thumb closed in 
(called mail-dom, ' bare-fist '), was 5 inches : with the thumb 
extended (called airtem-fist) , 6 inches. f We constantly 
meet with such measures as " a cow 20 fists in girth," " a 
spear-handle 12 fists in length." 

In a part of the Law (iv. 77) relating to fences and 
their measurement, a foot differing from the tabular one 
given above is mentioned : not the length of the whole 
foot, but as far as the separation of the big toe. This foot 
was considered as 10 inches : but it was rarely used. 

* Joyce's Keating, p. 37. For the political subdivisions of Ireland, see 
vol. 1., pp. 39, 40, supra : and for various other modern land measures, see 
Ware, Antiqq., 224 ; Sullivan's Introduction to O'Curry, 96 ; Reeves' 
paper " On the Townland Distribution of Ireland," Proc. Roy. Ir. Acad., 
vii. 473 : " On the Territorial Divisions of the Country," by Sir Thomas 
Larcom, prefixed to the " Relief Correspondence of the Commissioners of 
Public Works " : and Joyce, Irish Names of Places, 1. 241. 

f Br. Laws, 11. 241. 


Lengths and distances were often roughly indicated by 
sight and sound : a custom that prevailed among nearly 
all ancient peoples. A chief named Coirbre promised 
Cuangus all the territory he could see to the north of 
Sliab Cise, near Assaroe at Ballyshannon, as a reward 
for expelling St. Patrick* : and many other examples of 
this sight-measurement might be given. But distances 
were much oftener estimated by sound. In connexion 
with the law of distress certain distances, called in the 
Senchus M6r " magh-spaces," were made use of : and the 
old commentator defines a magh-space to be " as far as 
" the sound of the bell (i.e. the small handbell of those 
" times] or the crow of a barn-door cock could be heard. "f 
In the " Second Vision of Adamnan " it is stated with 
regard to a certain church that " neither the saints nor the 
" angels come nearer to it than where one hears the voice 
" of a bell that is struck at the church."J 

A man felling a tree was " bound by law to give 
warning as far as his voice could reach," so as to avoid 
danger to cattle or people.§ In some places these old 
measures are remembered in tradition to the present day. 
In the parish of Termonmaguirk in Tyrone there is an 
old burial-ground called Relig-na-man (Irish Reilig-na- 
mbari), the ' cemetery of the women,' where none but 
women are buried. It is about half a mile from the 
church-ruin of Termonmaguirk, and the people of the 
place give this traditional account of its foundation. 
The body of a certain woman of bad character was 
brought to be buried in the church of Termonmaguirk ; 
but St. Columkille forbade it, and directed that the body 
should be buried at a spot where the sound of a bell 
struck at the church began to go out of hearing : and 
he left an injunction that this new cemetery (now Relig- 

* Trip. Life, 149. J Rev. Celt., xn. 425. 

f Br. Laws, n. 107, 109. § Br. Laws, ill. 227. 


na-man) should never be entered by a living woman or by 
a dead man.* 

The crow of a cock and the sound of a bell, as distance 
measures, are very often met with. The " glockenklang " 
or bell-clang was also used by the ancient Germans to 
measure distances. f 

Other vague modes of estimating lengths were used. 
A certain legal distance is laid down in the Law (iv. 139) 
as being as far as a youth could cast a rod. The legal 
size of the faithche [faha] or green round a house depended 
on the rank of the owner, and the unit of measure was the 
distance a man could cast a spear standing at the house 
(p. 61, supra). Very often the human face is taken as 
the standard of size of a ring or crescent of gold — or 
silver — to be given as a tribute, or fine, or present. In 
the Battle of Rossnaree (25) it is stated that the Clanna 
Dedad of Munster proposed to give to Concobar, king 
of Emain, among other valuable things, " the breadth of 
his face of red gold," as an inducement to refrain from 
invading them. Many other like instances of this standard 
might be cited. We may form some vague idea of the 
value of such a ring or crescent from an expression in 
an ancient poem quoted in Cormac's Glossary (p. no) : — 
" Seven ounces of refined gold for my great friend's noble 
face. "J So in the Welsh tale of Bran wen the daughter of 
Llyr, Bendigeid Vran offers the offended hero Matholwch 
a plate of gold of the breadth of his face to appease his 

2. Capacity. 

The standard unit of capacity adopted by the Irish was 
the full of a hen-eggshell of moderate size, which perhaps 
was as good a standard as could be found at the time. 

* Reeves, Adamnan, 283. 

f Stokes (in Rev. Celt., xn. 440) refers for the use of the glockenklang 
to J. Grimm, Deutsche Rechtsalterthumer, 2te Ausg. 76. 

% See other instances in Mesca Ulad, 55 ; and Book of Rights, 243. 
§ Mabinogion, 30. 


Beginning with this there is given in the Book of Aicill* 
the following table of measures of capacity : — 

12 hen-eggshell-fulls, = i meisrin. 

12 meisrins, . . = i ollderbh. 

12 ollderbhs, . . = i olpatrick or oilmedach. 

Another measure, the olfeine, is half an olpatrick : but one 
of the law-tractsf gives an olfeine as two olpatricks. I 
find by actual trial that twelve times the full of an average- 
size hen-eggshell will fill a modern imperial pint ; so that 
a meisrin was equivalent to a pint, an ollderbh to 12, 
and an olpatrick or oilmedach to 144 pints. But there 
seems reason to think that the olpatrick was sometimes 
reckoned as one-fourth of this size, namely, as containing 
36 pints. 

A sellann, equal to four eggshells, was often used in 
measuring honey : it occurs frequently in the Rule of the 
Culdees : in which also a bochtan, equal to 12 eggshells, is 
mentioned as a measure for ale, milk, or whey. J 

A miach or sack was much used in measuring corn 
and malt : and fines for trespass were estimated, and 
payments of various kinds were made, in sacks, so that 
the miach must have been always much of the same 
size. As a standard of value it will be mentioned at 
page 386. 

As there were vague measures in length, so also in 
capacity. In measuring honey in large quantities four 
sizes of vessels are often mentioned. A " milch-cow 
vessel," or barrel, was one which, when full, a person of 
ordinary strength could lift as high as his knee : a " heifer- 
vessel," which he could raise to his navel ; a smaller 
" heifer- vessel " to his loins ; and a " dairt (or still smaller) 
heifer- vessel," which he could raise over his head.§ 

* Br. Laws, in. 335, bottom. f Br. Laws, in. 337, note i. 

J Reeves, Culd., 84, 85 : and Corm. Gloss., 134, under " Pinginn." 
§ Br. Laws, iv. 165, note 2. 


3. Weight. 

The smallest weight used was a grain of wheat. We 
read indeed, in an ancient Irish passage in the Book of 
Ballymote quoted by Petrie (R. Towers, 218), that there 
was a smaller weight called an atom : 24 atoms in a grain : 
but this is evidently fanciful, like some of the minute time 
divisions (p. 387, infra). An attempt was made to render 
the grain-standard definite and uniform by these two 
regulations : — First, the grains should be taken from wheat 
that grew in " land of three roots," i.e. land of the best 
kind (p. 269, supra). Secondly, the grains should be in a 
medium condition as to dryness. The following is the 
table of weight founded on the average grain of wheat* : — 

8 grains, 1 pinginn or penny of silver. 

3 pinginns, . . 1 screpall. 

24 screpalls, . . 1 unga or ounce. 

The pinginn and screpall will be again under notice in 
section 4 (p. 381). 

The unga or ounce (576 grains of wheat or about 432 
grains Troy) was the standard used in weighing metals. The 
word seems to have been borrowed from the Latin uncia. 
That the Irish did not borrow the standard itself, but had 
it from the most ancient times, appears from the fact that 
there was an older native word mann for the ounce. In 
the ninth century the word unga had come into general 
use, and mann had become obsolete, so that Cormac 
thought it necessary to explain it in his Glossary : — 
" Mann, that is unga or ounce." A verse is then quoted 
from Sencha, a celebrated law-giver and poet of far remote 
time, to show its application : and the Glossary adds : — 
" Mann then is ' bright,' that is, a refined ounce."f 

♦Petrie, Round Towers, 218, top : Corm. Gloss., 134, under "Pfssfre." 
Prof. Ridgeway reckons that four of those wheat-grains were equal to three 
grains Troy. f Corm. Gloss., no : Irish Text in three Irish Gloss., 29. 


There was a weight called dirna, of which the exact 
value is not known. It was very much greater than the 
ounce, as we know from the poem of Colman Mac Lenine, 
in which an ounce is contrasted with the much heavier 
dirna.* From an old passage quoted in Cormac's Glossary 
(72, " Fir "), it would seem that a dirna of silver was the 
value of a white cow. In Petrie's R. Towers (p. 219) is a 
quotation from the Brehon Laws in which a weight called 
a dinnra is mentioned, and it is stated that a dinnra of 
red bronze contains six ounces. Probably dirna and 
dinnra are the same : one being changed to the other 
by metathesis. 

The pound weight was used, and was designated by 
the word pun, which is probably a loan-word from Lat. 
pondus. From a passage in the Story of Mongan in the 
Book of the Dun Cow, it would appear that a pound of 
silver had twelve ounces and a pound of gold nine. Said 
Mongan to the poor scholar : — " Go now till you reach the 
" sith [shee] of Lethed Oidni, and bring me a precious 
" stone which I have there : and take for thyself a pound 
" of white silver [pun findairgii] in which are twelve 
** ounces . . . thou wilt [also] find a pound of gold, in 
" which are nine ounces."f But perhaps pun here means 
merely a mass or lump. 

From numerous references in the old writings, we learn 
that the ancient Irish had balances of different kinds and 
sizes, and with different names. The most usual Irish 
term for a balance, and also for the beam of a balance, was 
med or meadh [ma], which is the word in general use at the 
present day. Cormac's Glossary (p. 134) explains the word 
puincern [punkern] as meaning two things : — First, a cern 
or dish for measuring a commodity called sella (probably 
some kind of corn) : Keating has a " a cern of arbhar or 
corn " (p. 78, supra) : second, " a beam for weighing cattle, 

* Corm. Gloss., 10, u : O'Curry, Man. & Cust., II. 245. 
t Voyage of Bran., 1 55, 95 


namely, the notched beam." Here the author plainly 
implies a distinction between a plain beam and one that 
was notched (indmeach) ; and he takes care to explain 
that the beam called puincem used for weighing cattle 
was a notched one. But it appears from other passages 
in the glossary that there were smaller notched beams 
for weighing lighter commodities. Thus puingcne (p. 134) 
is explained " a screpall or scruple of the notched beam " : 
and again under " cimb " (p. 39) this quotation is given 
from the Bretha Nemed : — " A cimb or tribute of bronze 
since I placed the bronze in the notched balance." I take 
it that the balance with a " notched beam " (tned indmeach) 
was a steelyard — a balance having a single weight movable 

Fig. 328 « 

The small steelyard found in use by Thomas Dineley in the seventeenth century. 
(From Kilk. Archaeol. Journ., 1858-9, p. 56.) 

along a graduated beam from notch to notch, which by its 
distance from the fulcrum or suspension point indicated 
the weight of the commodity — identical with our modern 
steelyard. As bearing upon this point it is well to observe 
that an old steelyard of bronze was found in 1864 in 
a rath near Ballyshannon in Donegal, ornamented and 
carefully graduated : the material — bronze — indicating 
great antiquity.* Thomas Dineley, an Englishman, 
travelling in Ireland in the seventeenth century, found 
a " stilyard " in use for weighing foreign coins, of which 
he gives an illustration (reproduced here, figure 328) : 
probably a descendant of the ancient puincem. 

Another balance, which must have been small, is 
noticed in Cormac's Glossary (p. 134) in these words :— 

* Proc. Roy. Ir. Academy, vin. p. 476. 


" Pissire [peeshere], that is, a broad-headed beam for 
" weighing one pinginn of weight. One pinginn then is the 
" burden of that beam." From the epithet ' broad-headed ' 
we may conclude that it was a steelyard. Pissire is derived 
from piss [peesh], an old name for the pinginn or penn}^ 
Still another kind of balance was called laithe [lay-he], 
always used in this plural form (sing, laith). It is 
explained in Cormac's Glossary, but more fully by 
O'Clery in his Glossary : — " Laithe, a balance for weigh- 
ing gold or silver." From an expression in Cormac's 
explanation (etir laithe Lugba, ' between the scales of 
Lugba '), as well as from the fact that the plural laithe 
is always used (like our " scales "), we may infer that the 
balance so designated was the ordinary scales of two 
dishes. We know that the Irish had balances of this 
kind ; for on one of the Monasterboice crosses there is 
a representation of the general judgment in which the 
Archangel Michael is seen weighing souls in a two-dish 
balance : the dishes being deep like bowls. A small 
bronze balance of this kind — now in the National 
Museum — beautifully finished, was found in i860 in an 
excavation at Kilmainham near Dublin : and another 
about the same time in a crannoge in Ulster.* The 
Kilmainham one however is probably Danish. 

4. Standards of Value and Mediums of Exchange. 

In early stages of society in Ireland, as in all other 
countries, buying and selling and other commercial 
transactions were carried on by means of payment in 
kind : and there is hardly any description of valuable 
articles that was not used for this purpose. It will be 
seen in many parts of this book that payments were made 
for purchases, tribute, fines, &c, in cows, sacks of corn, 
salted pigs, butter, mantles, and so-forth : the parties 

* Proc. Roy. Ir. Academy, vn., p. 156, „, and p. 368. 


determining the values according to the customs of the 
place. But mixed up with this barter in kind, gold and 
silver told out by weight, and — after the middle of the 
eighth century — silver coins, were used as mediums of 

That the Irish were acquainted with the use of coined 
money, at least as early as the eighth century, is proved 
by the records ; and indeed might be anticipated without 
the help of records, inasmuch as there was in those times 
much intercourse, both by traffic and missionary enterprise, 
between Ireland and the Continent, where coined money 
was then in constant circulation. A celebrated Irish poet 
named Rumann, who died a.d. 747, once paid a cuairt or 
professional visitation (see vol. I., p. 449, supra) to the 
Galls or foreigners of Dublin, and composed a poem for 
them. They at first refused to give him anything, but 
ultimately agreed that he should name his own reward 
whereupon he demanded two pinginns from every good 
Gall and one from every bad Gall. The result was that 
to a man they gave him two pinginns each.* Half a 
century later we have another record indicating a familiar 
acquaintance in Ireland with the use of coined money. 
There is extant a letter written about the year 790 by the 
illustrious churchman and scholar, Alcuin, to Colcu the 
Wise, head of the great school of Clonmacnoise, stating 
that he had sent, with the letter, a number of sicli (small 
coins), some from himself and some from his great master 
Charlemagne, and a quantity of pure olive oil (then scarce 
in Ireland) for use in religious rites. 

The coins in circulation among the Irish were the 
pinginn and the screpall (or, as it was often called, the steal), 
both of silver : according to the authorities quoted by 
Petrie (R. Towers, 218, 219), the pinginn weighed 8 grains 
of wheat, and the screpall was equal to 3 pinginns and 
weighed accordingly 24 grains ( = 18 grs. Troy). It is 

* Petrie, Round Towers, 353. 


curious that in another and older authority, Cormac's 
Glossary (134, under " Pisire "), the weight of the pinginn 
is given as 7 grains : which Petrie (R. T., 220) conjectures 
may mean that while 8 grains was the normal weight, the 
pinginn then in circulation usually weighed only 7, on 
account of wear. According to M. d'Arbois de Jubainville, 
the screpall is mentioned in one of the St. Gall eighth- 
century glosses : which is the oldest Irish notice of it yet 

The two words screpall and sical, both meaning the 
same coin, are borrowed from the Latin scrupulus and 
siclus (this last being itself borrowed from the Hebrew 
shekel). In Professor Ridgeway's opinion they were 

Fig. 339. Ik;. 310. 

Irish bracteatc coins : now in the National Museum, Dublin. 

(From Petrie, Round Towers, p. 278.) 

borrowed from the Latin before the time of Constantine, 
i.e., before the beginning of the fourth century a.d. But 
the Irish had more than one native name for the screpall, 
which we find in various forms in the different authorities : 
puingene, opuingc, oiffing, faing, fang. Cormac's Glossary 
(p. 134) gives the native word piss [peesh] as another name 
for a pinginn-weight. 

Many specimens of the pinginn and of the screpall 
are preserved in the National Museum. The pinginns are 
what are called " bracteate " coins, i.e. struck only on one 
side ; but the screpalls are impressed on both sides. 

We have seen that the Irish were familiar with the use 
of coins in the eighth and ninth centuries ; and the question 

* Rev. Celt., xvni., p. 114. 


has often been debated whether they coined money for 
themselves before the tenth century. It would be certainly 
strange if they did not, seeing that they had such constant 
intercourse with Britain and the Continent in the early 
centuries of Christianity ; and they were — as we have 
seen — in advance of most nations of Europe in the Middle 
Ages in artistic metal-working. There is one circumstance 
that strongly favours the opinion that they had a native 
mint, namely, that they had a coin with a native name, 
crosdc or crosdg, which — both coin and name — fell out of 
use when the pinginns and screpalls had become well 
established as the Irish currency. The name — which 
signifies " little cross " — indicates that it was stamped on 
the side with the figure of a cross. That this was a very 
ancient native coin, quite independent of pinginns and 
screpalls, is rendered pretty certain by three circum- 
stances : — First, the native name crosdc ; secondly, that it 
fell out of use when the coins with the borrowed names 
" pinginn " and " screpall " came into use ; and, thirdly^ 
and most strongly of all, that in point of value it did not 
fit in with the tabular arrangement of these two last-named 
coins, for, according to the native records, it was equal to 
" two pinginns and a quarter of a pinginn," that is to say, 
it weighed eighteen grains.* But at any rate this matter 
of coinage is comparatively unimportant as affecting 
civilisation, for it has been pointed out that some of the 
greatest nations of antiquity did not coin money, or coined 
it only at a late period of their career. 

From the very beginning of our records gold and silver 
were used as a medium of exchange, sometimes as ingots, 
but more commonly in the form of rings, bracelets, and 
other ornaments. They were weighed by the ounce, f 
which, as we have seen, was equal in weight to 576 grains 
of wheat, or to 432 grains Troy ; and there is the best 

* See " Cros6c " in O'Donovan's Supplement to O'Reilly ; and Br. Laws 
v. 437, f For instances see Irish Miscell., 1846, pp. 133, 143, 147. 


reason to believe that, in order to facilitate interchange of 
this kind, gold and silver rings of various forms, as well as 
other gold and silver ornaments, were generally or always 
made of definite weights. Notices of this custom are 
found everywhere in the literature. In Cormac's Glossary 
(p. 22), a briar is defined as " a pin of one ounce of gold." 
At a.d. 1 150 the Four Masters record that Murkertagh 
O'Loghlin, king of Ireland, gave the abbot of Deny, 
among other presents, " a gold ring [fail 6ir) in which 
were five ungas " : in the next year (1151) Cu-Uladh 
O'Flynn gave the same abbot " a ring of gold in which 
were two ounces " : and Turlogh 0' Conor king of Ireland 
gave the archbishop of Armagh " a gold ring of twenty 
ounces." In an ancient document in the Book of Armagh 
certain payments are made for a purchase, among them a 
muince or necklet of three ounces of gold.* St. Finnen 
once found a gold ring, and gave it to a chief as the price 
of a certain slave's freedom, for which an ounce of gold 
had been demanded : the ring was weighed, and it was 
found to contain exactly an ounce, f Such examples might 
be multiplied indefinitely. That this custom existed in 
Ireland is rendered all the more certain by Caesar's record 
that in his time the people of Britain " used brass or iron 
rings fixed at a certain weight as their money. "J But in 
Ireland, gold, as being comparatively abundant, was used 
instead of the inferior metals. What is even more to the 
point, the practice seems to have been universal in other 
countries : — " I have already shown " — says Professor 
Ridgeway (p. 399) — " the universality all over the world 
of making gold ornaments after a fixed weight. "§ 

It may be considered certain that in Ireland the open 
gold rings called bunne-do-at (now often called fibula : 

* Trip. Life, 341. t Stokes, Lives of SS., 225, bottom. 

J Commentaries, v. xv. 

§ See also on all this M. de Jubainville in Revue Archaeologique, 1888, 
on " Des Bijoux et de l'argenterie employes comme prix d'achat." 


see p. 241, supra), as well as other gold ornamental 
articles, were used as money. But besides those called 
bunne-do-at, there are in the National Museum, a great 
number — fifty or more — of very small open gold rings, from 
i to f inch in diameter, without the terminal knobs or ats : 
they are bunnes simply, not bunne-do-ats. One is figured 
here, its natural size, and another is shown in the upper 
part of fig. 21, vol. 1., p. 33, supra. These, from their great 
numbers, and from their simple, unornamental construc- 
tion, have all the appearance of having been used mainly 
as currency. Professor Ridgeway has carefully investigated 
this question in his work " The Origin of Metallic Currency 
and Weight Standards," and he has fixed on 
the weight of the smallest specimen in the 
museum — 15 grains Troy — as the standard. 
He shows moreover that all the larger rings 
in the Museum are very nearly multiples of fig. 331. 

. _,, «... . - . Gold bunnl or 

this. These little rings then were used in rin*. tun **»: open. 

,. . . . , . but without the do- 

ordinary business transactions, as we use «*. used as money. 
coins now. As the crosoc weighed 18 grains M^um!" 6 ™ 
of wheat, i.e. 13.5 grains Troy, it is likely JJJJ ST™* 
that it was intended to be in accordance with 
this standard : in other words, that the smallest of these 
little rings represented the value of a crosoc. Gold rings 
offered in payment were tested, as we see in case of the 
ounce ring cited above, paid for the slave's freedom : just 
as they now weigh gold coins in banks. 

A full-grown cow, or ox, was in ancient times a very 
general standard of value, not only in Ireland, but all over 
the civilised world : and was considered equal in value to 
one ounce of gold. In this case — as an article of pay- 
ment — a cow was in Ireland generally called a sed [shade]. 
Cows or seds were very often used both in actual payments 
and in estimating amounts. Next above the sed was the 
cumal, which was originally applied to a bondmaid : but 
the word came to be used very generally to signify the 



value of a bondmaid, which was counted as three seds. 
The words sid and cumal are however sometimes used 
very loosely to denote variable values. Thus in one of 
the Law Glosses it is stated that the best sid is a milch 
cow, and the worst sed a dartaid or yearling heifer* : and 
in Cormac's Glossary (p. 29), under the word clithar-sit, 
there is a sort of classification of sids. So also the cumal : 
in a certain Law Gloss a cumal of six cows is mentioned.! 
But the text generally draws attention to exceptional cases 
of this kind : and in all ordinary statements of value in 
these standards, a sed may be taken as a cow, and a cumal 
as three seds. I 

For general convenience it was laid down that where 
the payment for anything was half a cumal or less, it might 
be legally made in one kind of goods — cows, or horses, or 
silver : from half a cumal to a cumal, it should be in two 
kinds : above a cumal, in three. Whenever horned cattle 
were given in payment, one-third of them should be oxen ; 
when horses, one-third should be mares ; and silver payment 
should include one-third of manufactured articles. But under 
mutual agreement payments might be made in any way.§ 

A miach or sack of corn — generally of oats or barley — 
which for convenience sake must have been always made 
of uniform size — was very often used as a standard of 
value : it is indeed adopted in the Brehon Law as the 
almost universal standard in estimating fines for trespass, 
and payments for grazing. || Thus for trespass over a full 
fence there was a fine of four miachs of oats or barley : the 
price that purchased the grazing of certain lands is twelve 
miachs : and the expense of feeding cattle under certain 
circumstances is a miach for every animal per month. ^ 

* Br. Laws, iv. 29. See also 11. 277, bottom ; and ill. 43, bottom, 
t Ibid., iv. 25, „. 

X See " Sed " in the Index to vol. v. Brehon Laws. 
§ See for all these arrangements, Br. Laws, ill. 151, 153. 
|| Br. Laws, iv., all through the tract on Judgments of Co-Tenancy, 
p. 69. If Br. Laws, iv., pp. 83, 105, 107. 



We have no means of ascertaining the exact contents of 
a miach : but we know its value ; for it is stated several 
times in the Brehon Law that a miach was worth a 
screpall of silver.* A miach or sack is often mentioned 
as a standard of value, without any intimation as to what 
it contains : but in all such cases it is to be understood as 
a sack of oats or barley. 

5. Time. 

In the works of some ancient writers who touch on 
technical chronology, such as Bede, Rhabanus, Isidore, &c, 
are to be found subdivisions of time, based on the day in 
the higher parts, but in the lower descending to such 
minuteness as to lead to the conclusion that the smallest 
measures were purely ideal, and never intended for prac- 
tical application. The ancient Irish also had their time 
divisions, with minute denominations, a specimen of which 
is given in the Tale of the Battle of Moyrath (p. 109). 
This may be tabulated as followsf : — 



Equivale'nts in our 

present time 


^ of a second. 


i ostent, 

376 atoms, 

1 min. 36 sec. 

1 bratha, . 

564 „ 

2 „ 24 „ 

1 pars (part), 

1 minuit (minute), 

94° » 
1,410 „ 

4 „ ° „ 
6 „ „ 

1 pongc (point), 

3.525 „ 

15 „ „ 

1 uair (hour), 

14,100 „ 

60 „ „ 

1 cadar (quarter of a day), 


6 hours. 

After this follow a day (called variously in Irish la, Ida, lae, 
lathe, dia, die) ; a week (sechtman) ; a month {mi) ; a season 

* For instance, Br. Laws, i. 61, note i ; and n. 251, 8 . 

t See also Moyrath, p. 331. For another statement of Irish time mea- 
sures, see Stokes, Trip. Life, Introd., cliv. In the above Table the Irish 
atom is sufficiently minute : but the Venerable Bede's smallest measure is 
seven or eight times smaller still, being only the thirtieth part of a second, 


(treimse) ; a year (bliadain) ; a saegal or seculum ; an aeis 
or aeon. As all but one of the Irish words used in the 
first eight items of the above enumeration are borrowed 
from the Latin, we may take for granted that the table 
itself was borrowed from the Latin writers, but probably 
modified. The exception is bratha [braha], a native Irish 
word, meaning a ' twinkling of an eye.' 

The Irish divided their year, in the first instance, into 
two equal parts, each of which was afterwards subdivided 
into two parts or quarters. The four quarters were 
called — Errach, now Earrach [arragh], Spring ; Samrad, 
now Samhradh [sowra], Summer ; Fogmar, now Foghmhar 
[fowar], Autumn ; Gemred, now Geimhridh [gevre], Winter : 
and they began on the first days of February, May, August, 
and November, respectively. We have historical testimony 
that games — which will be described in chapter xxix. — 
were celebrated at the beginning of Summer, Autumn, and 
Winter ; but we have no account of any such celebrations 
at the beginning of Spring. These divisions of the year 
and the festivities by which they were ushered in originated 
with the Pagan Irish, and were continued into Christian 

Errach or Spring began on the first of February. This 
day was called oimelc, imolg, or imbulc : the first form 
oimelc is given in Cormac's Glossary (p. 127, " 6i "), where 
it is derived from 6i, a sheep, and melc or melg, milk : 
" di-melg, ' ewe-milk,' for that is the time the sheep's milk 
comes." That oimelc is the first of February we know 
from Peter O'Connell's Dictionary, where oimelc is identi- 
fied with Feil Brighde (St. Brigit's feast day), which has 
been, and is still, the Irish name for the first of February 
all through Ireland, the old Pagan name oimelc, being 
obsolete for centuries. 

In Cormac's Glossary (p. 151) Samrad, Summer, is 
fancifully derived from the Hebrew sam, the sun, and the 
Irish rad, a course : " the course which the sun runs : then 


most its brightness and its height delight." Whatever may 
be the true derivation, the word is obviously cognate with 
the English Summer. The first day of May was the 
beginning of Summer. It was called Belltaine or Beltene 
[beltina], which is the name for the ist May still always 
used by speakers of Irish ; and it is well known in Scot- 
land, where Beltane has quite taken its place as an English 
word : — 

" Ours is no sapling, chance sown by the fountain. 
Blooming at Beltane, in winter to fade." 

— Lady of the Lake. 

Another name for May Day, according to Cormac's Glos- 
sary (p. 36), is Cedsoman. 

Autumn was called Fogmar, Fogamar, or Foghmhar, 
which is still its name : according to Cormac's Glossary 
(p. 74), Fogamar was also the name of the last month of 
Autumn, i.e. October. Autumn began on the ist August, 
Lammas day. This day has two ancient Irish names : — 
Bron-trogain and Lugnasad [Loo'nasa]. The first is derived 
in an old Irish glossary* from trogan, a name for the earth 
or ground, and bron, bringing forth : — Bron-trogain, the 
bringing forth of fruits by the earth or ground. Bron 
properly signifies ' sorrow ' or distress : the idea here 
being — in the words of the old Tale of the " Wooing of 
Emer "f — " It is then the earth sorrows under [the weight 
of] its fruit " (is and do-broni trogan fua torthib). In the 
Story of the Agallamh or " Colloquy of the Ancients " — as 
well as in the old glossary referred to above — the first 
day of August, or the " trogan-month," is identified with 
Lugnasad ,1 which is still the Irish name of the first of 
August everywhere in Ireland. (See page 439, below 
for the origin of this name.) 

* Quoted by O'Donovan in Book of Rights, liii. 

f Rev. Celt., xi. 443. 

\ Silva Gad., 216, 30 . and Stokes, Acallamh, line 4760 


Samain, Samuin, or Samhuin [so win], the first of 
November, was the first day of Gemred or Winter. The 
name is compounded of the two words, sam, which was an 
old word for Samrad or Summer, and fuin, an ancient 
word for end : that is to say, the end of Summer : " for," 
the old authority adds, " the whole year was [originally] 
" divided into two parts — Summer from ist May to ist 
" November, and Winter from ist November to ist May."* 
The name samain is still used even among the English- 
speaking people in Scotland and the north of Ireland, in 
the form of sowin or sowins, which is the name of a sort of 
flummery usually made about the ist November. The 
term gemred for winter is a derivative from the older and 
simpler word geim, meaning the same thing. 

For certain legal purposes connected with grazing and 
trespass, the ancient Irish had another division of the year 
into two unequal parts : — the Summer division from the 
ist March to the 31st July, five months ; and the Winter 
division from ist August to the 28th of February, seven 
months, f 

O 'Donovan stated in 1847 (Book of Rights, lii) that 
the season with which the Pagan Irish began their year 
could not (then) be determined. Some years later O' Curry 
asserted that according to the authority of an ancient Irish 
poem, of which he had a copy, the year began on the 
ist February. J We must presume that this is correct ; but 
he has not given the stanza in which the statement is 
made, and I have never seen the poem. 

Occasionally time was measured by the fortnight 
(coicthiges : pron. co-keess'). Fothad Airgthech on one 
occasion rested on a certain hill " till the end of three fort- 
nights." King Concobar — says the " Battle of Rosnaree " 
(p. 3) — was ill in Emain " for the time of three fortnights " : 
and in another part of the same Tale (p. 19) Sencha says 

* Sick Bed, Atlantis, i. 370, note 2 : and Book of Rights, liii. 

•j- Br. Laws, iv. 79, 89, 91. J Sick Bed, Atlantis, 1. 370, note 2. 


to Concobar, " I will ask a truce of battle till the distant 
end of a fortnight in addition to a month." 

The ancient Irish counted time rather by nights than 
by days. Thus in the Life of St. Fechin we are told : — 
" Moses was forty nights on Mount Sinai without drink, 
without food."* In coupling together day and night they 
always put the night first : in other words, the night 
belonging to any particular day was the night preceding. 
In the " Vision of Mac Conglinne " a certain thing is spoken 
of as happening on Oidche Domnaig, the " night of Sun- 
day," where it is obvious from the context that the night 
in question was the night preceding, or what we of the 
present day would call " Sunday eve " or " Saturday 
night. "f All this is a survival of what appears to have 
been the universal practice among the Celtic nations of 
old : for Caesar J describes the Gauls as measuring the 
lapse of time, not by days but by nights : and calculating 
months, years, and birthdays in such a way as to make the 
night precede the day. Tacitus§ states that the Germans 
also gave precedence to the night, and the same custom 
prevailed among the Jews. Traces of all this still remain 
in the English language in the words fortnight and 
sennight (i.e. fourteen nights and seven nights), and in 
such words as Christmas-eve and Hallow-eve. In express- 
ing a length of time by nights, the Irish commonly in- 
cluded the two nights at the beginning and end, and 
hence the word coicthigis for a fortnight, which literally 
means " fifteen-night " ; like the Welsh wythnos (" eight- 
night ") for a week. 

The Irish used the word nomaid or nomad to denote 
a time, the length of which has not been precisely deter- 
mined. It evidently means nine time-spaces of some kind, 

* Rev. Celt., xn. 435. On this custom of measuring time by nights, 
see Rhys, Hibbert Lectures, 360. 

t Mac Conglinne, 18, 20, and 134 : see also Adamnan, 230, last line 
but one. 

t Bell. Gall., vi. xviii. § Germania, cap. xi. 


from noi, nine. Some take it to mean ' nine nights,' like 
the Latin nundinutn : and it has been interpreted ' nine 
days,' ' five days and four nights ' ; ' the ninth day.' The 
probability is that its meaning varied, so that two or more 
of these may be correct. 

6. Enumeration. 

The decimal system was in general use. The mode 
of enumeration was usually the same as we have now in 
English, the largest numbers coming first and the smallest 
last. The Four Masters give all their dates in this order : 
and Keating reads 2628 as " two thousand and six hundred 
and twenty-eight." But very often this order was reversed, 
both in the old and in the more recent writings. In an 
ancient poem quoted by Keating, 197 years is given as 
" seven years, ninety and a hundred " : and in another 
passage 1130 is read " thirty on a hundred on a thousand." 
Frequently the two systems are mixed, and other denomi- 
nations besides decimal are brought in, of which the 
following are examples : — (432 years), " twelve years and 
twenty and four hundred " : (1978), " eight and seventy, 
a thousand and nine hundred " : (1130) ships, " ten ships, 
twenty, on a hundred, on a thousand." Sean Buidhe 
O'Clery calls 1453 years " a thousand years and four 
hundred years, and thirteen years, and twice twenty." 

It is remarkable that seven is sometimes called " great 
six." Thus mdir-sheis-ear is found in old authorities to 
denote seven persons, literally " great-six persons." This 
custom as well as the word mdir-sheis-ear [more-hesher] 
still continues in use. 

Sculpture on Window : Cathedral Church, Glendalough : Beranger, 1779. 
(From Pctrie's Round Towers.) 



Section i. Roads, Bridges, and Causeways* 

oads. — That the country was well provided with 
roads we know, partly from our ancient litera- 
ture, and partly from the general use of chariots. 
They were not indeed anything like our present 
hard, smooth roads, but constructed according to 
the knowledge and needs of the period, sometimes laid 
with wood and stone, sometimes not, but always open 
and level enough for car and horse traffic. There were 
five main roads leading from Tara through the country 
in different directions : and numerous roads — all with 
distinct names — are mentioned in the annals. Many of 
the old roads are still traceable : and some are in use at 
the present day, but so improved to meet modern require- 
ments as to efface all marks of antiquity. 

The ancient Irish classified their roads in regard to 
size and use into seven kinds, which are named and partly 
described in an interesting passage in Cormac's Glossary 
(p. 141). Cormac gives, as two general terms for a road 
of any kind, conair and cai, which are living words at the 
present day. The following are the names of the seven 
kinds of conairs or ways : they are given here, not in the 



Glossary order, but generally according to size : — Slige, 
Ramut, Bdthar, Rot, Set, Tuagrotd and Ldmrota. 

The slige [slee] was the largest of all : it was a main 
high-road. Cormac says " it was made for the passing of 
" chariots by each other, for the meeting of two chariots 
" [of the largest size], i.e. a king's chariot and a bishop's 
" chariot, so that each of them may go [freely] by the other." 

" Ramut (or ramat) " — says the Glossary — " is greater 
" [i.e. wider] than a rot : it is an open space or way which 
" is in front of [i.e. leads to] the forts of kings : and every 
" neighbour whose land comes up to it is bound to cleanse 
" [his own part of] it." A ramat is mentioned in the 
Senchus M6r, as subject to certain laws ; and the gloss 
gives the following description of it : — " A ramat, i.e. a 
" great road to which there is no fence [meaning that it is 
" open on both sides] and to which run all small by-roads : 
" and the fine for not cleansing the roads has a stay of three 
" days."* There is here no mention of a king's fort — as 
there is in Cormac's Glossary — from which we may infer 
that the ramuts were not used exclusively in connexion 
with the residences of kings. 

"A Bdthar " [boher] — says the passage in the Glossary 
— " two cows fit upon it, one lengthwise, the other athwart : 
" their calves or their yearlings fit on it along with them 
" [i.e. each calf walking beside its mother] : for if the calves 
" were behind them, the cow that followed would gore " 
[the calf in front of her]. Bdthar is still the common word 
for a road, and the diminutive bohereen or boreen (Irish 
bdithrin) is a familiar Anglo-Irish word for a little road 
or country lane. 

Rot (pron. rote : sometimes written rat), according to 
the Glossary, is compounded of ro, great, and set, a way : 
i.e. ro-shet [ro-hait], a great set or way — i.e. a road which 
is greater than a sit. " A rdt was made for the horses of a 
mansion, and there is room on it for a one-horse chariot." 

* Br. Laws, I. 233. 


The gloss on the Senchus Mor* defines rot : — " a small (i.e. 
narrow road), to which there is a fence " (clad), namely, a 
raised bank or " ditch " on each side. Rot, written in 
modern Irish rod [road], is still in use, and is evidently 
cognate with the English word road : for Cormac's deri- 
vation, above, is fanciful. 

Set [shate], the Glossary says, is less (i.e. narrower) than 
a rot, and is " a path of one animal," i.e. wide enough for 
a single cow or horse. 

A tuagrota is a small road, a farmer's road, such as he 
makes when he is permitted or purchases a right of way 
from his farm to an adjacent main road, or to a mountain 
for the convenience of sending cattle to graze on it, or of 
bringing home turf. 

A lamrota (i.e. a. hand-road : 1dm, a hand) is a small 
by-road, made for convenience of communication to con- 
nect two sliges or main roads. " Ldmhrdd, as much as to 
say, rdd Idimh le rdd eile, ' a road beside another road ' 
(O'Cl. Gloss.). 

The five main roads leading from Tara are mentioned 
in our oldest authorities, as, for instance, in the Story of 
Bruden Da Derga in the Book of the Dun Cow. They 
were all called slige. I. Slige Asail [slee-assil] ran from 
Tara due west towards Lough Owel in Westmeath, and 
thence probably in a north-west direction : it divided the 
ancient kingdom of Meath into two equal parts, North 
and South, f 2. Slige Midluachra extended northwards 
towards Slane, through the Moyry Pass north of Dundalk,| 
round the base of Slieve Fuaid near Newtown Hamilton, to 
Emain, and on to Dunseverick on the north coast of Antrim 
(Faraday's Tain, p. 59), portions of the present northern 
highway run along its site. 3. Slige Cualann ran south- 
east through Dublin, across the Liffey by the hurdle-bridge 

* Br. Laws, I. 233. 

f Book of Rights, Introd., lviii : Three Fragm., 77, 8> 9 . 

% Through the Moyry Pass : see Miss Stokes's Inscr., 11. 28 bot. 


that gave the city the ancient name of Baile-atha-cliath 
(the town of the hurdle-ford : now pron. Blaa-clee) : crossed 
the Dodder near Donnybrook : then south, still through 
the old district of Cualann, which it first entered a little 
north of Dublin, and from which it took its name (the slige 
or road of Cualann), and on by Bray, keeping near the 
coast. Fifty years ago a part of this road was traceable 
between Dublin and Bray. 4. Slige Data, the south- 
western road, running from Tara towards, and through, 
Ossory in the present Co. Kilkenny. This old name is 
still applied to the road from Kells to Carrick-on-Suir by 
Windgap. 5. Slige Mor (" great highway ") led south- 
west from Tara till it joined the Esker-Riada* near Clonard, 
along which it mostly continued till it reached Galway. 
Portions of this road along the old Esker are still in use, 
being traversed by the present highway. 

Besides these five great highways, which are constantly 
referred to, the Annals and other old documents notice 
numerous individual roads. In the Four Masters we find 
thirty-seven ancient roads mentioned with the general 
name bealach [ballagh], nearly all with descriptive epithets, 
such as Bealach Mughna, Mughain's or Mooan's Pass, now 
Ballaghmoon, near Carlow. Many of these are still com- 
memorated in the names of townlands. This word bealach, 
which is not included in Cormac's List of road-names, 
though in existence long before his time, is still in use. It 
means a pass with a road or path constructed through it. 
Another generic word for a road or way is raen or raon. 

* Esker-Riada, a long, natural, wavy ridge formed of gravel, running 
almost across the whole country from Dublin to Galway. It was much 
celebrated in olden times, and divided Ireland into two equal parts, Leth- 
Conn (' Conn's half ') on the north, and Leth-Mow (' Mow's half ') on the 
south. It may be seen marked on the map in the first volume of this 
book (squares 33, 34, 35, 36). The Irish eiscir means a sand-hill, and Had, 
travelling by chariot, horse, or boat : Eiscir-Riada, the ' sand-hill of 
chariot-driving." For the origin of the names Leth-Conn and Leth-Mow, 
see Joyce, Short History of Ireland, p. 131. 


In old times the roads seem to have been very well 
looked after : and the regulations for making them and 
keeping them in repair are set forth with much detail in 
the Brehon Laws/ The Book of Aicill lays down that that 
part of a main road {prim-rot) passing through a tuath or 
territory belongs more to the king of the tuath than to an 
inferior chief of the same tuath whose land adjoins the 
road : but that a by-road (for-rot), if bounding or passing 
through the minor chief's land, belongs more to him than 
to the king of the tuath. If any person injured a road, he 
had to pay compensation to the king or chief, or both, of 
the territory or district : and for the reason stated, if it was 
a main road, the king got a larger part of the fine than the 
chief ; but if a by-road, the chief got more than the king.* 
Care was taken that the roads were kept clean. According 
to Cormac's Glossary (p. 142), a road of whatever class had 
to be cleaned on at least three occasions : — the time of 
horse-racing, time of winter, and time of war ; which 
included clearing of brushwood, of water, and of weeds.: 
a statement also found in the gloss of the Senchus Mor.f 
The Glossary goes on to say that the road was thus cleaned 
in order that neither chariots going on a journey nor 
horses going to a fair should be soiled : and it was kept 
clear of brambles and weeds lest any one going [on horse- 
back] to battle or elsewhere might be upset. 

As illustrating the liberal and kindly spirit of the 
Brehon Laws, it is worthy of mention : — if a man's farm 
was so situated as that there was no way out of it except 
through his neighbour's land, he was entitled to purchase 
from him a tuagrotae or small roadway ; or if he did not 
do so, he could claim a passage in cases of necessity, but 
under certain restrictions. If he had occasion to drive his 
cattle out through his neighbour's farm, six persons were 
to be sent with them to prevent them spreading over the 

* Br. Laws, in. 305, 307, 309. 

f Ibid., 1. 129, n. See also vol. iv. 145, ^ 


land, three from the owner of the cattle and three from the 
owner of the land. This no doubt was mainly intended to 
prevent the acquisition of an unlimited right of passage by 
long usage.* 

We find similar equitable ideas running through the 
rules laid down for the public right of way. Under 
ordinary circumstances there was a fine for breaking 
through fences. But it was justifiable to make gaps in 
a man's hedges or fences where it was necessary for the 
passage of an army on the march, or of persons bringing a 
corpse to be buried, or for the passage of carts bringing 
provisions to an army, or for bringing building materials 
for a mill, a church, or the fort of a king. No compensation 
was allowed for these, provided no other convenient way 
could be found : but in all cases the breaches should be 
closed up after the passage, so as to leave the fence as 
perfect as at first, otherwise it was a case for damages. 
This rule is laid down also, which is very characteristic of 
the Irish Celt, that in every case permission was to be 
asked, which of course was always granted : — " Leave is 
" asked about them all, for it is an old maxim with the 
" Feini ' every supplication is pleasant.' " If the gap was 
not closed, the damages therefor were increased if leave 
had not been asked, f 

It seems that certain persons were bound to make a 
high-road (slige) through a wood if required, in time of 
war ; and to make and keep clean a rot at certain other 
times : but the statement on this point made by the 
glossator gives no information as to ways and means, or 
compensation. As an illustration of the favour with which 
road-work was regarded by all sections of the community, 
it may be mentioned that according to a statement in the 
ancient Irish " Life of St. Fechin," when a general three 
days' fast was enjoined with a certain object, a fast of two 

* Br. Laws, iv. 157, and Introd., cxxxii. 
t Ibid., 155, 157. 


days and one night was accepted from anyone who was 
engaged in making or repairing a bridge, or a causeway 
through a marsh.* 

From the evidence adduced in this short section, we 
may conclude that Giraldus Cambrensis was mistaken 
when he asserted — as he did in his " Topography " (I., iv.) 
— that Ireland " is truly a desert land, without roads, but 
well watered." It may be supposed that he referred to 
the uninhabited districts, covered with forest, bog, or 
marsh, which were extensive in his time and for long 
after : in which case he made no mistake. 

Bridges. — There is no evidence to show that the Irish 
built stone bridges before the Anglo-Norman invasion. 
The Senchus M6r lays down a law for the erection of a 
bridge, the gloss on which notices a distinction between 
stone and wooden bridges, f But the gloss is of much later 
date than the original text ; and no conclusion can be 
drawn from this passage as to the erection of stone bridges 
before the twelfth century. The Annals relate how Aed 
Allen, king of Ireland (a.d. .734-743), on one occasion 
stated that he had an intention to build a bridge at 
Clonard [across the Boyne], and to build it " marvellously " 
{i.e. on an unusually grand scale), " so that my name might 
live on it for ever " : % but even this grand bridge was 
no doubt intended to be a timber one. Lynch, in his 
Cambrensis Eversus (11. 193, top), says : — " I have not 
" treated of [ancient Irish] bridges because I have not 
" been able to ascertain whether they were of stone or 
" of planks." 

Droichet, the Irish term for a bridge, is a native word. 
Cormac's Glossary (p. 54) gives three alternative derivations 
of it, one of which is doroichet, ' he passes,' " for everyone 
" passes over it from one side to the other of the water or 
" of the trench." The place chosen for the erection of a 

* Rev. Celt., xn. 431. f Br. Laws, 1. 125, 9j 135. 

% Three Fragm., 15. 


bridge was very usually where the river had already been 
crossed by a ford ; for besides the convenience of retaining 
the previously existing roads, the point most easily fordable 
was in general most suitable for a bridge. Bridges were 
very often built of planks laid across the stream from bank 
to bank, if it was narrow enough, or supported on rests of 
natural rock or on artificial piers, if the river was wide : a 
kind of bridge occasionally used at the present day. 
There was a plank-bridge across the Shannon in the time 
of Brian Boru, near his palace of Kincora, that is, either at 
the very place where the bridge of Killaloe now stands, or 
near it. For we read in the " War of the Irish with the 
Danes " (p. 145), that, shortly before the Battle of Clontarf, 
when Mailmora, king of Leinster, retired in anger from 
Kincora, a messenger from Brian followed him, and " over- 
" took him at the end of the plank-bridge of Killaloe on 
" the east side." Sometimes bridges were constructed of 
strong hurdles supported on piles. A bridge of this kind 
across the Liffey gave Dublin its old name, Baile-atha- 
cliath (see p. 396, supra). These timber bridges of the 
several kinds were extremely common, and they are 
frequently mentioned in old authorities. The fourteenth 
abbot of Iona, from a.d. 726 to 752, was Cilline, who was 
surnamed Droichtech, i.e. the bridge-maker, from the 
number of bridges he got built ; and Fiachna, the son of 
Aed Roin, king of Ulidia in the eighth century, was called 
Fiachna Dubh Droichtech, Black Fiachna of the bridges, 
because " it was he that made Droichet-na-Feirsi and 
Droichet-Mona-daimh and others." These must have been 

Causeways. — In early ages, before the extension of 
cultivation and drainage, the roads through the country 
were often interrupted by bogs and morasses, which were 
made passable by causeways. They were variously con- 
structed ; but the materials were generally branches of 
trees, bushes, earth, and stones, placed in layers, and 


trampled down until they were sufficiently firm ; and they 
were called by the Irish name of tochar, now usually 
anglicised togher. These toghers were very common all 
over the country ; our annals record the construction of 
many in early ages, and some of these are still traceable. 
Sometimes a togher was covered over with planks laid 
across, forming what they call in America a corduroy 

2. Chariots and Cars. 

Our literature affords unquestionable evidence that 
chariots were used in Ireland from the most remote 
ages, both in private life and in war. They are men- 
tioned constantly, as quite common and familiar, in the 
ancient records, both legendary and historical, as well as 
in the Brehon Laws, where many rules are set forth 
regarding them. In the ancient historical tales in the 
Book of the Dun Cow and the Book of Leinster, the 
great chiefs, such as Cuculainn, Conall Cernach, Laegaire 
Buadach, &c, are constantly described as going to battle 
in chariots, each driven by a charioteer. At the Battle 
of Crinna, near Slane in Meath, about a.d. 254 (Four 
Masters, 226), Teige, the leader of the Munster forces, 
used a chariot, and was borne away in it from the 
field by his charioteer when severely wounded.* When 
St. Patrick was on his journey to Tara in 433, King 
Laegaire, as we are told in the Tripartite Life, went 
from Tara to Slane with nine chariots to arrest him for 
lighting the forbidden fire. We know from the best 
authority, such as the Book of Armagh and Adamnan's 
" Life of St. Columba," that SS. Patrick, Brigit, Colum- 
kille, Declan, &c, journeyed in chariots in their missionary 
progress through the country. And as Cuculainn's chario- 
teer, Loeg, is celebrated in the ancient tales, so St. Patrick 

* Lynch, Cambr Ev., II. 177 : Keat., 326. 



had a charioteer, Odran, who is equally well known in 
ecclesiastical history. 

The use of chariots continued without interruption 
both in military and civil life to a comparatively late 
period : and they always formed a prominent feature 
of fairs and other public meetings. In the time of 
St. Sechnall, fifth century, there was an aenach or fair at 
Dunshaughlin, which in spite of the saint's expostulations 
was held on the maigen or sacred precinct of the church 
(p. 44, supra) ; but — as the legend tells us — the earth 
swallowed up thirteen of the chariots with their horses 
and drivers, and the rest fled off the field glad to escape 
with their lives.* In the next chapter (p. 447) we shall see 
the regulations about chariots at the great fairs. Dermot, 
king of Ireland, when preparing for the Battle of Culdremne, 
a.d. 561, " gathered an immense army of horse, foot, and 
chariots."f Adamnan, in his " Life of St. Columba " 
(P- 33) > notices the Battle of Ondemone or M6in-M6r, 
fought a.d. 563, where the northern Hy Neill defeated 
the Dalaradians — a battle also recorded in the Annals : 
and the Dalaradian king, Eochaid Laib, " escaped," writes 
Adamnan, "sitting in his chariot" (currui insidens). 
Chariots played a prominent part in the great Battle of 
Moyrath, a.d. 637 : — " The snorting and neighing of their 
" steeds bounding under chariots, supporting and com- 
" manding the battle around them in every direction " 
(Moyrath, 193). Chariots are depicted on several of the 
high crosses (dating from the tenth to the thirteenth 
century) ; two of which are represented at p. 408, below. 
As is usual in case of important articles of every-day 
life in constant use, the bardic annalists have assigned an 
origin for the chariot : for we find it stated in the Book of 
Leinster that the first who invented chariots in Erin was 
Righairled, a prince of Munster, fourteenth in descent 

* Book of Hymns, 29, 30. 

f Cambr. Evers, 11. 177. See also FM., vol. 1., 193, note. 


from Eber Finn, the son of Milesius — long before the 
Christian era.* 

The usual Irish word for a chariot is carpat (now 
carbad), obviously cognate with the Latin carpentum, 
which is itself a Gaulish word. Adamnan always uses 
the Latin equivalent currus : but classical writers call the 
Gaulish and British chariots essedum, which however is 
another Gaulish word. For all three branches of the 
Celtic people, the Gauls, the Britons, and the Irish, used 
chariots. Carpat, as we see, is a native word, coming 
directly from Gaulish : but there are at least three other 
native terms, for a chariot, all given in Cormac's Glossary 
(p. n) : — A, ' a wain, or a car, or a chariot ' : corb, whence 
the personal name " Cormac," properly C orb-mac, ' chariot- 
son,' i.e. born in a chariot (Corm., 29) : and lastly cul, 
whence comes culgaire [3-syll.], i.e. the gdire, ' voice,' or 
creaking of a chariot ; and whence also a chariot-maker 
was called culmaire, which in the Glossary is denned saor 
denma carpait, ' an artificer who makes a chariot ' (Corm., 
pp. 39, 46). 

In the old romances there are several descriptions of 
Cuculainn's chariot, as well as of those belonging to other 
chiefs ; and in these and many other authorities details are 
given, from all which we can obtain a good general idea 
of the construction of the vehicle. They show, moreover, 
what might be expected, that there were varieties in shape, 
make, and materials, f The body ^Irish cret) was made of 
wickerwork, supported by an outer frame of strong wooden 
bars ; and it was frequently ornamented with tin, a practice 
which also prevailed among the Gauls. The ordinary one- 
or two-horse chariot had two shafts : fertas, a shaft, plural, 
fertse, feirse, or feirtse. The shafts were made of hard 
wood, often of holly. The charioteer of Orlam, son of 

* For more about the use of chariots, see vol. i., pr>. 87, 89, 90, supra. 
f See Crowe's Essay on the Irish Chariot, Kilk. Arch. Journ., 1870-71, 
P- 4!3- 


Ailill and Maive, on one occasion was employed cutting 
chariot-poles from a holly-tree in a wood* ; and in the 
Brehon Law we are told that the holly was counted among 
the noble trees because the feirse of chariots were made 
from it.f O'Donovan translates feirse in this Law passage 
by " axle-trees ' : and as holly is extremely hard and tough 
it would naturally be used for axle-trees as well as shafts. 
But that fertas was the usual word for the chariot-shaft 
there is no doubt. St. Brendan says to Iarlaithe (or 
Jarlath) of Tuam : " Let a new chariot be built by thee ; 
"... and wheresoever the two shafts (dd fheriais) of the 
" chariot shall break, there thy resurrection shall be."J 
Again in the Fled Bricrenn, Findabair, looking out from 
her high-up greenan, sees a hero coming in a two-wheeled 
chariot, which she describes to her mother — and among 
other things its feirtse or fertas-es — as hard and straight like 
a sword.§ Here the feirtsi (pi.) were obviously the two 
front shafts : for there was only one axle, and even that out 
of sight. Many other passages might be cited in which 
the two shafts are described in similar terms. But this 
word feirtsi was also applied to the two hind-shafts, on 
which the chariot rested when it was put by and thrown 
back, as we see in our present carts. On one occasion 
Cuculainn, driving his chariot, had a wild ox tied between 
the two feirtse behind. || 

In a two-horse chariot there was a pole (sithbe : pron. 
sheeva) between the two horses. King Laegaire describes 
Cuculainn's chariot as having hard, sword-straight fertas-es, 
and a sithbe [with ornaments] of white silver, ^j This pole 
is what Caesar calls temo when he describes the Britons as 
running along on it standing on the yoke (jugum) while 
fighting, and running back again to their seat in the 

* Miss Hull, Cuch. Saga, p. 155 : $ Stokes, Lives of SS., line 3495 
Lady Gregory's Cuchulain, p. 196 : § Kilk. Archa?ol. Journ., 1870-1, 

LL, 68, a, 29. p. 376, l6 . 

f Br. Laws, iv 151, 3 . see also || Ibid., p. 420, M . 

p. 287, supra. Vlbid., 376, l6> I7 ; 4*4, «. 


chariot.* A one-horse chariot had two shafts but no pole. 
A two-wheeled chariot, whether with one or two horses, 
was in very general use. There was a cuing or yoke 
between the two horses — also called mam [maum] — on 
which the shafts depended : Cuculainn once broke the 
cuing of Conall Cernach's chariot by striking it with a 
stone, on which the chariot fell down and Conall tumbled 
out so as to dislocate his shoulder. 

There were two words for a wheel, roth and droch, both 
of which are given in Cormac's Glossary (p. 61), and the 
latter corresponding with Gr. trochos. The wheels were 
spoked — sometimes six spokes, sometimes eight — and were 
from three to four and a half feet high, as we see by several 
delineations of chariots on the high crosses (p. 408, below). 
They were shod all round, generally with iron : Cuculainn's 
chariot, as we are told in the Tain, had rotha iarnaidi, 
" iron wheels," i.e. was iron-shod : and many other such 
references might be cited, f This corresponds with what 
we know of the ancient British chariots, of which some 
specimens have lately been found in burial-mounds, with 
iron rims on the wheels. Sometimes the Irish chariot is 
described as bronze-wheeled : by which is meant that the 
wheels were tired with bronze. Some chariots had four 
wheels : in the Battle of Rossnaree (p. 19), King Concobar 
tells Cuculainn to bring horses and to yoke to them " four- 
wheeled chariots " (carfiait chethir Had) : and we know that 
four-wheeled chariots were also in use among the Gauls. 
The axle pin was fixed immovable in the vehicle, and the 
wheels revolved on it, and were kept in their place by 
linch-pins. In some of the Latin Lives of the Saints the 
linch-pin is called either obex or roseta : now called in Irish 
dealg-roithledin [dallag-rolaun], " the pin of the wheel ' : 
roithledn, modern Irish for a wheel. J 

* Gallic War, iv. xxxiii. xiv. 417 ; and Kilk. Arch. Journ., 

f Miss Hull, Cuch. Saga, p. 176 : 1870-1, 414, I4 . 
LL, 78, a, 32. See also Rev. Celt., J Adamn., 172, d; 174, g. 


There was commonly an awning or hood overhead, 
often called pupall, which is one of the words for a tent, 
and is borrowed from the Latin papilio. Cuculainn's 
chariot is described as having a pupall cor cor da, a " purple 
hood," from which we may infer that it was of some kind 
of cloth.* But there was a native word for the hood — 
anbluth — to which more than once is appended an epithet 
that points to another material as in use. Findabair, 
describing the chariots of Laegaire the Victorious and of 
Conall Cernach, says in each case that there was anbluth 
n-in n-etegndith udsa creit charpait,\ " an awning of the 
wings of birds over the body of the chariot," showing that 
these great warriors used birds' plumage to roof their 
chariots, as ladies did for the roofs of their greenans 
(p. 30, supra). 

Kings, queens, and chieftains of high rank rode in 
chariots, luxuriously fitted up and ornamented with gold, 
silver, and feathers. Cuculainn, travelling in his chariot, 
orders his charioteer, on the approach of night, to spread 
for him the cushions and skins (joirtce ocus forgaimin) 
of the. chariot, preparing to sleep on them. J But with 
all this, the Irish chariot, like those of the Romans and 
other nations, was a rough springless machine, and made 
a great deal of noise. They evidently took pride in the 
noise : and the more distinguished the person riding in a 
chariot, the greater was supposed to be the creaking and 
rattle, as is often boastfully remarked by the old Irish 
writers, " a chariot under a king " being the noisiest of 

We occasionally come across expressions that enable 
us to arrive at some distant estimate of the value of a 
chariot. One of the annalists|| mentions a chariot — 

* Kilk. Arch. Journ., 1870-1, 377, „. 

t Windisch, Ir. Texte, 1. 365, " Anbluth " : LU, 106, a, 3 and 38. 
X Sick Bed : Atlantis, 11. 374, bot. of note. 

§ Kilk. Arch. Journ., 1870-1, p. 414 (noise twice) : Sick Bed, Atlantis, 
I. in, 2nd v. || Three Fragm., 45 : and Rev. Celt., xxiv., 59. 


evidently rather a good one — two-wheeled, no doubt, 
though this is not specified — as worth four cumals or 
twelve cows, representing £150 or £160 of our money. 
And one portion of the reward promised by Queen Maive 
to Ferdiad to fight Cuculainn was a [royal] chariot worth 
eighty-four cows — something like a thousand pounds.* 

The principal person in the chariot, the warrior or 
master, or chariot-chief, was commonly called err or 
eirr,\ and sometimes cairpthech (' chariot- warrior ') . The 
charioteer or driver was called ara. Loeg, Cuculainn's 
charioteer, is described in the Book of the Dun Cow as 
wearing a frock (inar) of deerskin, close fitting, so as not 
to impede the free action of his arms : over this was thrown 
a loose mantle — for-brat, i.e. an ' over-mantle ' (p. 200, 
supra). He wore a many-coloured helmet, from which 
fell a curtain behind down over his shoulders. Across 
his forehead was a gold-coloured band or fillet (gipni), 
which was worn as a special mark of a charioteer, or, as 
the old account says, "asa token of his charioteership -to 
distinguish him from his lord.' 'J Whether the chariot 
belonged to an ecclesiastic or a warrior, there was a special 
seat for the master, and an inferior one for the charioteer, 
as in case of St. Patrick and his charioteer Odran.§ In 
the " Phantom chariot of Cuculainn," the charioteer sits in 
front of the hero (ar a belaib) . But generally the charioteer 
sat on the right of the champion, as we know from Cormac's 
Glossary (p. 80), where we are told that fochla is the name 
for the champion's seat [in a chariot], which word also 
signifies both the north and the left-hand side. (See deas 
and tuaith, page 521, infra.) From this term, fochla 
fennida (' seat of the champion ') came to signify, in an 
extended sense, any distinguished seat at a banquet or 

* Sick Bed, Atlantis, II. 375, last par. of note, 
f Eirr glosses " curruum princeps " in Z., 255, „, 
X For all this, see O'Curry, Man. & Cust., 1. 290 : and Kilk. Arch. 
Journ., 1870-1, p. 423. 

§ Trip. Life, 219 : Three Ir. Gloss, Pref., xxxix. 




elsewhere. For a like reason the word faitsi, which 
signifies south, or right hand, means also the charioteer's 
seat (Corm. 80). 

On several of the high crosses chariots are carved, 
as, for instance, on those of Clonmacnoise, Tuam, and 
Monasterboice. The chariots represented here, from one 
of the Clonmacnoise crosses, have each only one horse and 
one pair of wheels : but two-horse chariots were more 
usual, and seem to have been a common vehicle for 
travelling.* The chariot ordinarily used in battle had 
two wheels and two horses. That four horses were some- 
times used is plain from the record in the Annals, of the 

Fig. 333. 

Ancient Irish chariots on base of cross at Clonmacnoise : ninth century. (From Wood-Martin's Pagan 

Ireland, p. 247.) They are also figured in Miss Stokes's Christian Inscriptions, I., PI. xxxiii. 

death of a prehistoric King Roitheachtaigh [Rohaghty : 
whose name signifies " possessor of wheels "], FM, a.m. 
4176, with the remark, " he was the first that drove a 
chariot with four horses in Erin." 

With rare exceptions, only two persons rode in a 
chariot, whether in battle or in everyday life : viz. the 
master (or mistress) and the driver : a custom which pre- 
vailed also among the Gauls, f Chariots were generally 
drawn by horses, especially those of chiefs and military 
men. But ordinary persons, and non-military people in 

* Two horses : see Fled Brier., 55, 61, 63 : and Bee Fola, 175. 
also De Jubainville, La Civil, des Celtes, p. 338. 
t De Jubainville, La Civil, des Celtes p. 331. 



general, often employed oxen : St. Patrick's chariot was 
drawn by two oxen.* 

Besides the chariots hitherto mentioned, both for travel- 
ling and for fighting, there was a special war-chariot 
furnished with scythes and spikes, like those of the Gauls 
and ancient Britons, which is repeatedly mentioned in the 
Tales of the Tain : most often in connexion with Cuculainn. 
It was called carpat serda, i.e. " scythed chariot," from 
sen, a saw, scythe, or sickle. It is thus described in the 
Book of Leinster : — " The hero of valour [Cuculainn] 
" leaped into the scythed battle-chariot, with its iron 
" points, with its sharp edges and hooks, with its hard 
" spikes, with its sharp nails, projected from its shafts and 
" straps and tackle. "f There is a similar description in the 
Book of the Dun Cow.J In O'Clery's Glossary is this 
entry : — " Searrdha, an edge : carbad searrdha, a chariot 
in which were sharp edges or sickles " ; and in another 
more ancient authority we read : — " It is why it was 
" designated sen da from the iron saws or scythes which 
" would be in array out of it."§ These accounts in the lay 
literature of the use of the scythed chariot are curiously 
corroborated in one of the ecclesiastical pieces, the con- 
mentary in the Amra of St. Columkille. In the Amra 
itself Columba is designated " chariot through battle " : on 
which the ancient commentator utters this remark, or 
rather prayer : — " That is, as a scythed chariot {car-pat 
" serda), namely, a chariot armed with swords, goes 
" through a battle (or a battalion), so may my soul go to 
" heaven through the battle (or battalion) of demons. "|| 

* Trip. Life, 253, 292. 

f LL, 78, a, 19 : Miss Hull, Cuch. Saga, 176 : Kilk. Arch. Journ. 
1870-1, p. 415. 

% LU, 80, a, 22 : Man. & Cust., 1. 300. See also LL, 76, b, A6 . and 79, 
b, 3 . § Kilk. Arch. Journ., 1870-71, p. 416. 

|| Stokes, Amra, Rev. Celt., xx. 149. See the original in LU, 6, b, 
30. For scythed- and war-chariots, see De Jubainville, La Civil, des 
Celtes, 339-341 : also 382 et seq. 


Farmers and people in general used rough carts, com- 
monly called can, for work of various kinds, but they are 
hardly noticed in the ancient literature. The name can, 
which is used in Cormac's Glossary (p. 44), is cognate with 
Latin carrus and English car. In the Senchus M6r, a 
chariot, as denoted by carpat, is distinguished from a can, 
which is explained in the gloss as a cart for corn or dung :* 
Another old name for a common cart or wagon or wain 
was fin [fain], which glosses plaustrum in one of the eighth- 
century mss. of Zeuss (Z., 19, 1 : 776, 19). These carts, 
whether called fin or can, were drawn by oxen trained 
specially for the purpose, as we know by many references.! 
They had probably solid wheels — such as the people used 
in later times — spoked wheels being expensive. 

3. Horse-riding. 

Horses were put to the same uses as at present : — 
riding, drawing chariots, racing ; and more rarely plough- 
ing, drawing carts, and as pack-animals : all which uses 
are mentioned in our old literature. The horse is known 
by various names. Ech signifies any horse of a superior 
kind — a war-horse, a steed ridden by a chief, a chariot- 
horse, &c. : cognate with Latin equus, and Greek hippos. 
Marc, another word for horse, is explained in O'Davoren's 
Glossary, ech no lair, " a steed or a mare " : hence the 
common word marcach, a. horseman : marcach, equestor, 
Z. 60, 10. Capdll, meaning a horse of any kind — a term 
existing in varied forms in several European languages — is 
the word now in universal use by Irish-speakers. Gearrdn, 
a hack-horse, means equus castratus, a gelding, from gearr, 
to cut. In ancient Irish documents this word denoted the 
common beast of burden : in the Anglicised form garron, 
it was constantly used by the Anglo-Irish writers of the 
time of Elizabeth : and garron or garraun is in general use 

* Br. Laws, i. 166, 33; 171, t . f For instance, Todd, St. Patk., 167. 


at the present day in Ireland among speakers of English 
to denote a heavily-worked, half-broken-down old horse. 
Other terms for a horse, such as fell and gobur, both of 
which are given in Cormac's Glossary (80, 83), need not be 
dwelt upon. The common word for a foal was, and is, 
serrach or searrach [sharragh]. 

The Senchus Mor provides against " over- tying " or 
" over-fettering " of horses when taken in distress ; i.e. it 
was forbidden to fetter so tightly as to cause suffering or 
injury : for example, the head was not to be held down by 
a rope tied to the leg and carried tightly round the neck.* 
In every chief's house there was an echaire [eh'ara] or 
horse-boy or stable-boy — sometimes called gilla-scuir, 
* horse-stud boy ' — who stabled the horses at night, and 
let them out in the morning, f 

Horses were often let run half wild in droves about the 
mountains and plains ; and whenever any were wanted, a 
sufficient number were driven home and trained in. A 
part of the stipend the king of Hy Blathmaic (the district 
round Newtownards in Down) received from the king of 
Ulidia consisted of " eight steeds not driven from the 
mountains J " : that is to say, not wild, but fully trained 
in. Before the Battle of Rossnaree (p. 19), when King 
Concobar wanted some horses, he says to Cuculainn : — 
" Well, O Cuculainn, let the horses {gaby a) of the plain 
" of Murthemni be caught by thee ; and let four-wheeled 
" chariots be harnessed to them," which was accordingly 
done : for Cuculainn, demigod as he was, subdued and 
tamed them on the spot. 

Besides grazing in the fields, horses were fed on corn 
of various kinds. When the Red Branch heroes in the 
palace of Ailill and Maive got their choice of food for 
their horses, Cuculainn chose barley grain for his, and 
Conall and Laegaire took airthend two years old for 

* Br. Laws, i. 169, s; 175, 8 . f Br. Laws, 111. 419, bottom. 

% Book of Rights, 163. 


theirs : where airthend probably means oats.* An ogaire, 
an inferior chief, was supposed to keep a horse which was 
used both for working and riding, f The higher classes 
of chiefs kept horses for riding exclusively, and others 
specially for work — that is, if they used horses for work 
at all. 

From many passages in the Brehon Laws and other 
old writings it appears that horses were often imported, 
and that those from Wales were specially prized. Pichan, 
the Munster chief, promises Mac Conglinne (p. 44) a gold 
ring and a Welsh steed (ech Bretnach) : and again (p. no) 
the same Mac Conglinne is promised a Welsh steed out 
of every [principal] house from Cam [Ui Neid\ to Cork. 
The glossator of the " Heptads " in the Brehon Laws 
(v. 221) includes British (i.e. Welsh) mares among the 
" foreign curiosities " mentioned in the text. According 
to the Book of Rights (p. 82) the king of Munster was 
bound to give the king of Ui Liathain, as stipend, among 
other things, " a steed and trappings brought from over 
sea." In like manner (p. 247) the king of Tara was to give 
to the king of Ui Briuin " a noble French steed." Even as 
late as the fourteenth century the " Tribes and Customs of 
Hy Many " records that a part of the tuarastal of the king 
of Connaught to his subject king O' Kelly of Hy Many 
was " ten steeds from across the boisterous brine. "J In 
the fourteenth, fifteenth, and sixteenth centuries, those 
Irish horses called hobbies were known all over Europe 
" and held in great esteem for their easy ample, . . . from 
" this kind of horse the Irish light-armed bodies of horse 
" were called hobellers."§ 

Giraldus Cambrensis|| tells us that in his time the Irish 
used no saddles in riding. Two hundred years later, 
Mac Murrogh Kavanagh had a splendid horse that cost 

* Fled Brier., 81. f Br. Laws, iv. 305, , 9 . 

X Hy Many, 93. See also Stokes, Lives of SS., line 3128, and p. 348. 
§ Sir James Ware, Antiqq., p. 166. || Top. Hib., in. x. 


him 400 cows, which he rode with wonderful swiftness 
without saddle down a hill to meet the earl of Gloucester* ; 
and the custom must have been very general at a still later 
time, for laws were made to compel the Irish and Anglo- 
Irish to ride like the English — with saddles. Yet this 
custom prevailed among the English themselves in early 
times, as well as among the ancient Britains, Gauls, and 
Romans, f 

But from the earliest times the higher classes of the 
Irish used a thick cloth called dillat, between them and 
the horse. Maildune, on one of the islands, saw a lady 
riding, having a " good 
adorned dillat under ^P^|* f^?\. 

her. "J In another part ■L^^^^Sw%^^^. 

of the same tale, a lady *q) W^Kp^| 
is seen riding on a horse ;, ^£^^^^^^^)^f^ J f 
" with an ech - dillat f /nJCa W vlM.^ 

[' horse- dillat '] under TTj | (CCjTjD IJi 1 
her." These two last *~*^wv_^ \"~* 

quotations prove too „ ^.333. - 

x * Grotesque representation of a horseman, given in the 

that ladieS DraCtised Book of Kells. Man's cap yellow ; cloak green, with 

* bright red and yellow border ; breeches green ; leg 

horSe-riding aS Well aS clothed; foot naked. Dillat yellow. (From Wilde's 

rr^-i i '11 r Catalogue.) 

men. The dillat often 

covered the whole animal, as is seen in the above illustra- 
tion taken from the Book of Kells (7th or 8th century) — 
and also in fig. 334, p. 417, infra, from the same old book. 
This word dillat originally meant an outer cloth, garment, 
or loose vest : in the Cymric Glosses of Zeuss it is given as 
the equivalent of vestimentum or vestis : and it is applied in 
this sense in the following Irish quotation from the Book of 
the Dun Cow : — Gabaid-seom dan a dillat n-oenig n-imbi in 
laa sin : " He puts his assembly-^7/atf or raiment about him 
that day."§ As dillat was originally applied to the cloth 

* Joyce, Short Hist, of Ireland, p. 330. f Ware, Antiqq., 159. 

% Rev. Celt., x. 63. 

§ LU, 81, a, 24 : see "Dillat" in Windisch's Worterb., Ir. Texte, 1. 481. 


thrown over the horse's back, the name was retained even 
when the cloth developed into a regular saddle : and dillat, 
or in its modern form diallaid, is now the common Irish 
name for saddle. Saddles of some kind were often used 
by the Irish before the time of Cambrensis, for they are 
frequently mentioned in the Book of Rights. The word 
used in this book is sadail, which however appears to be a 
loan-word from the Norse. Part of the stipend of the Hy 
Cennsealaigh from the king of Leinster was ten saddles 
(pi. saidle) : and the king of Ireland was bound to give the 
king of Luighne, among other things, twenty steeds with 
saddles.* The Senchus M6r glossf mentions whalebone 
as being sometimes used for making saddle-trees — cldr- 
sadall : but this is a late authority. 

When the Irish employed the horse as a beast of 
burden, to carry things on his back, they used a pack- 
saddle. This is mentioned in Cormac's Glossary (p. 153), 
under the name sratkar [srahar], which is still used for it, 
and which is there derived from sreth, a range, because, as 
Cormac says, " it is set on the range of the [horse's] ribs." 
The use of the pack-saddle is also mentioned in the " Story 
of the Eruption of Lough Neagh " in the Book of the Dun 
Cow, where certain persons pack a srathar which is on the 
back of a great horse with an immense load of household 
goods. J 

Two kinds of bridle having two different names were in 
use. The single-rein bridle, called srian (Lat. frenum), 
was used in horse-riding. This rein (Irish sreth, pron. srah] 
was attached to a nose-band, not at the side but at the top, 
and came to the hand of the rider over the animal's fore- 
head, passing right between the eyes and ears, and being 
held in its place by a loop or ring in the face-band (called 
drech-ongdas, from drech, face) which ran across the horse's 

* Book of Rights, 209, 267. f Br - Laws, 1. 135, last par. 

X Kilk. Arch. Journ., 1870-1, p. 97. 


forehead and formed part of the bridle-gear.* This single 
rein was used to restrain merely : it could not be used to 
guide, which, as we shall presently see, was done by a 
horse-rod. Accordingly a horse for riding is often called 
ech-sreine, a " smm-steed."f 

The two-rein bridle, called all or fall, was used with 
chariot-horses. The charioteer, who sat too far from the 
horse's head to use a horse-rod in guiding, had to use 
double reins, both to guide and to restrain, like those of 
the present day. The two reins were called sreiha (pi. of 
sreth), and also aradna (now earadhain or earadhain-srdine) . 
The distinction between those two kinds of bridle and their 
uses is clearly set forth in a gloss on the Senchus MorJ : — 
" Srian, i.e. [having] one sreath or rein : all [so called from 
" oil or all, great] because it is greater than the srian on 
" account of having two reins on it : it is for the horses of 
" the chariot it is used." And this distinction is always 
observed in the Tales. Where horse-riders are mentioned 
they have srians or single-rein bridles : but in the descrip- 
tions of chariots the two horses have da n-all, " two alls,"' 
i.e. each steed has one all of two reins. Essi or esi seems to 
be another name for the all. Cuculainn says to Loeg : — 
Fosta latt essi fostada th' echraidi, which Stokes translates, 
" Fasten the securing reins of thy horses."§ Foill Mac 
Nechtain tells Cuculainn's charioteer not to unharness the 
horses from the chariot ; to which he answers that he is not 
going to do so, inasmuch as he still holds the esi and the 
aradna in his hands. || The bridle-bit was beilce or beilge 
[bailka, bailga], modern Irish beulmhach : from bel or beul, 
the mouth. 

* Kuno Meyer in Mac Congl., 89, 22 . and Glossary. 

f Br. Laws, iv. 326, „. 

X Br. Laws, 1. 139, bot. Other references to the all and srian, as dis- 
tinguished are, : — O'Curry, Man. & Cust., 11. 157, bot. and note, p. 158 : 
Kilk. Arch. Journ., 1870-1, pp. 376, I7; 413, 30 ,3 9 . 414, 7> l6 . Stokes, 
Lives of SS., line 317 : Courtship of Emer, 72, top line (LU, 122, b, 10) : 
Bee Fola, 175 : Fled Brier., 55 to 63 passim. 

§ Stokes in Rev. Celt., xiv. 419, top. || Miss Hull, 149 : LL, 66, b, 7. 


The bridle was often elaborately and expensively 
ornamented. Among the royal tributes of the Book of 
Rights (p. 57) we find " fifty steeds with costly bridles." 
In the Bruden Da Derga (p. 51) the king's retinue have 
" thrice fifty steeds with their thrice fifty bridles of red 
enamel (srian cruanmaith) on them." In the Crith Gabhlach 
the quality of the bridle is set down as indicating the rank 
of the man : — a certain class of chief (brugaid) is stated to 
have a riding-steed with a bridle of cruan or red enamel : 
another of higher rank (aire-desa), a steed with a bridle of 
silver ; and another still higher {aire-tuisi) with one of 
gold : meaning in all the cases that the bridles were 
adorned with the several materials.* Accordingly, special 
provisions were laid down in the Brehon Law (v. 415, 
417) for compensation to the owner of a bridle in case 
a borrower did not restore it ; from five or six cows up 
to eighteen or twenty, according to the rank of the several 
owners. In later times the Irish continued or became so 
extravagant in ornamenting their steeds that the Anglo- 
Irish parliament passed laws to restrain the use of over- 
expensive trappings, f 

In corroboration of all these accounts, portions of 
antique bridles and headstalls have been found from time 
to time, with enamelled ornamentation of beautiful work- 
manship, some of them now preserved in the National 
Museum. Petrie, in one of his letters, has an account of 
a headstall found near Boyle in Roscommon, all covered 
with a beautiful jet enamel J ; and Miss Stokes pictures 
and describes another, still more elaborate, of exquisite 
workmanship in coloured enamel.§ 

Giraldus Cambrensis says that the Irish in his time 
(n 85) used a sort of reins that served for both bridle and 

* Br. Laws, rv. 311, bot. ; 323, „. 327, l8 . f Ware, Antiqq., 160. 

\ Dr. William Stokes's Life of Petrie, p. 240, bot. 

§ Trans. Roy. Ir. Acad., xxx. 291, 293, bot., with plate. For another 
of a similar kind, pictured and described, see Kilk. Arch. Journ., 1856-7, 
p. 423. For more about enamelled bridles, see vol. 1., p. 559, supra. 


bit, and did not prevent the horse from grazing.* The 
common people generally used a halter, which was, and is 
still, called adastar or adhastar [sounded nearly as oyster]. 
Giraldus states that the Irish did not use spurs, but urged 
on and guided their horses with a rod having a hooked goad 
at the end. In this he is correct ; for we find frequent 
notices of the horse-rod — called echlasc and slatt — in Irish 
literature. In the Story of Aed Baclamh, we read that 
during a race at Tara, the riders, being dissatisfied with the 
pace, lashed their horses with their echlascs.] No mention 
is made of spurs, and none were used. This echlasc was 
commonly of yew or ash, and must 
have been — at least in some cases 
— pretty strong and heavy, not a 
mere switch : Mailmora, king of 
Leinster, when leaving Kincora in 
a rage, struck King Brian's mes- 
senger with the eachlasc of yew, 
" and broke all the bones of his 

Grotesque representation of horse- 

head."t The rider guided with the "«"• usi »e horse-rod, given in the 

Book of Kells (seventh century.) 

echlasc by touching one side or the (From wiidvs catalog, P . 3 oo.> 
other of the horse's head. In the 

Tain bo Fraich (p. 137) we read of certain noble riders 
having echlascs [with ornaments] of bronze and with 
hooked goads (boccan) of gold at the end. In later times 
the word echlasc came, by a natural extension of meaning, 
to be applied to a whip, or a strong rod with a lash or 
scourge at the end, which was used both for striking and 
guiding. Maildune, coming to a certain island, saw a horse- 
race a good way off, " and the strokes of their echlascs were 
heard by him "§ — notwithstanding the distance. Here the 
echlasc was no doubt the slatt with a scourge. 

In some of the above passages the word for goad is 
boccan, which is still in use to denote any hook or crook. 

* Top. Hib., in. x. X Wars of GG., 147, top line, 

t Silva Gad., 72 : Ir. Text, 68 ; „. § Rev. Celt., ix. 467. 



But the more usual word for a goad in the Tales is brot, 
and its diminutive bruitne [brit-ne] : " into the horses Loeg 
" drove the brot ; he plied the horse-switch (slat) towards 
road and wayfaring."* In the " Phantom Chariot of 
Cuculainn " we read : — [Cuculainn] " had a goadlet 
(bruitne) in his hand with which he urged on the 
horses." f 

Horse-riders also used a whip (sraigell or srogell : bor- 
rowed from Lat. flagellum). In the Bruden Da Derga, 
Conaire's three outriders (ritiri) had each a sraigell in hand J: 
and another rider is mentioned in the " Vision of Mac Con- 
glinne " who grasped a srogell. Srogell glosses flagellum and 
flagrum in Zeuss (80, 25 ; 769, i 4 ) ; and O'Clery explains 
the verb sroiglim by do sgiursudh, ' to scourge.' Whether 
the same rider carried both a rod and a whip is doubtful ; 
probably not. The custom of suspending little bells from 
the necks of horses is noticed in vol. 1., p. 376. 

It would appear that horses, as well as other domestic 
and pet animals, were sometimes wholly or partly dyed, 
for ornament. In the story of " The Courtship of Ferb," 
we are told that in the train of Prince Mani were fifty 
white horses with red ears, yoked in pairs to chariots, and 
having their long manes and tails dyed purple§ : and the 
circumstance is mentioned as if the practice was usual. 
From an expression in the story of young Ciaran (told at 
p. 360, supra), we may infer that other animals also were 
dyed. When the piece of cloth was taken out of the pot 
for the third time, not only was it dyed an intense blue, 
but what remained of the colouring liquor in the pot, after- 
wards " made blue all the dogs and the cats and the trees 
that it touched " (Stokes's Lives of SS., 267). The seven 
dogs that accompanied Prince Mani appear to have been 

* Death of Goll and Garb, Rev. Celt., xiv. 401. 

t Kilk. Arch. Journ., 1870-1, p. 378, top line. See also LU, 122, b, ^ 

X O'Curry, Man. & Cust., II. 146, note 212. 

§LL, 253, a, 11, 12 ; Ir. Texte, 462, 463 : Leahy, p. 5. 


dyed : for the narrative says that they were of every colour 
that could be imagined. See also Sullivan's observations 
in his Introd., p. 405, in which he states that dyeing pet 
animals is practised in India, where a blue dog is a 

Horsemen rode without stirrups : and every man was 
trained to spring from the ground by an ech-Uim or ' steed- 
leap ' on to the back of his horse.* The chief, Dicho, 
St. Patrick's first convert, was on one occasion a captive 
in Tara with some others. But they made their way out 
of the prison by the help of a cleric, and finding horses 
ready bridled on the lawn, " they leap on their horses " 
(leangait for a n-eochu) and make their escape. This ready 
method of mounting continued to the beginning of the 
seventeenth century in both Ireland and Scotland : — 

" No foot Fitz-James in stirrup staid, 
No grasp upon the saddle laid, 
But wreathed his left hand in the mane, 
And lightly bounded from the plain." 

— Lady of the Lake. 

It was considered necessary that every young man 
belonging to the upper classes should be taught horse- 
riding ; and so important was this that even the Brehon 
Law interfered, just as the law of our day requires children 
to learn reading. The Senchus Mor prescribes, among 
other accomplishments, horse-riding " for the sons of chiefs 
in fosterage. "f They began very young ; for the same 
authority (Br. Laws, 11. 159) tells us that up to the age of 
seven years the horse should be supplied by the father : 
after that by the foster-father. If the foster-father 
neglected this part of the child's instruction, he forfeited 
two-thirds of the fosterage fee. A horse was not given 
to the sons of the Feini or farmer grades while in fosterage, 

* See Silva Gad., 296, x8: and O Conor, Dissert. 70, note. 
f Br. Laws, 11. 157, bot. 


" for horsemanship is not taught to them " (Br. LL. n., 
p. 161). But the law took care that young farmers should 
be taught such things as were considered fit for them (see 
I. 441, supra). No doubt the horse supplied for learning 
was a commonplace animal : but the Law required that the 
son of a king should during fosterage be supplied with a 
high-class animal in time of races.* 

A simile found everywhere in the Irish and also in 
Welsh tales is : — " Like flocks of birds over the heroes' 
" heads were the sods thrown up by the hooves [or shoes] 
" of the steeds." Another not quite so common : — " Like 
" a flock of swans pouring over a vast plain was the foam 
" which the steeds flung from them over their bridles." 
" Like the smoke from a royal hostel was the dust and 
" the breath and the dense vapour, because of the vehe- 
" mence of the driving which Laeg gave Cuculainn's two 
" horses."f 

The period of the introduction of the practice of shoeing 
horses among the natives of Europe is not very clearly 
determined. The Romans had for this purpose a sort of 
sock faced with iron, which could be readily taken off and 
put on as occasion required, and which was used only 
in rough places. That the ancient Irish protected the 
horse's hoofs by a shoe of some kind is plainly shown 
by the records. This shoe is called cru in the oldest Irish 
documents : it is given with this meaning in all modern 
dictionaries, and cru is still the living word for a horseshoe, 
not only in Irish, but in Scotch Gaelic and Manx. But 
as cru was also used for ungula or hoof,| the inquiry needs 
to be conducted with some caution. That the word was 
intended to designate a horsehoe in at least some of the 
records (which serves our purpose as well as if it was so 

* Br. Laws, 11., Pref., xliv. 
t Rev. Celt., xiv. 417 : also LL, no, a, top. 

J Cru etch glosses ungulus ; Stokes, Ir. Glosses, 442 : crua glosses 
ungula in Sg. 46, b, 13 ; Stokes in Rev. Celt., xin., 469. 


used in all) is clear from several passages. In the " Demon 
Chariot " we read that showers of sods were thrown up 
from the " shoes " of Cuculainn's horses (a chruib nan ech). 
Here the word used is cru, which must mean ' shoe,' not 
hoof, for a few lines farther on, where the hoof is specially 
mentioned, it is called bos* : and accordingly Crowe here 
properly translates cru by ' shoe,' and bos by ' hoof.' An 
entry in the Four Masters, under a.d. 1384, goes to con- 
firm all this : — " Tomaltach Mac Dorcy, chief of Kinel- 
" Duachain, was killed by his own knife while he was 
" shoeing a horse " (ag cur cru : * putting on a horseshoe '). 
It seems plain from the preceding that the ancient Irish 
shod their horses : but nothing is known that would indi- 
cate the particular form of shoe used. Many specimens of 
iron horseshoes may be seen in the National Museum, 
generally lighter than the present shoes ; but there are 
no data on which to found an opinion as to their age. 

Giraldus Cambrensisf says that in his time the Irish 
women rode astride like the men, " their legs sticking out 
on both sides of the horse " : but I have not found any 
confirmation of this. Indeed the dress universally worn 
then by women would render it impossible for them to 
ride in this manner. It is likely that Giraldus may have 
witnessed an accidental and exceptional instance, as we 
may sometimes witness now. Four centuries after the 
period of his record, Spenser (View, 102) tells us that the 
Irish women rode, not astride, but on the wrong side of 
the horse, i.e. — as he says — " I meane with their faces 
towards the right side" [of the horse]. In this he is 
correct ; and the fashion moreover came down from old 
times ; for in a delineation of the flight to Egypt sculptured 
on the high cross of Moone Abbey, the Blessed Virgin sits 
on the ass with her face to the right — i.e. her left hand 

* Kilk. Arch. Journ., 1870-1, p. 376, 2> 9; 377, 2p I0 . For bos or has, 
' hoof,' see also Fled Bricrenn, p. 58, 8 . and " Bass " in Windisch's 
Worterbuch. f Top. Hib., in. xxvj. 


towards the ass's head, while St. Joseph leads the animal 
by a halter.* The concurrence of these two unquestion- 
able authorities, one before the time of Giraldus, the other 
after, renders it all the more likely that he is mistaken. 

In old times in Ireland, horse-riding as a mode of loco- 
motion in ordinary life was not very general. But nobles 
commonly rode and were very proud of their steeds and 
trappings. Horses were also kept and carefully trained 
for sporting purposes, chiefly racing, which, as we shall 
see in next chapter, was a favourite amusement. A poet, 
praising Sechnasach, king of Ireland (a.d. 664-671), says 
that his house was " full of bridles " and " full of horse- 
rods " (srianach and echloscach) .f 

The ass hardly figures at all in ancient Irish literature, 
so that it cannot have been much used. 

4. Communication by Water. 
The boats used by the ancient Irish way be roughly 
classified as of three kindsj : — canoes hollowed out from 
the trunks of trees ; curraghs, or wicker-boats ; and ordinary 
vessels — ships or boats — propelled by sails, or oars, or both 
combined, as occasion required. In the Brehon Law Tract 
called the " Small Primer " (v. 105), vessels are classified as 
(1) ler-longa, " sea-ships " (ler, the sea), viz. large vessels fit 
for long voyages ; (2) barca or barks, small coasting vessels 
not suitable for long voyages, which are also called serrcinn 
(sing, serrcenn, ' saw-head ') ; (3) curraghs. This classifi- 
cation — which is a good one — has reference, not so much 
to the vessels themselves as to their several builders, in 
order to set forth their privileges ; which explains why 
single-piece canoes are not included, inasmuch as they 
required small technical skill to make them. 

* O'Neill's Crosses, p. 7, and pi. xvii. f Rev. Celt., XIII. 97. 

% This general classification will be quite sufficient here : but that 
vessels were much varied in shape according to the purposes they were 
intended to serve is obvious from the fact that Adamnan mentions ten 
different kinds by their Latin names, which may be seen in Adamn., 
176, note b. 


The single-piece canoes were very common, especially 
in connexion with crannoges, where they were used to 
communicate with shore. Many of these have in late 
times been found in bogs at the bottom of dried-up lakes 
and near old crannoges, varying in length from 50 or 60 
feet down to six or eight : and numbers of them may 
been seen in the National Museum in Dublin. Adamnan 
(p. 176) mentions boats of pine and oak, which the monks 
of Iona dragged overland, and then used them for convey- 
ing across the water great timbers both for houses and for 
ships. These were single-piece boats, and must have been 
of considerable size. 

The curragh (Irish form curach, connected with Latin 
corium, a hide) was the best known of all the Irish boats. 

FIG. 335. 

Single-piece canoe : in the National Museum, Dublin : 23 feet long ; 2j£ feet broad. (From Wilde's 

Catalogue, p. 203.) 

It was made of a wicker-work frame, called in Irish cliab- 
curaich [cleev-curry], i.e. curach-ba.sket, covered with hides 
which were stitched together with thongs.* Some curraghs 
had a double hide-covering, some a triple. These boats 
are constantly mentioned in lay as well as in ecclesiastical 
literature, and also by Continental writers, the earliest of 
whom is Solinus in the third century. They are used still 
round the coasts, but tarred canvas is employed instead of 
skins. They were propelled by oars or sails according to 
circumstances. When there were two or more hide-cover- 
ings, they were probably placed in contact. A curragh of 
one hide was, of course, the least safe : and in the Life of 
St. Patrick we are told how an Ulsterman named Mac Cuill, 
a converted sinner, committed himself to the sea in a one- 
hide boat without oar or rudder, in accordance with a 

* See p. 427, line 13, infra ; and p. 370, supra. 


penance imposed on him by St. Patrick, as already 
mentioned (vol. L, p. 214). In Muirchu's Latin narrative 
of this incident, the boat is called navis unius pellis, while 
in the Irish Tripartite Life it is curach oen seiched, ' a 
curragh of one hide,' the exact equivalent of the Latin.* 
Maildune, intending to make a voyage in the Atlantic, had 
a boat of three hides constructed — n6i tre-chodlidi (codal, 
a hide) : here the boat is called ndi : but a little farther on 
it is called a curach.] Many curraghs were so small and 
light as to be easily carried on a man's back from creek 
to creek overland, as Giraldus says the Welsh were 
accustomed to carry their wicker boatsj : and as people 
sometimes do to this day in Ireland. 

The mode of constructing curraghs has been described 
by foreign as well as by Irish writers. Julius Caesar§ tells 
how he had some curraghs made for his use after the 
model of those used by the Britons (" ships of the kind 
that his knowledge of Britain had taught him ") ; and 
twelve centuries later Giraldus describes in similar terms 
the Irish curraghs as he saw them.|| But the most detailed 
and accurate account we have of the building of a curragh 
is in the Latin narrative of the Voyage of St. Brendan. 
The Saint and his companions " using iron tools [saws, 
" hammers, chisels, &c] prepared a very light vessel, with 
" wickerwork sides and ribs, after the manner of that 
" country, and covered it with cow-hide, tanned in oak- 
" bark (rubricatis in cortice roborina : ' reddened in oak- 
" bark,' p. 368, supra), tarring its joints : and they put on 
" board provisions for forty days, with butter enough to 
" dress hides for covering the boat [whenever the covering 
" needed repair], and all utensils necessary for the use of 
" the crew."T[ From all these accounts, which might be 

* Trip. Life, 222, „. 288, l8 . || Top. Hib., in. xxvii. 

t Rev. Celt., ix. 459, 460. ^f Brendan., Cardinal Moran, 90 ; 

% Descr. of Wales, 1. xvii. O'Donohue, 119. 

§ Bell. Civ., 1. liy 


corroborated by many others, we see that curraghs, when 
intended for long voyages, were made large and strong, 
furnished with masts and solid decks and seats, and having 
the hides tanned.* 

By far the greatest part of the water-communication 
round the coasts and across the narrow seas, as well as in 
the lakes and rivers, of Great Britain and Ireland, was 
carried on in those early days by curraghs, which indeed 
were used also in other parts of Europe. The Anglo- 
Saxon Annals and Florence of Worcester relate that 
three learned Irishmen, desirous of leading a religious 
life, went on board a boat which was made of two or 
three ox-hides, and with provisions for a week, and sailing 
wherever Providence led them, landed in Cornwall, whence 
they were brought to the great King Alfred, f We know 
that in the fourth, fifth, and sixth centuries the Irish sent 
numerous plundering expeditions to Britain, as mentioned 
in vol. 1., p. 73, et seq. These voyages they made in 
curraghs : and Gildas pictures hordes of them as landing 
from such vessels (de curicis).% 

The native records corroborate these, so far as the 
general use of the curragh is concerned. In Cormac's 
Glossary (p. 41) we are informed that Breccan, grandson 
of Niall of the Nine Hostages, had a fleet of fifty curraghs 
trading between Ireland and Scotland, till they were all 
swallowed up in the terrible whirlpool near Rathlin Island, 
which thenceforward took the name of Coire-Bhreccain 
[corrie-vreckan], Breccan's caldron or whirlpool^ When 
the Irish chief Mac Con gathered an army in Britain and 
Scotland for the invasion of Ireland, leading ultimately to 

* See also Silva Gad., 386. t Ogygia : in. xxiv. 

% Adamn., 169, note k. 

§ This whirlpool, which is still well known, but now called Slugnamara 
(' swallow of the sea '), lies between Rathlin and the coast of Antrim. It 
was the original Corrievreckan ; but its name was borrowed for the 
dangerous whirlpool between the islands of Scarba and Jura, in Scotland, 
mentioned in The Lord of the Isles, See Ir. Names of Places, 11. 433. 


the great battle of Mucruime, and the accession of Mac Con 
as king, he conveyed them in vessels of various sizes, so 
that between Scotland and Ireland " there was " — says the 
old account — " a continuous bridge of curraghs "* : and 
when — as we read in the story of " The Siege of Etar " — 
the Ulster Forces were besieged in Ben Edair (Howth) 
by Leinstermen, they sent north requesting their friends 
to come from Ulster either by land or "in curraghs " 
(i curchaib) to relieve them.f 

Many of the ordinary vessels used by the Irish in 
foreign commerce must have been large ; otherwise they 
could not have traded with Continental ports, as we know 
they did (p. 429, farther on). In the Book of Rights (p. 39), 
it is mentioned that part of the yearly tribute from the 
king of Cashel to the king of Ireland consisted of " ten 
ships with beds," as much as to say they were large enough 
to contain sleeping-berths. 

The most general Irish name for a ship is long ; which 
in Cormac's Glossary (p. 105) is derived from the Saxon 
lang (Eng. long) ; but it is more likely that both the Irish 
and Saxon words are cognate with Latin longa. Some- 
times the word lestar (' vessel ') was used. A ship filled 
with men for a warlike excursion was often called laech- 
lestar, ' hero-vessel ' (laech, hero), which was the Irish term 
for a " man-of-war. "J Bare, another Irish word for ship, 
is not, as might be supposed, a loan-word from English, for 
it is used in our oldest manuscript tales. In the Senchus 
M6r§ the word noe is used as an equivalent for curach, 
where a " noe of one hide " is mentioned ; and in Cormac's 
Glossary (p. 125) the same word is given in the form of 
nai, as meaning a ship, and derived from Lat. navis : here 

* O'Grady, Silva Gad., 352 bot. ; Ir. Text, 314, top. 

f LL, 115, b, 22. For more about curraghs and boats in general, see 
Ware, Antiqq., xxiv. : Lynch, Cambr. Ev., chap. xii. : O'Flaherty, Ogyg. 
in. xxxiv. : and Kilk. Arch. Journ., 1852-3, p. 71. 

% Mac Congl., 34, bot. 

§ Br. Laws, 1. 170, top line : see also p. 424, supra 


no doubt, as in many similar cases, the two words are 
cognate and derived from a source older than either. This 
word noe or nai has been long obsolete : but the diminutive 
form noemhog or naomhog, which is pronounced naevogue, 
is still in use for a curragh in the south of Ireland. In 
Kilkee and elsewhere on the western coast you will see 
plenty of canvas-covered curraghs : but they do not call 
them curraghs or naevognes : " canoe " is now the word in 
Kilkee. It is curious to see the middle v sound, which is 
lost in noe, restored in naomhog. Other names for vessels 
will be brought in as we go along. 

In some old texts the word laidheng [lee-ang] is used 
for a boat : and from the expression, " stitched laidheng," 
in one of them, we may infer that it was another name for 
a curragh.* In O'Clery's Glossary libheam [livern] is 
given as the equivalent of long, a ship. The Irish for an 
oar is rdmh or ramha, which seems a loan-word from Lat. 
remus. An oar or paddle was sometimes called sluasat, 
which is the common word for a shovel. A sail is called 
seol [shole] or brat : and a mast, crann, literally a " tree.' 
The usual word for a pilot is luamaire [loomere]. For 
the names of various other parts of ships and boats, see 
" Mac Conglinne," p. 84. 

There were two words to denote a fleet of ships or 
boats : — hinges [ling-as], which is merely a derivative from 
long ; and cobhlach [cowlagh], from cabhal or cobhal, a ship, 
which seems connected with, or perhaps borrowed from, 
Lat. caupulus, a small ship. What great numbers of boats 
were in the fleets that navigated the rivers and lakes may 
be gathered from a record of the Four Masters under a.d. 
751, of the shipwreck of the people of Dealbhna Nuadhat 
in the present county Roscommon, on Lough Ree, when 
twenty-nine out of thirty vessels were lost in a storm, and 
their crews were drowned with their chief. Tigernach 

* Wars of GG., 40, s . Moylena, 45, „. 


records that " the large fleet of Cormac mac Airt (third 
century) was over the sea for the space of three years." 

Ferry-boats were in common use in rivers ; and they 
are often mentioned in the Brehon Laws as subject to 
strict regulations. Cormac's Glossary explains the word 
ethur or ethar : modern form eithear : both pron. eher) as 
a boat that " goes from brink to brink of the river " ; 
O'Clery's Glossary gives the same word as equivalent to 
arthrach iomchuir, a ' boat for carrying ' or (ferrying) : 
and the Senchus M6r gives a similar explanation : — 
" Ethur [a boat which] ferries from bank to bank."* These 
ferryboats were sometimes owned by individuals, and were 
sometimes the common property of the people living round 
the ferry. If a church or monastery happened to be near 
a river where there was no bridge or ford, the inmates 
usually kept a little ferry-boat for their own convenience 
and for the free use of travellers, f 

A ferry-boat, when not in use, was commonly kept high 
and dry on the bank for its better preservation. Where a 
boat was common property, it was usually given in charge 
to one individual, who was then the regular ferryman. 
When it was not so in charge, each person used it as he 
needed : but he was responsible for leaving it on the bank 
safe and uninjured when he was done with it. Those 
using it had to bring it down to the water — or help to do 
so — and where possible bring it back again : and the law 
is very precise in regulating the fine for any injury in 
moving it either way. J We have seen (p. 66) that the 
ferry-boat of a crannoge was often kept floating in the 
middle of the channel, with connecting ropes extending to 
land on both sides. Pleasure boating parties were usual 
in those days as well as now ; and young folk were just 
as inclined to indulge in boisterous merriment ; of which 

* Br. Laws, i. 126, top line. 

f Br. Laws, in. 211, l8 . Dr. Healy, Irel. Anc. Sch., 427, l6 . 

J For all these rules about ferry-boats, Br. Laws, in. 209, 211. 


it would seem the Brehon Law was in a way conscious ; 
for it prescribes compensation in case the boat was injured 
during a pleasure excursion. 

5. Foreign Commerce. 

Many passages referring to the communication of 
Ireland with the outer world in ancient times will be 
found scattered through this book : but it will be con- 
venient to collect here under one heading a few special 
notices bearing on the point. In the native Irish literature, 
as well as in the writings of English, Anglo-Irish, and 
foreign authors, there are many statements showing the 
intercourse and trade of Ireland, both outwards and 
inwards, with Britain and Continental countries. To begin 
with early foreign testimony : — The island was known to 
the Phoenicians, who probably visited it ; and Greek 
writers mention it under the names Iernis and Ierne, 
and as the Sacred Island inhabited by the Hiberni. 
Ptolemy, writing in the second century, who is known to 
have derived his information from Phoenician authorities, 
has given a description of Ireland much more accurate 
than that which he has left us of Great Britain. And 
that the people of Ireland carried on considerable trade 
with foreign countries in those early ages we know from 
the statement of Tacitus, that in his time — the end of the 
first century — the harbours of Ireland were better known 
to commercial nations than those of Britain.* The natural 
inference from these scattered but pregnant notices is that 
the country had settled institutions and a certain degree 
of civilisation — with more or less foreign commerce — as 
early at least as the beginning of the Christian era. 

These accounts, and others from foreign sources that 
might be cited, are fully confirmed by the native records. 
There are numerous passages in Irish literature — in the 

* For all the above, see Moore, Hist, of Irel., vol. 1., chap, i., and the 
authorities he refers to. 


Book of Rights, for instance — in which axe mentioned 
articles of luxury, dress, gold and silver ornaments, swords, 
shields, slaves, &c, imported from foreign lands. One of 
the Law Tracts mentions a foreign axe (probably from 
Gaul) as in use in Ireland, and in terms too implying that 
it was highly valued.* In another authority, O'Davoren's 
Glossary, a foreign axe is noticed in the following curious 
terms : — " A foreign axe {Gall-Mail) perfect . . . with its 
two black ears : "f and here it is set down as worth sixteen 
scripuls, or about the value of a good milch cow. The 
following incident in the Life of St. Columbanus is illus- 
trative of the intercourse between Ireland and the coast of 
Gaul in the beginning of the seventh century : — When the 
authorities of Nantes wished to get rid of him by sending 
him back to Ireland, in the year 610, there was no diffi- 
culty about the conveyance, for they found ready in the 
harbour a ship which was " engaged in the commerce of 
the Scots."| 

A section of the Book of Aicill (Br. Laws, in. 423) 
is given up to muir-bretha or ' sea-laws,' namely, those 
relating to trading vessels arriving on the Irish coast, 
some from Britain, some from Continental countries : a 
circumstance that of itself indicates constant traffic by 
sea. This section mentions, as one of the mediums of 
exchange between the Irish and their foreign visitors, 
an " escup-vessel " of wine or of honey : and Cormac's 
Glossarj'§ explains a " wine-escop " (epscop or escop-fina) 
as being " a vessel for measuring wine among the mer- 
chants of the Norsemen and Franks." In the account of 
the great triennial fair of Carman in Kildare (see page 
444, infra) we are told that there were three markets, 
one of which was " a market of foreigners selling articles 

* O'Curry, Man. & Cust., n. 29, bot. 

t Corm. Gloss., xn. : Three Ir. Gloss. 70, under " Cailech." 

X Reeves, Adamn., 57, note d ; Lanigan, 11. 282. 

§ Corm., p. 67 : see also " Esbicul," p. 69, same Glossary. 


of gold and silver," who sold " gold [ornaments] and noble 
clothes " : so that the fame of this fair found its way to the 
Continent and attracted foreign merchants with their goods. 
In the beginning of the fifth century, when St. Patrick, 
escaping from slavery — as we are told in his " Confession " 
— arrived on some part of the coast of Ireland, he found a 
ship about to set sail : engaged of course in commerce.* 

This commerce was not confined to the coasts. In 
the " Life of St. Ciaran " it is related that on a certain 
occasion a cask of wine was brought by merchants to 
Clonmacnoise from the land of the Franks, f Wine was 
imported too at a much earlier time than this, as we 
know from the Memoir of St. Patrick written by Muirchu 
in the seventh century, which mentions that when the saint 
arrived at Tara on Easter Sunday, a.d. 433, he found King 
Laegaire [Laery] and his nobles feasting and drinking 
wine in the palace (manducandibus illis et bibentibus 
vinum in palatio Temoriae).J 

In the native legends and semi-legendary history, as 
in the strictly historical Irish writings, there are constant 
allusions to foreign intercourse and intermarriages : all 
reflecting historical reality. In the Battle of Rossnaree 
(Book of Leinster) it is stated that an embassy was sent to 
some foreign countries from Concobar mac Nessa, and that 
the pilot who went with them was Cano the foreigner 
(Cano Gall), " to teach them the way over the surface of 
the sea."§ The wife of Eochaidh, king of the Firbolgs, 
was Taillte, daughter of the king of Spain, from whom 
Tailltenn in Meath took its name.|| The various royal 
families of Ireland, from the fifth to the eighth century, 
intermarried among those of Scotland and Britain quite 
as much as among those of their own country : so that in 

* Trip. Life, 362, top. Meyer, in the Courtship of Emer, 

f Stokes, Lives of SS., 276 : p. 303, note 5, illustrative of the 

Adamnan, 57, note d. intercourse of the Irish with the 

% Trip. Life, 282. Scandinavians. 
§ Rossnaree, 13. See also Kuno || O'Curry, Man. & Cust., 1. 148 


most of the great wars and battles in Ireland we read of 
the kings and chiefs of both sides being joined by con- 
tingents from their relatives in those countries.* 

Giraldus Cambrensis has been quoted (i. 164, supra) 
for the statement that slaves were imported in great 
numbers from England, the chief mart for this trade 
being Bristol : and our own records show that foreign 
slaves — " slaves without Gaelic," to quote the old writer's 
expression, i.e. not speaking Irish — were imported (vol L, 
p. 165, supra). The Senchus M6r, in setting forth the law 
of distress for certain articles, names among them a lock 
for securing things brought from beyond the sea : and the 
Gloss explains this as meaning young foreigners ; which 
possibly may point to slaves imported from the Continent. f 

The various articles mentioned here as brought from 
foreign lands were imported to supplement the home 
produce ; in which there was nothing more remarkable 
than our present importation of thousands of articles from 
foreign countries, all or most of which are also produced at 
home. The articles anciently imported were paid for in 
home commodities — skins and furs of various animals, 
wool and woollens, oatmeal, fish, salted hogs, &c. 

At a comparatively late time — the twelfth century — 
Giraldus Cambrensis tells us that the Irish exchanged 
their home produce — chiefly hides — with France, especially 
Poitou, for wine : which agrees with the incident related 
above about Clonmacnoise. Giraldus also relates that when 
the Anglo-Normans under Robert Fitzstephen came to 
attack Wexford in 1169, many ships lay in the harbour, 
among which was one lately arrived from the coast of 
Britain laden with corn and wine. 

Long after the Anglo-Norman Invasion the export 
and import trade continued. We know that in the 
thirteenth century Irish woollen cloth was exported to 

* As illustrative of this, see Dr. Healy, Anc. Ir. Sch., 25, bot. : Moyrath, 
45. 47> 65 ; and vol. 1., p. 79, supra. \ Br. Laws, 1. 127, 3 . 143, l6 . 


England : and in the next century we find Irish frieze — 
or " fryseware," as it is called — mentioned as being freed 
from " aulnage " or duty when exported to England. In 
1300, the army of Edward I., while in Scotland, was sup- 
plied with wheat, oats, oatmeal, pease, beans, wine, beer, 
salt, and hogs, purchased in Ireland.* The " Libel of 
English Policie " (about 1430), p. 199, has the following 
passage enumerating the exports of Ireland : — 

" I caste to speake of Ireland but a litle : 
Commodities of it I will entitle, 
Hides, and fish, Salmon, Hake, Herrynge 
Irish wooll, and linen cloth, faldinge [a coarse kind of cloth] 
And marterns [martens] goode ben her marchandie, 
Hertes [harts'] Hides, and other of Venerie [hunting]. 
Skinnes of Otter, Squirell and Irish hare, 
Of sheepe, Lambe, and Foxe, is her chaffare [merchandise], 
Felles [skins] of Kiddes, and Conies great plentie. 


Of siluer and golde there is the oore." 

This trade continued and increased as time went on, 
There appeared life and activity everywhere, and the 
country was becoming great and prosperous. But all 
this came to a sudden end ; for the manufactures and 
commercial prosperity of Ireland were swept off the face 
of the earth in the seventeenth century by the lawsf 
made to destroy Irish trade ; a blow which at once reduced 
the country to poverty, and from which it has never 

* " Introduction towards a History of Irish Commerce," by William 
Pinkerton ; Ulst. Journ. Archaeol., in. 177. 

f For these laws, see Jsyce, A Child's History of Ireland, p. 394. 


Ornament : composed from the Book of Kclls, 



Section i. The Great Conventions and Fairs. 

ook their rise in Funeral Games.— Public 
assemblies of different kinds, held 
periodically, for various purposes and 
with several designations, formed a marked 
and important feature of social life in 
ancient Ireland. Most of the great meet- 
ings, by whatever name known, had their origin in Funeral 
Games. Tara, Tailltenn, Tlachtga, Ushnagh, Cruachan, 
Emain Macha, and other less prominent meeting-places, 
are well known as ancient pagan cemeteries, in all of which 
many illustrious semi-historical personages were interred : 
and many sepulchral monuments remain in them to this 
day. In the account given in the Book of Ballymote of 
the triennial fair of Carman or Garman, in Kildare, we 
are told that when old Garman, a chief who was contem- 
porary with the heroes of the Red Branch, was dying, 
" they made his grave there ; and he begged of them to 
" institute a fair of mourning (aenach n-guba) for him, and 
" that the fair and the place should bear his name for 
" ever " : and accordingly the place took his name (Carman) 



and the Fair was held there for ages afterwards.* The 
double purpose is shown very clearly in the account of 
the origin of Carn-Amhalgaidh [Awly], near Killala : — 
" Carn-Amhalgaigh, i.e. of Amhalgaidh, son of Fiachra- 
" Ealgach, son of Dathi, son of Fiachra. It was by him 
" that this earn was formed, for the purpose of holding a 
" meeting (aenach) of the Hy Amhalgaidh around it every 
" year, and to view his ships and fleets going and coming, 
" and as a place of interment for himself, "f In the Dinn- 
senchus, as well as in other authorities, we are told that 
Oenach Macha, i.e. the annual fair-meeting at Emain, was 
established to lament the death of Queen Macha of the 
Golden Hair, who had founded the palace there. J 

Important affairs of various kinds, national or local, 
were transacted at these meetings. The laws were publicly 
promulgated or rehearsed to make the people familiar with 
them. There were councils or courts to consider divers 
local matters — questions affecting the rights, privileges, 
and customary usages of the people of the district or 
province — acts of tyranny or infringement of rights by 
powerful persons on their weaker neighbours — disputes 
about property — the levying of fines — the imposition of 
taxes for the construction or repair of roads — the means 
of defence to meet a threatened invasion, and so forth. 
These several functions were discharged by persons 
specially qualified. In all the fairs there were markets for 
the sale and purchase of commodities, whether produced 
at home or imported. 

Some meetings were established and convened chiefly 
for the transaction of serious business : but even at these 
there were sports in abundance : in others the main object 
was the celebration of games : but advantage was taken of 

* O'Curry, Man. & Cust., 529, s . For another account see same vol., 
P- 535, verse 19 : and LL, 215, a. 

f Book of Lecan, cited in Petrie, Round Towers, 108 
X Stokes, Rennes Dinds., in Rev. Celt., xvi. 45. 


the occasions to discuss and settle important affairs, as will 
be described further on. The word Fes or Feis [faish], 
which literally means a feast or celebration, cognate with 
Latin festum and English feast, was generally applied to 
the three great meetings of Tara, Croghan, and Emain (FSis 
Temrach, Feis Cruachan, and Feis Emna, respectively). 
These were not meetings for the general mass of the 
people, but conventions of delegates who represented the 
kingdoms and sub-kingdoms, i.e. the states in general of 
all Ireland, who sat and deliberated under the presidency 
of the supreme monarch.* 

The Feis-Temrach or convention of Tara, according to 
the old tradition, was founded by Ollam Fodla [Ollav- 
Fola], who was king of Ireland seven or eight centuries 
before the Christian era. It was originally held, or 
intended to be held, every third year, but within the 
period covered by our authentic records, it was generally 
convened only once by each king, namely at the beginning 
of his reign, or if at any other time it was on some special 
emergency. The provincial kings, the minor kings and 
chiefs, and the most distinguished representatives of the 
learned professions — the ollaves of history, law, poetry, 
medicine, &c. — attended. It lasted — as we read in some 
authorities — for seven days, namely from the third day 
before Samain (ist November) to the third day after it : 
but according to other accounts it continued for a whole 
month, i.e. " a fortnight before Samain and the day of 
Samain, and a fortnight after " ; and still another makes it 
six weeks, f Possibly the sports lasted for a week, like 
those of Carman (p. 441, below) : but the meetings of 
delegates and ollaves for the discussion of important 
public affairs were held on — like our present parliaments 
— for some weeks longer. Each provincial king had a 

* Keating, 414, 418. 

t Stokes, Lives of SS., Pref., xxxiii : Stokes, Acallamh, line 5367, and 
p. 47, two last lines ; see also Silva Gad. 142, bottom. 


separate house for himself and his retinue during the time ; 
and there was one house for their queens, with private 
apartments for each, with her attendant ladies. There 
was still another house called Relta na bh-filedh [Railtha- 
na- villa], the ' star of the poets,' for the accommodation 
of the poets and ollaves of all the professions, where 
also these learned men held their sittings.* Every day 
the king of Ireland feasted the company in the great 
banqueting-hall — or, as it was called, the Tech Midchuarta 
or ' mead-circling-hall " — which was large enough for a 
goodly company : for even in its present ruined state it 
is 759 feet long by 46 feet wide. 

In the same hall were held the formal meetings for the 
transaction of important business, such as proclaiming the 
Laws, making new regulations for the whole country where 
necessary, examining and checking the historical records 
of the kingdom, and correcting them if found defective or 
wrong. All these functions were discharged by experts : 
and at the end of the fits the whole proceedings were 
written by properly qualified ollaves in the national record 
called the Saltair of Tara. These are the accounts left 
us in our oldest traditions. That the meetings were 
held here is not however a matter of tradition, but of 
unquestionable history. The last Feis Temrach was con- 
vened by Dermot king of Ireland, in a.d. 560, after which 
Tara was abandoned as a royal residence, on account of a 
curse pronounced against it in very solemn fashion by 
St. Ruadan of Lorrha in Tipperary. 

According to the account given by Keating, who took 
the statement from old authorities now no longer existing, 
the conventions of Emain and Croghan were largely con- 
cerned with industrial affairs. The ollaves and nobles — 
as already stated at page 329 — selected from many candi- 
dates a number, the best of each craft, who were, as we 
should express it, " certificated " as persons duly qualified 

* Keating, 414, from old authorities. 


to practise their several trades, each in his own district : 
which gave them at once legal standing and legal rights 
in all cases of dispute.* 

The dal [dawl] was a meeting convened for some 
special purpose commonly connected with the tribe or 
district : a folkmote.f A mordal or arddal (tndr, great : 
drd, high) was a great, or chief, or very important assembly. 
These two last terms are often applied to such assemblies 
as those of Tara, Tailltenn, and Ushnagh. 

The aenach or fair was an assembly of the people of 
every grade without distinction : it was the most common 
kind of large public meeting, and its main object was 
the celebration of games, athletic exercises, sports, and 
pastimes of all kinds. In Cormac's Glossary! an aenach 
is well characterised as a place where there were " food 
" and precious raiment, downs and quilts, ale and flesh- 
" meat, chessmen and chessboards, horses and chariots, 
" greyhounds and playthings besides." In a still older 
authority, the story of the Sick Bed of Cuculainn§ in the 
Book of the Dun Cow, copied from the Yellow Book of 
Slane, we read : — " That was the period of time which the 
" Ultonians devoted to the holding of the fair of Samain 
" in the plain of Murthemne [the level part of the County 
" Louth] every year : and nothing whatever was done by 
" them during that time but games and races, pleasure and 
" amusement, eating and feasting : and it is from this 
" circumstance that the Trenae Samna (' three days of 
" Samain ') are still observed throughout Erin." 

The Fairs of Tailltenn, Tlachtga, Ushnagh, The Curragh, 
Nenagh, Aenach-Beag. — The most important of the Aenachs 
were those of Tailltenn, Tlachtga, and Ushnagh. The Fair 
of Tailltenn, || now Teltown on the Blackwater, midway 

* See O'Conor, Dissert., 42 : and O'Flaherty, Ogyg., Part in., chap, lvi.: 
Keating, 419. t Da h forum, Zeuss, 71, ^ 

X Corm., page 129, " Ore treith." § Atlantis, 1. 371. 

|| I have had the advantage of perusing Mr. Edward Gwynn's Todd 
Lecture on the Aenach Tailtenn, of which he lent me the manuscript. 


between Navan and Kells, was attended by people from 
the whole of Ireland, as well as from Scotland, and was 
the most celebrated of all for its athletic games and 
sports : corresponding closely with the Olympic, Isthmian, 
and other games of Greece. It was held yearly on the 
ist August, and on the days preceding and following. 
What vast numbers were congregated during these games 
will be seen from the Four Masters' record of the last 
official aenach held there, a.d. 1169, by Roderick O'Conor, 
king of Ireland, when the horses and chariots alone, 
exclusive of the people on foot, extended in a continuous 
line from Tailltenn to Mullach-Aiti, now the Hill of Lloyd 
near Kells, a distance of more than six English miles. 
This aenach was originally instituted, according to the 
old legend, by the De Dannan king Lugad, or Lug of the 
Long Arms, to mourn and commemorate his foster-mother, 
Tailltiu, who was buried there under a mound, and from 
whom the place took its name. From Lug the first of 
August was named Lugnasad, meaning the nasad or games, 
of Lug : a name still in use. 

Marriages formed a special feature of this fair. " From 
" all the surrounding districts the young people came with 
" their parents, bachelors and maidens being kept apart in 
" separate places, while the fathers and mothers made 
" matches, arranged the details, and settled the dowries. 
" After this the couples were married, the ceremonies being 
" always performed at a particular spot."* Hence, accord- 
ing to Cormac's Glossary (p. 48), a hillock there had the 
name of Tulach-na-Coibche, " the hill of the buying," 
where the bride-price was paid. All this is remembered 
in tradition to the present day : and the people of the 
place point out the spot where the marriages were per- 
formed, which they call " Marriage Hollow." The remains 
of several immense forts are still to be seen at Teltown, 

* From Joyce, Short History of Ireland, 90. 


even larger than those at Tara, though not in such good 

O'Donovan carefully examined this historic site in 1836 
for the Ordnance Survey, and found among the people 
vivid traditions of the old customs. Though the younger 
generation, when speaking English, called it Teltown, the 
older Irish-speaking people never used any name but 
Tailltenn. They told him too that games were carried 
on there " down to 30 years ago " — i.e. to 1806 — but that, 
on account of the increasing manufacture of pottheen 
whiskey — instead of the old native drinks, ale and mead — 
there were quarrels and scenes of violence, so that the 
magistrates at last put a stop to the meetings, f 

The meetings at Tlachtga and Ushnagh, which have 
already been mentioned, seem to have been mainly pagan 
religious celebrations : but games, buying and selling, and 
conferences on local affairs, were carried on there as at the 
other assemblies. One of the most noted of all the fairs 
was Aenach Col main on the Curragh of Kildare, which is 
noticed in sect. 5 below (p. 464) in connexion with races. 
The memory of one important fair is preserved in the name 
of Nenagh in Tipperary, in which the initial N is the Irish 
article an, ' the ' : N-enagh, ' the fair.' The yearly fair 
held here was called Aenach-Urmhumhan [Enagh-Uroon], 
meaning the Fair of Urmhumha, i.e. of Ormond or East 
Munster : and the old people still call Nenagh " Aenach- 
Uroon " ; but they have quite forgotten the meaning 
of uroon. So also Monasteranenagh in Limerick, the 
' Monastery of the Fair,' which in old times, before 
the monastery was thought of, was called Aenach-beag, 
' Little Fair,' to distinguish it from the Great Fair of 
Nenagh. I 

* See Wilde, Boyne, 149, 150 : Stokes, Life of Petrie, 366. 
f O'Donovan's Ord. Surv. Letters, Roy. Ir. Acad. (Meath) • Letter on 
the parish of Teltown. 

J For more on these see Irish Names of Places, 1. 204-206. 


2. The Fair of Carman. 

The people of Leinster held a provincial aenach at 
Carman, a place situated probably in South Kildare, once 
every three years, which began on Lughnasad [Loonasa], 
i.e. the first of August, and ended on the sixth. It was 
considered so important to hold this fair that in case the 
Leinstermen should ever neglect it — a very unlikely thing — 
the poem in the Book of Leinster (p. 215, a) giving an 
account of the celebration, threatens them with many 
evils — early greyness ; baldness (see p. 182, supra) ; feeble- 
ness ; kings without wisdom or dignified manners, without 
hospitality, without truthfulness. But if the fair was duly 
held, they were promised various blessings — plenty and 
prosperity, corn, milk, fruit, and fish, in abundance ; and 
freedom from subjection to any other province.* 

Fortunately we have, in the Book of Leinster, the Book 
of Ballymote, and some other ancient manuscripts pretty 
full descriptions — chiefly poems — of this particular aenach. 
The poem in the Book of Leinster was written by a poet 
named Fulartach, about a.d. 1000. The several accounts, 
which are printed in the second vol. of O'Curry's " Man. 
& Cust." (Appendix) differ somewhat in detail : but the 
following abridged description, drawn from all, will give 
a good idea of the arrangements and proceedings, not only 
of this fair, but of all others wheresoever held.f The 
representative character of the Fair of Carman, as intended 
for the whole of Leinster, will be seen from the statement 
in one of the old accounts that forty-seven sub-kings or 
chiefs of the province [with their people] attended it : — 
viz. sixteen from Carman itself and the surrounding dis- 
tricts ; eight from the territory of Hy Donoghue around 

* O'Curry, Man. & Cust., 11., pp. 547 and 531. 

t Cuan O'Lochain's Poem in the Book of Leinster (LL, 200, b ; Man. 
& Cust., 1. 148) describes the Fair of Tailltenn somewhat similarly, but 
not nearly in such detail. 


the River Dodder near Dublin ; twelve from the Plain of 
Maistiu, i.e. the district round Mullamast in Kildare ; five 
from Fidh-Gabhla, now Figile in King's County ; and six 
from Ossory.* 

There was much formality in the arrangements. While 
the chief men were sitting in council under the king of 
Leinster, who presided over all, those belonging to the 
several sub-kingdoms had special places allotted to them 
in the council-house or enclosure, which were jealously 
insisted on. The forud or sitting-place of the king of 
Ossory — a sub-kingdom of Leinster — was on the right 
hand of the king of Leinster, and that of the king of Offaly 
on his left : and each of the other sub-kings had his own 
special place assigned to him. Each day but the last 
appears to have been given up to the games of some 
particular tribe or class. For instance, we are told the 
people of Ossory had a special day for themselves for 
what was called the " steed contest of the Ossorians," 
i.e. for their chariot and horse races, f Another day was 
set apart for royalty, when roydamnas or crown princes 
contended, and none others were permitted to enter 
(Man. & Cust., II., p. 539, last verse). 

Women played a conspicuous part in this fair, and of 
course in all others. There were special cluichi or games 
for them in the afternoon which are called cluichi ban 
Laigen iar lo, ' the games of the women of Leinster in 
the evening,' but what kind they were we are not told. 
To the Laisig (i.e. the people of Laighis or Leix) was 
entrusted the important and delicate duty of superin- 
tending these games ; and they were responsible not only 
for the good order of the proceedings, but also for the 
safety of the jewellery, which the Leinster women wore 
in abundance, and which they had to lay aside during 
the games. J 

* O'Curry, Man. & Cust., 11., p. 539. t Ibid., 529, par. 4 

j Ibid., 539, verses 44, 45. 


The women had airechts or councils of their own to 
discuss those subjects specially pertaining to women : and 
at these assemblies no man was permitted to be present : 
while, on the other hand, no woman was allowed to enter 
the special council meetings of the men (Man. & Cust., n., 
p. 543, ver. 55). In those formal sittings that were open 
to both sexes, the women were seated with their own people, 
in the special places set apart for the representatives of their 
respective tribes (ibid., p. 529, par. 4). 

Conspicuous among the entertainments and art-per- 
formances was the recitation of poems and romantic tales 
of all the various kinds mentioned in I. 533, supra, like 
the recitations of the Rhapsodists among the Greeks : — 
" The Tales of the Fena of Erin " — says the old account — 
" a never- wearying entertainment : stories of destructions, 
" cattle-preys, courtships, rhapsodies, battle-odes, royal 
" precepts, and the truthful instruction of Fithil the Sage : 
" poets and learned men with their tablets and books of 
" trees (1. 480, supra) : deep poetry, and Dinnsenchus or. 
" History of Places : the wise precepts of Carbery and 
" Cormac Mac Art " (Man. & Cust., 11. 543). For all of 
these there were sure to be special audiences who listened 
with delight to the fascinating lore of old times. 

Music always formed a prominent part of the amuse- 
ments : and among the musical instruments are mentioned 
cruits or harps ; timpans ; trumpets ; wide-mouthed horns; 
cuisig or pipes ; and there were plenty of harpers ; pipers ; 
fiddlers ; bone-men (cnamh-fhir) , i.e. castanet-players ; 
" tube-players " ; and fir-congail or ' chain-men,' probably 
men who shook music from chains furnished with little 
bells like those already described in vol. 1., page 586, 
supra. In another part of the fair the people gave them- 
selves up to uproarious fun, crowded round showmen, 
jugglers, and clowns with grotesque masks or painted 
faces, making hideous distortions, all bellowing and roaring 
out their rough jests to the laughing crowds : for there 


were " professors of every art, both the noble arts and the 
base arts."* There were also performers of horsemanship, 
who delighted their audiences with feats of activity and 
skill on horseback, such as we see in modern circuses. 
The Brehon Law (v. 109) mentions " equestrians, namely, 
those who stand on the backs of horses at fairs." 

Prizes were awarded to the best performers of " every 
" dan or art that was just or lawful to be sold, or rewarded, 
" or exhibited, or listened to " : which excluded from any 
prize, showmen and all other exhibitors of the baser sortf : 
and at the close of the proceedings the coveted trophy — 
always a thing of value, generally a gold ring, or some 
other jewel — was publicly presented by some important 
person, such as a king, a queen, or a chief. 

Special portions of the fair-green were set apart for 
another very important function — buying and selling. We 
are told that there were " three [principal] markets : viz. a 
" market of food and clothes : a market of live stock and 
" of horses ; while a third was railed off for the use of 
" foreign merchants with gold and silver articles and 
" fine raiment to sell. "J There was the " slope of the 
embroidering women," who did their work in presence of 
spectators (Man. & Cust., 11., p. 547, verse 76). A special 
space was assigned for cooking (verse 76), which must have 
been on an extensive scale to feed such multitudes. 

On each day of the fair there was a conference of the 
brehons, chiefs, and leading men in general, to regulate the 
fiscal and other local affairs of the province for that and 
the two following years : or, as the old account has it, " for 
" considering the judgments and rights of the province for 
" three years. "§ 

Possibly some readers may think it strange that in all 
this detailed list of amusements we do not find a word 

* O'Curry, Man. & Cust., n. 545, verse 63 : and 531, note, line 16. For 

showmen's face-distortions, see p. 486, below. f Ibid., 531, note, line 20 

X Ibid., 531 ; 547, verse 75. § Ibid., 543, verses 53, 54 ; 530 note. 


about dancing. There is, in fact, no evidence that the 
ancient Irish ever danced to music, or danced at all, i.e. 
in our sense of the word " dancing " ; but very strong 
negative evidence that they did not. Though we have in 
the old literature many other passages in which the several 
amusements at popular gatherings are enumerated,* on no- 
one of them is dancing mentioned. This curious fact has 
been already noticed by O'Curry, who, after all his vast 
reading of native literature, says : — " As far as I have ever 
" read there is no reference that can be identified as con- 
" taining a clear allusion to dancing in any of our really 
" ancient ms. books."f So also Stokes : — " Dancing is not 
" mentioned in the documents now published [in his Trip. 
" Life], nor, indeed, in any Irish mss. that I have read." 
(Trip. Life, clviii). We have now two Irish words for 
dancing : — damhsa, which may be passed over, as it is 
obviously a modern adaptation of the English " dance " : 
and rinncedh, which is a native word, derived, according 
to 0'Curry,| from rinn, an old word for foot (see Corra. 
145, "Rind"), and the termination cedh: so that rinncedh 
literally means ' footing.' But it does not seem to be an 
old word. 

What appears to be a sort of confirmation of all this 
occurs in a passage in a Homily on the Passion of John 
the Baptist, published by Dr. Atkinson from the Liber 
Brecc, where the Irish homilist is giving a free rendering 
of the Bible narrative about Herodias' daughter dancing 
before King Herod. He had before him the Vulgate word 
saltavit (' she danced ') : but it would almost appear as if 
he did not know how to render it into Irish, possibly not 
knowing what " dancing " was.§ His words are that the 
girl was skilled in clesaighecht ocus lemenda ocus opairecht, 

* As, for instance, in the Courtship of Emer, 69. 

j O'Curry, Man. & Cust., 11. 406. J Ibid., 11. 407. 

§ Something like what happened when the Irish annotator, having 
never seen a fiber or beaver, but knowing well what an otter was, explained 
fiber by dobran, or ' otter.' (See p. 462 farther on.) 


' juggling and leaping and activity ' ; where lemenda simply 
means ' leaping/ but not ' dancing ' as such.* So far as I 
am aware, the Irish words leim, Uimenn, Umenda mean 
exactly the same as the English words " leap," " leaping," 
and nothing more. When we now wish to express in Irish 
the special sense of " dancing," the word leimenn, ' leaping,' 
will not answer : we have to employ a different word 
(rincedh or damhsa) : just as we have to do in English. 
But the Irish translator had no word but one : and accord- 
ingly used the Irish word lemenda in its primary sense of 
leaping merely, to represent " saltavit." In the " Circuit of 
Muirchertach mac Neill " (p. 45), which celebrates an event 
that took place a.d. 941, the English translation has — 
" Music we had on the plain and in our tents, listening to 
its strains we danced awhile " : but the three last words are 
inserted by O'Donovan (in italics), and have no corre- 
sponding words in the original. Yet the men kept time 
with the music — as the poem expresses it — " by the shaking 
of our hard cloaks " : but not by dancing. 

When the evening of the last day had come, and all 
was ended, the men of the entire assembly stood up and 
made a great clash with their spears, each man striking 
the handle of the next man's spear with the handle of his 
own : which was the signal for the crowds to disperse, f 
It always took two years to make the preparations for 
the holding of this fair 4 After the introduction of 
Christianity in the fifth century, the pagan customs were 
discontinued, and Christian ceremonies were introduced. 
Each day was ushered in with a religious exercise, and on 
the next day after the fair there was a grand ceremonial : 
Masses and adorations and singing of hymns.§ But beyond 
this there was little or no change. 

We have seen that a fair-green was usually called 
faithche [faha] : and a small portion of the faithche of 

* Homilies, 66, 307 : Matthew, xiv. 6. % Man - & Cust., 11. 531, note, line 4. 
t Man. & Cust., n. 545, verse 70. § Ibid., 11. 545, verse 67. 


some forgotten fair-sports still retains the name in " The 
Fay the," a level spot near the present town of Wexford. 
The correspondence between these fairs and the Greek 
celebrations for similar purposes will be obvious to every- 
one : and it is worth observing that the Carman festival 
bore a closer resemblance to the Isthmian games, where 
there were contests in poetry and music, than to those of 
Olympia, where there were none. 

3. General Regulations for Meetings. 

The accounts that have come down to us show that 
the ancient Irish were very careful that there should be 
no quarrelling or fighting, or unseemly disturbance of any 
kind that might " spoil sport," at the formal dais or aenachs, 
or meetings, for whatever purpose convened. The Senchus 
Mor, and the glosses and commentaries on it, mention fines 
for creating disturbance or being guilty of any misconduct 
while the people were assembled* : and any serious breach 
of rule was punished with death. The Poem on Carman 
says : — 

" Whoever [seriously] transgresses the law of the assembly, 
Which Benen with accuracy indelibly wrote, 
Cannot be spared upon family composition, 
But he must die for his transgression."! 

Whatever causes of quarrel may have existed between 
clans or individuals, whatever grudges may have been 
nurtured, all had to be repressed during these meetings. 
Even proceedings likely to lead to disputes were forbidden, 
such as elopements, repudiation of wives by husbands, or 
the reverse. There were to be no distraints or other pro- 
cesses for the recovery of debts, so that a debtor, however 
deeply involved, might enjoy himself here with perfect 
safety and freedom from arrest. J Hence we find the old 

* Br. Laws, 1. 231, bottom ; 175, bottom ; 177, I2 . 233, last line, 
f Man. & Cust., 11. 543, verse 56. J Ibid., 11. 543, verses 53, 54, 55. 


writer boasting, with natural pride, " the Gentiles of the 
Gael," i.e. the Irish pagans of pre-Christian times, cele- 
brated the fair of Carman " without breach of law, without 
crime, without violence, without dishonour."* Similarly, 
Cuan O'Lochain is at great pains to detail the precautions 
for peace at Tailltenn (LL, 200, b, 40-47). 

A very few cases of serious violation of the law are 
recorded, as when " Fogartach O'Carney disturbed the 
fair [of Tailltenn], for he killed Maelruba the son of 
Dubhsleibhe " (FM, 715). But the annalists record them 
as exceptional : as in the present case, where they 
obviously look upon the breach of the peace as the 
important and unusual circumstance. The reader will 
perceive that all this runs parallel with the sacred armis- 
tice proclaimed by the Greeks at their Olympic and 
Isthmian games. Indeed an expression in Cuan O'Loch- 
ain's poem is almost identical with some phrases in the 
Greek accounts, where he tells us (LL, 200, b, 46) that 
among the multitudes attending the fair of Tailltenn, 
whether from Ireland or Scotland, there was " one uni- 
versal fair-truce " (oen chair de oenig). Where such vast 
numbers of chariots were congregated there was always 
liability to accidents. The law took cognisance of these ; 
and provision was made that in case a chariot should be 
broken, or anyone was injured by furious driving, or should 
any other accident occur, the persons responsible should 
be made liable, but should at the same time be protected 
from vexatious prosecutions, f 

The Law made provision for having the fair-green, and 
particularly that part of it devoted to special purposes, kept 
in proper order. This duty was assigned to certain persons 
of the neighbourhood, whose business it was to clear away 
the brambles and rubbish immediately before the fair, and 
to keep the spaces clear during the sports : and for the 

* Man. & Cust., 11. 537, verse 31. f Br. Laws, in. 265, bottom. 


neglect of this duty there was a penalty.* They were of 
course paid in some way, but on this point we have no 
information. When to anyone was assigned the task of 
making a fair-green, he had to furnish it with fences and 
mounds (claide and ferta), wherever they were required 
for such purposes as jumping, racing, special assemblies, 
&c. : and here also was provided a penalty if the structures 
were not properly made.f 

Besides the large fairs or other assemblies, there were 
smaller meetings for special purposes, such as councils of 
representative men to deliberate on local matters. These 
were generally held in the open air on little hills, J and were 
called airecht, oirecht, or oirechtas,§ from oire or aire, a 
chief or leading man ; for the local king or chief always 
presided at them. The custom of holding oirechts was 
continued down to the end of the sixteenth century. 
Spenser (View, 126) notices them as carried on in his time : 
and the word was familiarly used in an anglicised form by 
English writers of the time of Elizabeth. In the agreement 
between the Anglo-Irish council of Dublin and O'Reilly,* 
in 1583, it is laid down : — " He [O'Reilly] shall not 
" assemble the queen's people [i.e. his own people, over 
" whom in common with all the rest of the Irish the 
" queen claimed to be sovereign lady] upon hills, or use 
" any Irachtes or paries upon hills."|| 

A hill of this kind, set apart for meetings, was some- 
times called tulach, which is a name for any small hill, or 
tulach airechtais, or ard-airechtais, ' hill of meeting ' ; and 
also ri-bheann, i.e. royal benn or hill : i.e. devoted to the 
king's business.^ But it was also designated by the special 

* Br. Laws, 1. 123, 34 . 129, I3 . f Br. Laws, 1. 157, 34 . 159, last line. 

% Moyrath, 67, bottom : Br. Laws, in. 297, l6 . 

§ See " Aireachta " in O'Donovan, Supplement. The above words 
will be found in Irish dictionaries. 

|| Hardiman, Irish Minstrelsy, 11. 159. 

^f Br. Laws, 1. 175, bottom: 177, I2 . Sick Bed, Atlantis, 1. 384, l8 . 
Moyrath, 92, 5 . O'Donovan, Supplement, " Ribheann." 

G I 


name aibinn or aiminn [eevinj, which the gloss on the 
Senchus Mor explains by suide-ddla [see-dawla] i.e. the 
' seat of the ddl or meeting ' : ' convention-seat.'* Hills 
devoted to this particular purpose were held in much 
veneration, and were not to be put to any other use. 
Ferflatha O'Gnive, the ollave poet of the O'Neills of 
Clannaboy in the time of Elizabeth, lamenting the decay 
of the old Gaelic customs, says that now, alas, the sacred 
meeting hills are no longer frequented : they are tilled 
and cropped and used as common market-places, f 

Great care was taken that they should be kept in 
proper order : and anyone who stripped sods from the 
surface or dug into them for any purpose was fined. Cows 
were not permitted to graze on a convention-hill : and if 
the smooth surface happened to get broken up from any 
cause it should be strewn with fine clay which was to be 
trampled down and made smooth before the meeting. If 
the meeting had to be held while the hill happened to be 
bare of grass, or rough, or dirty, the person having the 
management of the ddl should have cloths of some kind 
spread under the feet of kings, and rushes for the other 
chief people. J The very name of these assembly-hills 
seems to indicate that they were deliberately selected for 
their pretty appearance : for aibinn (or aiminn as other 
authorities have it) denotes anything beautiful, cognate 
with Latin amoen-us : and indeed the Law tells us ex- 
pressly that, as the name of a meeting-hill, this is the 
sense it bears.§ 

At small meetings held in a building or any other con- 
fined space, the president, when he wanted silence, shook 
what was called the " chain of attention " (slabra estechta), 
which was hung with little bells or loose links that gave 
forth a musical sound. In the story of the Sons of Turenn 

* Br. Laws, i. 167, 35; 170, IS . 171, 19 . 

f Hardiman, Irish Minstrelsy, II. 106, second verse. 

} Br. Laws, 1. 171, so- in. 297. § Ibid., in. 297 ; iv. 215, m 


(p. 185) we read that Luga of the Long Arms, sitting 
beside the king of Ireland in the hall of Tara, and wishing 
to address the assembly, ordered the chain of attention of 
the court to- be shaken, which procured him silence. More 
often it was a branch hung with little bells : this was called 
craeb [crave], or craeb sida. ' branch of peace.' At the feast 
in the house of Bricriu, when a dangerous dispute arose 
and there was a great and noisy contention, Sencha the 
brehon arose and shook the Craebh Sencha (i.e. ' Sencha's 
branch '), which produced instant silence and attention.* 
This musical branch with silver bells figures in many of 
the romantic tales, f There were other ways of procuring 
attention at feasts and meetings. When there arose some 
noisy confusion at a feast, King Concobar gave one of his 
usual signals by striking the bronze pillar that supported 
the canopy of his couch with his silver do or wand. Some- 
times the president hushed all talk and noise by merely 
standing up, like the Speaker in the House of Commons. 
At the feast at Dun-da-bend (now Mountsandal over the 
Bann near Coleraine), while the talk and enjoyment went 
on without restraint, Concobar, sitting on his " hero-seat " 
at the head, and wishing to speak to the assembly, rose 
up ; and " mute and silent became the Ultonians when 
" they saw the king standing, so that if a needle fell from 
" roof to floor, it would be heard." j 

4. Some Animals connected with Hunting and Sport. 

The Dog. — Dogs of all kinds were used by the people 
of Ireland quite as much in ancient times as they are now : 
but hunting-dogs have, as might be expected, impressed 
themselves most of all on the literature. By far the most 
celebrated of the native dogs was the Irish wolf-dog, noted 

* Fled Brier., 35 : Ir. Texte, 1. 267, a . See also Hyde, Ir. Lit., 296, top. 
f As in Voyage of Bran, 1., p. 2 : and the Story of Cormac Mac Art and 
the Musical Branch, Oss. Soc, in. 213. See vol. 1., p. 586, supra. 
% Mesca Ulad., 13, bottom. 


for its size and fierceness. There is no doubt that this 
gigantic animal existed in Ireland from the earliest times, 
as is proved by unquestionable authorities, one of which 
is quoted below : but it is curious that it is chiefly from 
English and foreign writers we get such precise informa- 
tion as enables us to form an idea of its actual size. It was 
so familiar at home that the native writers did not think 
it necessary to describe it. 

In the ancient Irish tales the hunting-dogs are con- 
stantly mentioned in terms of great admiration, as large 
and strong : but these references are vague, and many 
persons might regard them as high-sounding poetical 
exaggerations. There is nothing poetically vague however 
in the statement of Campion, the English Jesuit, who 
visited Ireland, and wrote a short history of it in 1571. 
He says (p. 13) : — " They [the Irish] are not without 
" wolves, and greyhounds to hunt them, bigger of bone 
" and limme than a colt." Twelve centuries before his 
time, a Roman citizen named Flavianus, who had visited 
Britain, presented seven Irish dogs (Scotici canes) to his 
brother Symmachus, a Roman consul, for the games at 
Rome (a.d. 391) — though we are left in the dark as to 
how he procured the animals — a gift which Symmachus 
acknowledges in a letter still extant : — " All Rome," he 
says, " viewed them with wonder and thought they must 
have been brought hither in iron cages."* Among the 
numerous passages in native Irish writings mentioning 
this great Irish greyhound I can find only one that gives 
an idea of its actual size, quoted by the Rev. Dr. Hogan 
(Wolf-dog, p. 164) from the Book of Lismore, a manuscript 
copied in the fifteenth century from much older sources, 
which states : — " Each of these hounds is as big as an ass." 

* This letter is referred to and partly quoted in Harris's Ware (Antiqq., 
166) : the original Latin passage may be seen in the Rev. Dr. Hogan's 
" Irish Wolfdog," page 153, and a translation of it at page n of the same 


From the fifteenth, to the eighteenth century Irish wolf- 
dogs were, it might be said, celebrated all over the world, 
so that they were sent as valuable presents to kings and 
emperors, princes, grand Turks, noblemen, queens, and 
highborn ladies, in all the chief cities of Europe, and even 
in India and Persia.* It is strange that Giraldus Cam- 
brensis does not notice these dogs : they must have been 
in the country in his time ; and if he had seen them he 
would certainly have mentioned them. After the final 
extinction of wolves in Ireland in the early part of the 
eighteenth century, the need for these great dogs ceased, 
and the race was let die out.f 

The word cu, genitive con, was generally applied to any 
fierce dog, this term being qualified by certain epithets to 
denote dogs of various kinds. A greyhound or hunting-dog, 
whether a wolf-dog or any other, was commonly called 
milchu. In Cormac's Glossary (p. 115) the s}7llable mil is 
explained mdl, a king, so that according to this authority 
milchu is cu-mdl, the hound of a king — which is I fear all 
fanciful. O'Davoren in his Glossary explains milchu by 
gadhar [gy-er ■ g hard] which is still the common word for a 
beagle or small hound. At the present time the most general 
name for a dog is madra or mada, which is also an old word. 

A watch-dog for a house was called archu, from ar or 
air, to watch. These watch-dogs were kept in every house 
of any consequence ; and they were tied up by day and 
let loose by night. In accordance with custom and law 
the watch-dogs of the farming classes were loosed earlier 
in the evening and tied up earlier in the morning than 
those of the chieftain grades : in the chiefs' houses so many 
people were coming and going that the dogs were kept 
tied up till bedtime to avoid danger to guests : whereas 
those of farmers were set free at cow-stalling. J 

* Preface to the Rev. Dr. Hogan's " Irish Wolfdog." 

f See note on the Irish Wolfdog in Stokes's Life of Petrie, p. 437. 

I Br. Laws, 1. 127, „. 145. ., in. 419. 


A shepherd's dog was called a cu-buachaill [coo-boohil], 
i.e. a ' dog-cowherd.' The gloss on the Senchus M6r 
alludes to a penalty for stealing one of these dogs, and 
mentions that they were of three kinds, but does not 
specify them. A story connected with Ireland, in a Norse 
work of the twelfth century — Snorro Sturleson's Chronicles 
of the Kings of Norway — written in Icelandic, shows that 
the ancient Irish trained these dogs as carefully as the 
shepherd's collie is trained at the present day. 

" While Olaf [or Amlaff] was in Ireland [in the ninth or tenth 
century] he was once upon an expedition which went by sea. As 
they required to make a foray for provisions on the coast, some of 
his men landed and drove a large herd of cattle down to the strand. 
Now a peasant came up and entreated Olaf to give him back the 
cows which belonged to him. Olaf told him to take his cows if he 
could distinguish them, ' but don't delay our march.' The peasant 
had with him a large house-dog, which he put in among the herd of 
cattle, in which many hundred head of beasts were driven together. 
The dog ran into the herd, and drove out exactly the number the peasant 
said he wanted ; and all were marked with the same mark, which 
showed that the animal knew the right beasts and was very sagacious. 
Olaf then asked the peasant if he would sell him the dog. ' I would 
rather give him to you,' said the peasant. Olaf immediately presented 
him with a gold ring in return, and promised him his friendship in 

It appears from some passages in the Laws, as well as 
from general Irish literature, that lapdogs were as much in 
favour in Ireland in old times as they are now : women of 
all classes, from queens down, kept them. We find them 
even in convents. The virgin saint Cruimtheris, who lived 
near Armagh in the time of St. Patrick, kept a lapdog 
which she fed on the milk of a doe.f Their importance 
in the eyes of the law is attested by the story of the first 
lapdog brought to Ireland (vol. I., p. 74, supra) : and a 
heavy fine was prescribed for stealing them : which was 

* Laing's Translation, quoted in Kilk. Archseol. Journ., vol. 1., p. 326. 
f Trip. Life, 233. 


recoverable by either husband or wife, though the lapdog 
was always the woman's personal property. 

The commonest name for a lapdog was oircne [urkina]. 
a diminutive of oirc [urk], which means, among other 
things, a little dog. Two other terms for these little pets 
are derivatives from the rootword mes or mess : — meschu, 
which is very often used by the old Irish writers, and 
which O'Clery makes equivalent to oirc ; and mesan or 
messan, which O'Clery's Glossary defines as cu-beag, ' little 
hound.' The Brehon Law (i. 153, bot.) is still more 
explicit in identifying this word with oircne, for the gloss 
on the Senchus Mor, explaining oircne rigna (the ' lapdog 
of a queen '), says it is identical with mesan. In Cormac's 
Glossary (p. 115) the word mesan is derived from messa, 
' the worst,' because a lapdog is " one that is the worst of 
hounds," i.e. I suppose as being merely a plaything, with 
no further use. This word is still current in Scotland even 
among the English-speaking people, and is often met with 
in Scott's novels. In the Heart of Midlothian (chap, xxix.), 
Madge speaks to Jeanie Deans about her " little messan 
dog " — her " puir bit doggie Snap." It has been already 
remarked that little bells were often hung from lapdogs' 

A dog that had the vicious habit of attacking lambs, or 
fowls, or domestic pets, had a muzzle (srublingi) of leather 
tied on his snout. From the words of the Book of Aicill 
it appears that dogs were often muzzled as a general 
precaution. The muzzle should be tested, so that in case 
a dog did mischief the owner might be enabled to mitigate 
damages by the plea that proper precaution was taken. 
An eye-cap, called in Irish Eirrgi, i.e. a covering of leather 
fastened over the eyes, was used f or " a dog which does not 
know its own people from the neighbours," which probably 
means that he was as ready to bite one of the family as to 
bite a stranger.* 

* Br. Laws, in. 417. 


Dogs were liable to run mad then as well as now. 
When a dog was found to be mad, it was hunted down 
and killed, its body was burned, and the ashes were thrown 
into a stream. Here is the quaint language of the Book of 
Aicill on this point : — " There is no benefit in proclaiming 
" it [i.e. sending round warning of a cu-conjaid or mad dog] 
" unless it be killed ; nor though it be killed unless it be 
" burned ; nor though it be burned unless its ashes have 
" been cast into a stream."* 

Some dogs were " lawful " (dlighedh), i.e. they were 
in some way recognised by law, which turned to the 
advantage of the owner in case of proceedings of any 
kind in a brehon's court regarding the dog. Others were 
said to be " unlawful " (indlighedh) , which did not mean 
that they were forbidden by law, but simply that they 
were not legally recognised, and the owner had therefore 
to take his chance in law proceedings, without any benefit 
from legal recognition. Some dogs again were fully lawful, 
some three quarters, and some half lawful ; but these terms 
are not defined in our copies of the laws. Certain dogs, 
stated to be lawful, are named ; but the statement is to me 
not clear. One example of a lawful dog is plain enough — 
a dog with a duly tested muzzle following a woman as a 
companion, f The Book of Aicill lays down detailed rules 
about dog-fights, in view of the injury that might be 
inflicted on bystanders or on other animals. J 

The Greeks, though they looked upon the dog as the 
friend of man, did not hold it in high esteem, and they did 
not use it in war. Among the Celts of Gaul it held a much 
higher place, and was trained to fight in battles. The high 
regard in which the Gauls held it remained among their 
descendants the ancient Irish ; for though I cannot find 
that dogs were employed in battle in Ireland, they were 
much valued and esteemed ; and they figure conspicuously 

* Br. Laws, in. 273. f See on all this, Br. Laws, m. 413. 

I Ibid., 193, to 199. 


in Irish literature. The best illustration of this is the 
very general custom of using cu as a name for men, so 
that large numbers of Irish personal and family names 
have cu or con as one of their components ; like Cucu- 
lainn, O'Conor, Macnamara (Mac Con-mar a), O'Connolly, 
Conway, Quin, Quinlan, &c* 

Wolves. — A common name for a wolf was cu-allaidh 
[coo-allee], i.e. ' wild-hound.' Another was mac-tire [mac- 
teera], which literally means ' son of the country,' in 
allusion to the wild places that were the haunts of these 
animals. Two other names fael and breach have long 
since fallen out of use, though they are commemorated 
in local names. Faelchu, which is formed from fael, and 
cu, a dog, is now a general name for a wolf. Cormac's 
Glossary (p. 87) gives glademain (pi.) as a collective name 
for wolves : which is derived from glaid or glaodh, a cry 
(pron. glay). 

In old times wolves were so numerous in the woods 
and fastnesses of Ireland as to constitute a formidable 
danger to the community : so that in Irish writings we 
meet with frequent notices of their ravages, and of the 
measures taken to guard against them. Sometimes when 
pressed with hunger they killed and ate human beings. 
But it may be said to have been the only really dangerous 
wild animal of the country ; and one of the old native 
writers, comparing Ireland to Paradise, and noticing its 
exemption from baneful reptiles, states that the wolf was 
its only noxious animal, f In later times, and probably 
in early ages as well, we know that these animals were 
hunted down by the great Irish wolf-dog : and they were 
also caught in traps. J 

* See, on all this, De Jubainville, La Civil, des Celtes, pp. 55-60 : Joyce, 
Irish Names of Places, 11. 156 : and a Series of Papers written in the Irish 
language by Mr. Thomas Flannery in the first volume of the Gaelic Journal 
(1882) on the word cu as used in Irish names. 

f Trip. Life, Introduction, xxx. note. 

t For wolf-traps in 1659, see Ulster Journ. Archaeol., ij. 281 


The war of extermination against wolves was not left 
to chance or to individual enterprise. We learn from the 
Senchus M6r and from the gloss upon it, that in various 
parts of the country there were organised efforts by the 
community to keep them down. Once a week a body of 
men made a regular raid on them under the direction of 
the chief : and it was a duty owed by every man to his 
chief to join these parties of attack in turn on the days 
appointed.* As the population and the extent of open 
cultivated land increased, wolves became less numerous 
and were held well in check ; but during the wars of the 
reign of Elizabeth, when the country was almost depopu- 
lated, they increased enormously and became bolder and 
fiercer, so that we often find notices of their ravages in the 
literature of those times, f 

Bears continued to exist in Scotland — according to 
Garmichael (11. 306) — so late as the sixteenth century : but 
they became extinct in Ireland at a much earlier period. 
The oldest list we have of the chief native wild animals is 
given incidentally in the treatise already mentioned (vol. 1. 
P- 345 > supra), written in the year 655, by Augustin, an 
Irish monk, then living in Carthage ; but bears are not 
among them. Next in point of antiquity comes the evi- 
dence of Bede, who states that the only noxious animals 
in his time in Ireland were the wolf and the fox. J In a 
Latin poem written early in the ninth century in praise of 
Ireland by Donatus, bishop of Fiesole, an Irishman, there 
is a more precise statement ; for one of the points of com- 
mendation is that it possessed no bears. Yet there were 
bears in Ireland at some very early time, while the country 
was inhabited by men ; for their bones are often found 
among the remains of human dwellings. Between 1840 
and 1846 the skulls of two bears were dug up in a cut- 
away bog§ : and quite recently the bones of numbers of 

* Br. Laws, I. 161, top. % Proc. Roy. Ir. Acad., IV. 417. 

f Iar C, 180. § Ibid., iv. 417. 



brown bears have been found in the caves of Kesh-Corran 
in Sligo. The Irish language retains a faint memory of 
these animals, inasmuch as it has a native name for a 
bear : — math, or more commonly in the compound form 
math-gamuin* : but beyond this the bear is totally lost to 
Irish history, so that it must have become extinct before 
our earliest legends began. 

Deer were plentiful in ancient Ireland, and they are 
noticed everywhere in the literature, both lay and ecclesi- 
astical. By far the most remarkable of the ancient deer 
of this country was the 
gigantic Irish elk, the 
bones of which are now 
often found buried deep 
in clay, sometimes with a 
thick layer of bog over it. 
It is well established that 
this stately creature lived 
in the country for some 
considerable time contem- 
poraneously with man : but 
it seems probable that it 
had disappeared before the 
time reached by our oldest 
writings : so that it is lost 

to history ; and those deer so often spoken of in Irish 
literature are not the great Irish elk, but animals like 
those of the present day.f The skeleton of the elk in 
the National Museum has antlers extending twelve feet 
from tip to tip : and as may be seen from the figure stands 
nearly twice the height of a man. 

The most common word for a deer is fiad or fiadh 
[feea], which originally meant wild ; but its' meaning has 

* See " Math " in Windisch. Worterb. Ir. Texte, i. : and " Math- 
gamuin " in O'Donovan's Supplement to O'Reilly. 

t See Kilk. Archaeol. Journ., 1849-51, p. 166 : and 1856-57, p. 155. 

Fig. 336. 

Skeleton of Irish Elk in the National Museum, 
Dublin. (From plate of the Royal Dublin Society.) 
Human skeleton put in for comparison. 


been gradually narrowed, and in Irish writings, as well as 
in the spoken language, it is almost universally applied 
to a deer. Seg is given in several glossaries, including 
Cormac's (p. 152), as a name for a wild deer; and both 
Cormac's (p. 68) and O'Clery's give erb as meaning a kind 
of deer ; but both -these words have been long obsolete. 
Os signifies a fawn. The celebrated Irish poet and 
warrior who lived in the third century of the Christian 
era, and whose name has been changed to Ossian by 
Macpherson, is called in Irish manuscripts Oisin [Osheen], 
which signifies a little fawn ; and the name is explained 
by a legend. 

The Hare. — It has been remarked above that the word 
fiadh [feea] was originally used in the sense of ' wild ' in 
general. The hare would appear to be the smallest animal 
to which the term was applied, if we may judge by the 
composition of its name gerr-fhiadh [gerree'j ; i.e. short 
or small fiadh, from gerr, short or deficient. In Cormac's 
Glossary (p. 133) is given patu [or ftata], another word 
for a hare, which still survives in the spoken language in 
the south of Ireland ; where the diminutive patachdn is 
used even by speakers of English to denote a leveret. 
The same glossary (p. 49) has cermna as still another name 
for a hare. Sometimes a hare was called mil-maighe 
[meel-mee], ' beast of the plain.' 

The Cat. — A cat is called by the same name with slight 
variations, in nearly all the languages of Europe : in Irish 
the common name is catt or cat ; but O'Davoren in his 
glossary gives eight Irish names for this animal. Cormac's 
Glossary derives catt from Latin cattus : but it is certain 
that the two words are merely cognate, i.e. both derived 
from an older source. Wild cats were in old times very 
plentiful : large wicked rough-looking creatures, very strong 
and active and very dangerous ; and the race is not yet 
quite extinct, for wild cats, nearly twice the size of our 
domestic animals, are still found in some solitary places. 


It was these animals that gave origin to the legend, very 
common in ancient Irish story, of a monstrous enchanted 
wild cat, dwelling in a cave, and a match for the bravest 
champion. One of these monsters, named Irusan, that 
had his dwelling in the cave of Knowth on the Boyne, 
once seized the poet Senchan in his mouth and ran off 
with him, till he was rescued by St. Ciaran.* Another 
tremendous cat named Luchthigem (' mouse-lord ' : luch, 
a mouse), lived in Derc-Ferna, now the cave of Dunmore 
near Kilkenny, till he was killed by a ban-gaisgidheach 
or female champion of Leinster.f Three monstrous cats 
dwelt in the cave of Croghan, from which Conall Cernach 
and Laegaire the Victorious had to fly for their lives ; but 
Cuculainn withstood them though he was not able to kill 
them .J Stories of demon cats have found their way down 
to modern Irish legend : see " Puss in Brogues," Irish 
Penny Journal, p. 346. 

Otters. — The otter has several names in Irish, the most 
usual in old writings being dobor-chu, ' water-hound ' (from 
dobor or dobur, an old word for water, common to Gaelic 
and Welsh). § It was also called madad- or madra-uisce, 
' water-dog.' O'Clery's Glossary explains dobor-chu by 
madra-uisce, which is now the general word for an otter ; 
though dobar-chu is still in use in some parts of the country. 
Cormac's Glossary (p. 40) gives the old word coinfodorne 
(pi. : the singular is confodomc) as meaning dobarchoin, 
i.e. otters ; and explains fodorne, the latter part of the 
word, as meaning fo-dobarnai, i.e. ' under dobur or water.' 
Con fodorne is given in O'Reilly's Dictionary in the more 
modern form confoirne. Another name for an otter, 
derived also from dobur, is dobran, a diminutive, which in 

* Tromdamh, pp. 81-85. 

t The Poem of Broccan the Pious, in praise of Leinster, in which this 
cat figures, will be found in LL, 43, b, 7 . it has been published, with trans- 
lation, by Mr. T. O. Russell in his Fior Chldirseach na h-Ereann, p. 118. 

I Fled Bricrenn, 73. 

§ Cormac's Glossary, 40, " Coinfodorne " ; and 53, " Dobur." 


Cormac's Glossary is given as the equivalent of dobarchu* 
In the " Story of the Eruption of Lough Neagh," Liban 
spends three hundred years as a salmon under the sea, 
accompanied by her lap-dog in the shape of a dobran or 
otter. f In the tract on Hy Many (p. 90, 16) the compound 
condobran is used for an otter. It is curious that in the 
Irish Glosses on Latin Declension edited by Stokes 
(No. 375), the old Irish writer explains the Latin fiber, a 
beaver, by dobran, an otter, which possibly may be owing to 
the circumstance that though he knew well what a dobran 
was, he had never seen a beaver, as there were none in 
Ireland, and thought that a fiber was the same animal 
as his own native otter (like what happened in case of 
Dancing, p. 445, and note, supra). 

Otters abounded in rivers and lakes, and were hunted, 
partly for sport and partly for their skins. In later times 
— and probably in the early ages — otter skins formed an 
important article of commerce, so that they were sometimes 
given as payment in kind for rent or tribute. We get an 
indication of the importance of these animals in the fact 
recorded in the Book of Lecan, that in the end of the 
fourteenth century the people of the district of Fidh 
Monach were entrusted with the special charge of the 
otters and of the fishing of O'Kelly, king of Hy Many 
(HyM, 93, top). 

Of the badger it will be enough to say here that it was 
called in Irish broc, and that the chase of the " heavy- 
sided low-bellied badger " was a favourite sport. 

5. Races. 

The Irish were passionately fond of racing, even more 
so than those of the present day. Everywhere, in all sorts 
of Irish literature, we read of races — kings, nobles, and 
common people attending them at every opportunity. 

* Cormac's Glossary, 53, " Dobur " ; the Irish in Three Ir. Gloss., p. 
15, " Dobur " J Kilk. Archaeol. Journ., 1870-1, p. 100, & 


The prominence of this sport at the aenachs or fairs is 
indicated in Cormac's Glossary (p. 127) by one of his 
derivations of the word oenach, a fair, which he says 
signifies ' contention of horses ' — as if racing was the main 
object of holding a fair. The popularity of the sport 
affected even the Law: for we find in the Senchus Mor 
a provision that young sons of kings and chiefs when 
in fosterage are to be supplied by the foster-fathers with 
horses in time of races.* 

A passage in the " Story of the Second Battle of 
Moytura " affords another indication of the universality of 
racing, where it is related that certain visitors arriving at a 
meeting were asked had they hounds (coin) and steeds 
(etch) for races : " for " — the story goes on — " at that time 
" when a body of men went to another assembly [in a 
" strange country or district] it was the custom to challenge 
" them to a friendly contest." Then " the hounds had a 
" coursing match (coin cocluiche, ' hound contest ') and the 
" horses ran a race " : after which the men themselves 
engaged in friendly sword-play, f But perhaps the best 
illustration of the passionate admiration of people high 
and low for this sport is that it is represented as one of 
the delights of the pagan heaven, as described to Bran by 
the fairy lady : — " The hosts run races — a delightful game — 
" along the plain of sports, the plain on which they hold 
" games ; a delight to the eye it is to look upon — it is a 
" glorious sweep of country. "% This shows, too, that one 
of the choice plains of heaven was specially set apart for 
sports. In the races held here there was moreover a 
variety, namely, races of curraghs on the water against 
chariots on the low-lying shore adjacent — " curragh con- 
tends against chariot "§ — from which we may conclude that 

* Br. Laws, 11. 155. t Rev - Celt., xn. 73. 

\ Kuno Meyer, in the Voyage of Bran, vol. I., p. 12, verse 23 ; and 
p. 4, verse 5. The " Plain of Sports " is Mag Mon in the original 
§ Kuno Meyer, in the Voyage of Bran, vol. I., p. 4, verse 5. 


this odd sort of race was also usual and popular in the 
mortal world above. The common name for a race of any 
kind was grafand or grafann : plural graifne : cur grafainn, 
' to run a race.' But this word was most generally confined 
in its application to a horse-race. Foot-racing does not 
appear to have been much practised. 

The Curragh of Kildare, or as it was anciently called, 
the " Curragh of the Liffey " (Cuirrech Life) was, as it is 
still, the most celebrated racecourse in all Ireland : and 
there are numerous notices of its sports in annals and tales. 
In the Bruden Da Derga it is stated that Conari, king of 
Ireland in the first century of the Christian era, went once 
with four chariots to the Cluichi or games of the Cuirrech 
Life*. The races were held here in connexion with the 
yearly fair, which was called Aenach Colmain or Aenach 
Lift, as being on the plain of the Liffey. It was the great 
fair-meeting of the southern half of Ireland, and especially 
of the kings of Leinster, when they resided at the palace of 
Dun-Ailinn (now Knockaulin), which was on the edge, and 
which, being on a flat detached hill, overlooked the Curragh 
and its multitudes. Though sports and pastimes of all 
kinds were carried on there, races constituted the special 
and most important feature : so that some of the annalists 
mention the Curragh under the name of " Curragh of the 
Races."* The games here were formally opened by the 
king, or one of the princes, of Leinster, and lasted for 
several days : and the great importance attached to them 
is indicated in the " Will of Cahirmore " in which that 
king bequeaths to his son Criffan the " leadership of 
" [i.e. the privileges of opening and patronising] the games 
" of the province of Leinster. "f 

In Cormac's Glossary cuirrech is used as a general 
word for a racecourse, and derived from the Latin cursus : 
and the scholiast who annotated St. Broccan's hymn on 

* See Three Fragm., 189, top line : see also FM, a.d. 825, note n ; 
and a.d. 040 and 954.. t Proc. Roy. Ir. Acad., ix. 348. 


St. Brigit, referring directly to the Curragh of Kildare 
gives the same derivation.* showing at what an early time 
the Curragh was recognised as a racecourse. But it seems 
more likely that the Irish word cuirrech is merely cognate 
with Lat. cursus, and not derived from it. In the hymn 
itself St. Brigit is designated as " the nun who used to run 
over the Curragh " ; for her convent was on the edge of it, 
and no doubt she often drove over the beautiful smooth 
sward in her chariot, f 

Numerous references to chariot-racing are met with in 
Irish literature. Cormac's Glossary in one place (p. 45) 
explains the word cuirrich by fich-carftait, the ' contest of 
a chariot ' [in a race]. During the first three centuries of 
the Christian era, chariot-racing was universal in Ireland ; 
and it was specially popular among the Red Branch 
Knights. Horse-racing was also very general, almost as 
much so indeed as racing with chariots. Maildune and 
his companions come to an island where they find gigantic 
people eagerly engaged in a horse-race. % The sport is 
alluded to in the ancient Notes on the Feilire of Aengus 
(p. 105, bot.), where a person striving to earn heaven by 
doing the will of God is compared to " a chariot that is 
driven under a king that bears off prizes." The Fena of 
Erin, as we have seen (vol. 1., p. 89, supra), did not use 
chariots, either in battle or in racing ; but they were 
devoted to horse-racing ; and in many passages referring 
to them, their chiefs are represented as indulging in this 
sport. § 

A racecourse (cete or cet) was sometimes in a king's 
faithche or exercise green : and the law in this case laid 
down rules as to how far the king was liable for accidents, 

* Todd Book of Hymns, 67, note /. 

f For much more on the Curragh, and on races, see Hennessy's Paper 
on " The Curragh of Kildare," Proc. Roy. Ir. Acad., vol. ix., p. 343. 

X Old Celtic Romances, p. 122. 

§ In the Story of Finn and the Phantoms, horse-races alone are carried 
on : no chariot-races : Rev. Celt. vn. 291. 

H I 


and when he was exempt : which rules applied to the 
owners of other racecourses. Thus if accidental collisions 
occurred from which injury resulted, he was exempt : but 
he was responsible if injury was caused by a chasm or a 
deep rut carelessly left unprotected, or not filled up. And 
the Law follows up this by a series of detailed regulations 
about the liabilities of the several parties in case of accidents 
on a course.* 

Coursing with greyhounds was another favourite amuse- 
ment. We have already seen (p. 463) how Irish visitors 
at a meeting in a distant land were challenged to a 
coursing match ; which came off with victory for the Irish 
hounds. The greyhounds (milchoin : singular milchu) 
mentioned in Cormac's Glossary as being always found 
at oenachs or fair-meetings, were for coursing contests, as 
part of the games carried on at the fair. A passage in the 
Crith Gabhlach, setting forth the distribution of a king's 
duties among the days of the week, assigns Wednesday 
for enjoying himself " witnessing greyhounds coursing. "f 

6. Chase and Capture of Wild Animals. 

Some wild animals were chased for sport, some for 
food, and some merely to extirpate them as being noxious : 
but it will be convenient to include all here in connexion 
with sports and pastimes. 

Our legendary annals relate that the first colonists to 
Ireland lived by hunting, fowling, and fishing : and though 
this record is legend, it presents a true picture of the mode 
of subsistence in primitive times, when the country was 
nearly all covered with forest and bog, and there was little 
open land for either tillage or pasture. But even after 
much land had been cleared and the people had begun to 
keep herds and to grow food-crops, they continued to hunt, 
fowl, and fish with the three objects stated above. Every- 

* Br. Laws, in. pp. 255-263. f Ibid., iv. 335, bottom. 


where in our literature we meet with notices of hunting, 
and of various other methods by which wild animals were 
taken. The hunters led the chase chiefly on foot, with 
different breeds of hunting dogs, according to the animals 
to be chased. Maelfothartaig [Mailfoharty], the son of 
Ronan king of Leinster, visits the king of Scotland, who 
" had hounds for boars, hounds for deer, and hounds for 
hares."* The principal kinds of game were deer, wild 
pigs, badgers, otters, and wolves ; and hares and foxes 
were hunted with beagles for pure amusement. Pig- 
hunting was a favourite sport. 

For the larger and more dangerous game, such as wild 
boars, wolves, and deer, the hunters employed wolfhounds 
jnd other breeds of large dogs ; and in the romantic litera- 
ture we have many a passage describing the dangers of the 
chase, and the courage, skill, and swiftness of hunters and 
hounds. The tales also reflect the immense delight those 
observant and nature-loving people took in the chase and 
all its joyous accompaniments. While Finn rested on the 
hill of Knockainy in the County Limerick, his companions 
hunted on the plain beneath ; " and it was sweet music to 
" Finn's ear the cry of the long-snouted dogs as they routed 
" the deer from their covers and the badgers from their 
" dens ; the pleasant emulating shouts of the youths ; the 
" whistling and signalling of the huntsmen ; and the 
" encouraging cheers of the heroes, as they spread them- 
" selves through the glens and woods, and over the broad 
" green plain of Cliach."f Cailte and his companions once 
" heard the musical concert (coicedul) of three packs of 
" hounds hunting round the head of Sliabh Lugda."J In 
another passage a man asks Cailte what pack is it that 
they hear : " That " — replies Cailte — " is the melodious 
chase by beagles after the swift and gentle hares. "§ Else- 
where Cailte describes the chase of " the heavy-sided, low- 

* Rev. Celt., XIII. 376. J Stokes, Acallamh, p. 206. 

f Old Celtic Romances, 226. § Ibid., p. 260. 


" bellied badgers : and behind the hunt they heard the 
" shout of the gillies, and the swiftest of the boys, and 
" the readiest of the warriors, and the men who were the 
" straightest spear-shots, and the strong attendants who 
" bore the heavy burdens." The same keen appreciation 
of the chase and its concomitants has descended to 
modern Irish sportsmen. The sweetest music in the 
world to Daniel O'ConnelTs ear was the cry of the Kerry 
beagles echoing among the woods and hills round Derry- 
nane ; and in the modern Anglo-Irish ballad of " Reynard 
the Fox " we are told how : — 

" Early next morning the woods they did resound 
With the echo of horns and the sweet cry of hounds."* 

Most of the details of the manner of trapping deer we 
learn from the Book of Aicill, in the third volume of the 
Brehon Laws, chiefly from p. 449, to p. 459. They were 
caught in a deep pit or pitfall (Irish cuithe or cuithech) 
with a trap, and a bir or spear fixed firmly in a wooden 
stock (Irish cep, pron. kep) in the bottom, point upwards ; 
the whole gin concealed by a brathlang or light covering 
of sods and brambles, f The spear either had a metallic 
head or was merely a stake of hard wood with a sharp 
point. Along with the spear there was in the pit a trap 
of some kind, the construction of which we do not know, 
called airndil or airnil, from which the spear was called 
bir airnil (or airndil), literally ' spear of a trap,' or a ' spear 
set for a trap. 'J That this is the correct sense appears 
further from Cormac's Glossary (p. 12), which derives 
airndel from air, noble, and indel (now inniV) a setting : — 
" air-indel, a noble setting [of a trap] it is." 

* This ballad and the air will be found in Joyce's Ancient Irish Music, 
p. 50. 

f Cuithe, Br. Laws, in. 452, I2: Cuithech, 272, 6 and Trip. Life, 186, 2S> 
where brathlang, ' a pit-cover,' is also used. 

% See " Bir " in O'Dovovan's Supplement to O'Reilly : and Br. Laws, 
in. 272, 9 . and 452, 3. 


The trap — which was also often called cuichi* — appears 
to have been set independently of the spear with its stock — 
though they were beside each other in the pit — so that 
either spear or trap might be removed without stirring the 
other : which appears from the expression in the Book of 
Aicill : — " If the deer [has escaped after falling in and] 
has carried off the spear-stock or the trap out of its place." 
Spear and trap were both set in the same pit, so that the 
deer that fell in was pretty sure to be secured by either or 

A passage in the Rennes Dinnsenchus giving the 
supposed origin of the name of the ancient district of 
Magh Cobha, in the present county Down, affords us some 
idea of how these traps acted : — " Coba [cova], the 
" cuchaire or trapper of Heremon [first Milesian king of 
" Ireland] son of Milesius : it is he that first prepared a 
" trap [airrchis] and a pit-fall [cuithech] in Erin : and he 
" himself put his foot in it to try if it was in trim, where- 
" upon his shinbone and his two forearms were fractured in 
" it : and his drinking-cup, after being emptied, fell down, 
" so that he died thereof [i.e. of the wound and thirst] : 
" whence is derived Mag Coba, Cova's plain." f This passage 
introduces a new name for a trap — airrchis : a word which 
is in a still older authority, Cormac's Glossary (p. 2, 
" Airches "), where it is explained " a trap or enclosure," 
and derived " ab arceo, because of its holding whatsoever 
is put in it." An ancient MS. quoted by O'Donovan, 
under this passage in the Glossary, states that an airches 
is a trap for catching wild hogs : but the passages just 
quoted from Cormac's Glossary and the Dinnsenchus show 
that the word was applied to a trap in general. The 
quotation by O'Donovan is interesting, however, as it 
shows that wild hogs were caught in traps as well as 
deer and other animals. We have already seen that 
wolves were caught in specially-constructed traps. 

* Br. Laws, in. 456, I4 and I7 . f Stokes, in Rev. Celt., xvi. 44. 


There were persons skilled in setting traps, following 
the occupation as a sort of trade, who either worked on 
their own account, or were employed by others : such a 
person was called a cuchaire or cuthchaire [cuh'ara], i.e. a 
trap-maker or 'trapper' (from cuichi). A trapper was, 
sometimes at least, one of the officials in the employment 
of a king ; as in the case of Coba. Some trappers or 
hunters and their pitfalls are designated " lawful " and 
others " unlawful."* Both are taken into consideration 
in the Book of Aicill, which, in setting forth fines for 
damages, distinguishes carefully between them, but gives 
no explanation of the distinction : because, as in hundreds 
of other cases, all understood the terms at the time, so 
that no definition was needed. Probably some sort of 
license or authoritative legal sanction was prescribed, the 
possession of which constituted " lawfulness," and its 
absence " unlawfulness." 

A deer-trap was obviously very dangerous to both 
people and cattle : and the Book of Aicill dwells minutely 
on the precautions that should be adopted and on the 
question of responsibility and damages in case of injuries 
from trap accidents. A deer-trap was generally set in a 
wild place — mountain, bog, or wood : but sometimes also 
in the faithche or green of the owner's residence, or 
between the green and the adjacent wild place. The 
law laid down that in all cases verbal notice should be 
sent over the nine holdings (nae n-orba) nearest to the 
place, f Sometimes when deer were observed hovering 
near, they were driven, so far as could be done, towards 
the trap ; and if they did any damage while being driven, 
either to the people or to domestic animals, the law pro- 
vided compensation.! Adamnan (p. 154) gives an account 
of the setting of a sharpened stake in a place frequented 
by wild animals, on which a poor man caught many 

* Br. Laws, III. 457. f On these points : — Br. Laws, m. 453 ; 

2 73i us 2 7 2 » 9. J Ibid., 457, bottom. 


deer to feed his family. As confirming the preceding 
records, remains of deer-traps with their sharpened stakes 
are now often found buried deep in bogs in various parts 
of Ireland.* 

Wooden traps of another kind, quite different from 
deer-traps, smaller and having no spear, all of much the 
same pattern, with doors or valves, springs, and triggers, 
are often found in bogs, of which the illustration (fig. 337) 
will give a good idea. There has been considerable 

Fig. 337- 

Three different views (top, side, bottom) of an otter-trap : found in a bog. 

(From Wood-Martin's Pagan Ireland, p. 407.) 

divergence of opinion regarding their exact use ; but it 
is now generally considered that they were otter-traps : 
the animal, while attempting to force its way through, 
being caught and held by the edge of the door or valve. 

There were traps and nets of several kinds to catch 
birds. The word sds [sauce], which means an engine or 
gin of any kind, is also applied to a bird -trap. A 
basket- shaped bird -crib, such as is used by boys at the 

*Kilk. Archaeol. Journ., 1879-82, p. 500, bottom. Ms. G. H. Kinahan 
has mentioned to me half a dozen places where he has seen the remains of 
old deertraps. See also his Geology of Ireland, p. 277. 


present day, was called cliabhdn [cleevaun], which is also 
the word for a child's cradle : a diminutive of cliabh 
[cleeve], a basket. A person very much frightened, or 
cowardly, or unreliable in time of sudden danger, was 
often compared to a bird on which a sds or a cleevaun 
has closed, and also to a salmon caught in a weir-trap. 
Cormac's Glossary (p. 152) explains the word sin [shain] 
as ' a net in which birds are caught,' and illustrates it by 
the derivative sin-bretha, ' bird-net laws ' : showing that 
the practice of catching birds by nets was general ; and 
that it was considered of such importance that special laws 
were laid down to regulate it. Birds were also caught, as 
they are still in the Orkneys and Hebrides, by men let down 
in baskets with ropes over the cliffs round the coasts ; 
as appears by several entries in the Brehon law-tract 
called " Heptads."* O'Flaherty, in his " Iar Connaught " 
(p. 67), states that, in his time, birds were caught in this 
manner at night by persons who brought down candles, 
the light of which fascinated the birds so that they were 
easily taken. I have seen birds caught by night in bushes 
by boys who carried lanterns : but of this practice I have 
found no mention in old Irish literature. 

Fish as an important article of food has been already 
spoken of (p. 133, supra). The general Irish word for a 
fish is iasc [eesk], cognate with Latin ftiscis and English 
fish : and a fisherman was an iascaire [eeskere : 3-syll.]. 
The people fished with the net and with hook and line, 
both in the sea and in lakes and rivers : as the records 
everywhere show. The slatt or fishing-rod was 10 or 12 
feet long : the line was called ruaim [roo-im] or ruaimnech : 
and the hook duban [dooan]. Net-fishing came under the 
cognisance of the law ; it is mentioned in the Senchus Mor ; 
and it appears from the gloss that a fishing-net was called 
cochull and lin [leen], both words in use still. The net was 
sometimes the common property of the fine or family 

* Br. Laws, v. 237, 20; 239, J9; 301, 20 - 


group of relations, each individual family having a right to 
use it in turn, or a claim to a share of the fish caught. 
Both salmon and eels were often caught with trident 
spears, or with spears of more than three prongs : and 
sometimes people followed the primitive plan of trans- 
fixing large fish with a single-point spear. Eels are still 
caught with forked-spears : and until lately salmon were 
taken in a similar way, the trident spears having handles 
about 5 feet long and a cord attached at the end. The 
spear was flung from the bank with aim that seldom 
missed : and spear and salmon were drawn back by the 
cord (see vol. L, p. 112, for a fishing-spear). 

Salmon-fishing was the most important of the fishing 
industries, and it is oftenest mentioned in the old writings : 
it is constantly noticed by Adamnan. A salmon is desig- 
nated by several Irish names : — bratan, ae, eo, tonnem, orcc, 
eicne [aikne], and linne, besides some others : but the first 
under the modern form braddn, is now the general name. 
It would seem that in old times bratan meant a young 
salmon, and eo a full-grown one : for we find this line in 
an ancient poem in the Book of Leinster : — " It is from 
the bratan that the eo comes : it is from the youth that 
the king comes."* 

Fishing-weirs on rivers were very common. The weir 
(called in Irish corad and sod) sometimes belonged to an 
individual, and sometimes — like the net — it was the common 
property of the fine. A man who had land adjoining the 
stream had the right to construct a weir for his own use : 
but according to law, he could not dam the stream more 
than one-third across, so that the fish might have freedom 
to pass up or down to the weirs belonging to others. If it 
was found that his dam went more than this, he had to 
give up two-thirds of the fish he caught to the wronged 
man, whether living above or below, f 

* LL, 148, a, l6 . " Contents," 34, a, 26 . 

t On all this see Br. Laws, 1. 131, 6i 7 . 205, 207 iv 211, bottom; 
213. .3. 

474 social And domestic life [part lit 

J. Camdn or Hurling, and other athletic games. 

Hurling or goaling has been a favourite game among 
the Irish from the earliest ages ; and those who remember 
the eagerness with which it was practised in many parts of 
Ireland sixty years ago, can well attest that it had not 
declined in popularity. Down to a recent period it was 
carried on with great spirit and vigour in the Phoenix 
Park, Dublin, where the men of Meath contended every 
year against the men of Kildare. It still continues, though 
less generally than formerly, to be a favourite pastime, and 
there is lately a movement to revive it. 

Our literature gives us many glimpses of the manner 
of carrying on this game in old times. It was played with 
a ball (liathrdid : pronounced leeroad) about four inches 
in diameter, made of some light elastic material, such as 
woollen yarn wound round and round, and covered with 
leather. Each player had a wooden hurley to strike the 
ball, generally of ash, about three feet long, carefully 
shaped and smoothed, with the lower end flat and curved. 
This was called camdn [commaun], a diminutive from cam, 
curved : but in old writings we find another name, lorg 
(i.e. ' staff '), also used. The game was called immdn, or in 
modern spelling iomdn [immaun], meaning ' driving ' or 
' urging ' : but now commonly camdn, from the camdn or 
hurley. In a regular match the players on each side were 
equal in number. It was played on a level grassy field, at 
each end of which was a narrow gap (bema) or goal, 
formed by two poles or bushes, or it might be a gap in 
the fence. The general name for the winning goal was 
bdire [bawre]. The play was commenced by throwing up 
the ball in the middle of the field : the players struck at it 
with their hurleys, the two parties in opposite directions 
towards the gaps ; and the game, or part of it, was ended 
when one party succeeded in driving it through their 
opponents' gap. It was usual for each party to station 


one of their most skilful men beside their own gap to 
intercept the ball in case it should be sent flying direct 
towards it : this man was said to stand cul [cool], or cul- 
bdire, ' rear-guard ' : cul meaning back. The preceding 
description shows generally how the game was played 
down to a period within my own memory ; and so far as 
can be judged from the old literature, it was much the 
same a thousand years ago. In old times the field on 
which the game was played was commonly the faithche 
[faha] or green of the dwelling (p. 61, supra) : but some- 
times a large level space was railed off for games of all 
kinds, which was called cluichi mag, ' game-plain,' or 
mag-mon, ' plain of sports.' 

The law required that the sons of kings and chiefs 
in fosterage should have their camans mounted or orna- 
mented — or perhaps ringed — with bronze or brass.* These 
bronze-mounted hurleys were valuable, as we may infer 
from a record in the annals, that on one occasion Feradach, 
king of Ossory towards the end of the sixth century,, 
collected into one place all his valuable and precious 
things — his ingots of silver, ornamented cups, &c. — among 
which are enumerated his camdin creduma, ' hurleys of 
bronze ' — that is to say, mounted with bronze, f 

Provision was made in the Brehon Law for compen- 
sation in case of accident by either ball or hurley ; and 
in the statements of these rules in the Book of Aicill we 
come across other features of the game. A player is 
mentioned as " striking the ball (liathroid) with his hurley 
" (lorg), from the hurling hole (poll na h-imdna) to the 
" place of the grifid — or from the place of the grifid to the 
" place of the division (comrann) — or from the hole (poll) 
" where it [the ball, then] is until it reaches the place where 
" it usually lies. "J Some of these terms and details are 
unintelligible to me. 

* Br. Laws, n. 147. f Three Fragments, p. 9. 

% Br Laws, in. 555. 


Hurling is often mentioned in the tales. Cuculainn, 
when a boy, going to Emain to visit his uncle King 
Concobar, took with him his playthings to shorten the 
road, among them a cammdn creduma ocus a liathrdit 
n-argdide, " his hurley of bronze and his ball of silver."* 
In the tale of the Destruction of Dinnree (see p. 95, supra), 
it is related that Moen Ollam, afterwards Labrad Loingsech 
(king of Ireland before the Christian era), when a boy, was 
dumb. But one day as he was playing immdn in the 
playgreen, he got a blow of a hurley on the shin, which 
gave him such a paroxysm of agony that he shouted out 
and spoke for the first time ; and ever after that he retained 
his speech, f It may be inferred from the careful provisions 
laid down in the law that the game sometimes resulted in 
injuries, generally by accident, but occasionally by design, 
in rare cases ending even in death : and in this respect, too, 
it resembled the game of our own day. But the mischief 
was seldom serious. 

Various other athletic exercises were practised, some 
of them like those we see at the present day. We find 
foot-racing and leaping mentioned, but much prominence 
is not given to them in the tales. There were contentions 
in wrestling and swimming. The boys of the Red Branch 
had a rough sort of game, in which each set strove to tear 
off the outer garments from their opponents ; and they 
had another, something analogous to goaling, where they 
endeavoured to put a ball through a small hole, which 
those opposed to them tried to prevent. % There was also 
the " loop and ball game " (cluiche luibe ocus liathroidi,) 
which was played on the green ; but nothing is said in the 
tales as to the manner of playing it.§ What was called 
the Roth-chless, or ' wheel-feat,' consisted in throwing a 
heavy circular disc or quoit upwards beside the wall of a 

* LL, 62, a, 4S: Miss Hull, Saga, p. 136. 

f Stokes's edition, Zeitschr. Celt. Phil., in. p. 10 

X Miss Hull, Saga, 138, 139 ; LL, 63, a, bottom. 

§ See Kuno Meyer, Ventry, 29, line 530 ; and p. 82, note 529. 


large house. At this the three great Red Branch heroes 
once contended inside the lofty hall. Laegaire the Victor- 
ious took the first trial and sent the disc half way up the 
side wall. Conall Cernach next : and he threw it to the 
ridge-pole. Lastly, Cuculainn caught it in mid-air as it 
came down after Conall's throw, and with a mighty effort 
hurled it right through the roof, amid the frantic shouts of 
the spectators.* Throwing a ball or quoit to a distance, 
and "putting" a heavy stone forward from hand and 
shoulder — both of which are noticed in the tales — were in 
no wise different from the corresponding strength trials of 
the present day. 

8. Chess. 

In ancient Ireland chess-playing was a favourite pas- 
time among the higher classes. Everywhere in the 
Romantic Tales we read of kings and chiefs amusing 
themselves with chess, and to be a good player was 
considered a necessary accomplishment of every man. 
of high position. At banquets and all other festive 
gatherings this was sure to be one of the leading features 
of the entertainment. In every chief's house there was 
accordingly at least one set of chess appliances for the 
use of the family and guests : and chess-boards were 
sometimes given as part of the tribute to kings, f Chess 
furniture was indeed considered in a manner a necessity, 
so much so that in this respect it is classified in the 
Brehon Law with food.| 

As to the general form and construction of the chess- 
board there can be no doubt, for Cormac's Glossary (p. 75) 
describes it with much exactness. This old authority states 
first, in regard to the game, that the play demands ciall 
and fdth [keeal, faw], i.e. attention and judgment : and it 
goes on to say that the fidchell or chess-board was divided 
into black and white compartments by straight lines : that 

* Fled Brier., 83. J Book of Rights, 39. J Br Law, 1. 143, xa , 


is to say, into black and white squares. The game was 
called fidchell or fidchellecht [fihel, fihelleght] : and fidchell 
was used to designate the chess-board. But this was also 
called cldr-fidchilli, cldr being the general name for a board 
or table. The chessmen were called fir-fidchilli, i.e. ' men 
of chess,' or collectively foirenn, which is the Irish word 
for a party or body of men in general. The whole set of 
furniture was called fidchellecht * or fidchell. 

The men, when not in use, were kept in a fer-bolg or 
' man-bag,' which was sometimes of brass or bronze wire 
woven. The chiefs took great delight in ornamenting 
their chessboards and men richly and elaborately with 
the precious metals and gems. We read in the " Story 
of the Battle of Mucrime," that when the Irish chief 
Mac Con was an exile in disguise at the court of the king 
of Scotland, the king's chessmen were of gold and silver : 
meaning ornamented with these metals, f The following 
quotation from a much older authority — the " Courtship 
of Etain " in the Book of the Dun Cow — is very instructive 
and very much to the point. Midir the fairy king of 
Bri-leith, comes on a visit to King Ochy : — " What brought 
thee hither ? " said Ochy. " To play chess with thee," 
answered Midir. " Art thou good at chess ? " said Ochy. 
" Let us try it," said Midir. " The queen is asleep," said 
Ochy, " and the house in which are the chessboard and ' 
men belongs to her." " Here I have as good a set of 
chess," said Midir. That was true indeed; for it was a 
board of silver and pure gold ; and every angle was illumi- 
nated with precious stones ; and the man-bag was of 
woven brass wire. "J In the Will of Cahirmore, king 
of Ireland in the second century, we are told that he 
bequeathed his chessboard and chessmen to his son Olioll 
Ceadach§ — an indication of their great value. 

* Book of Rights, 201, I7 . 

f Silva Gad., 351 : see also Keating, 290, top ; and FM, a.d. 9. 

% See O'Donovan, in Book of Rights, Pref ., lxi § Book of Rights, 201. 



The men were distinguished half and half, in some 
obvious way, to catch the eyes of the two players. Some- 
times they were black and white. The foirenn or party 
of chessmen of Crimthan Nia Nair, king of Ireland about 
the first century of the Christian era, are thus described : — 
" One-half of its foirenn was yellow gold, and the other 
half was fvndruine " (white 
bronze).* Many ancient 
chessmen have been found 
in bogs, in Lewis and other 
parts of Scotland : but so 
far as I know we have only 
a single specimen belong- 
ing to Ireland, which was 
found about 1817 in a bog 
in Meath, and which is now 
in the National Museum, 
Dublin. It is figured here, 
full size. We frequently 
read in the tales that a 
hero, while playing chess, 
becoming infuriated by 
some sudden attack or 
insulting speech, flings his 
chessman at the enemy 
and kills or disfigures him. 
When we remember that chessmen were sometimes made 
partly of metal and were two and a half inches long, 
we may well believe this. 

The game must, sometimes at least, have been a long 
one. When St. Adamnan came to confer with King 
Finachta, he found him engaged in a game of chess : but 
when his arrival was announced, the king, being aware 
that he had come on an unpleasant mission, refused to see 
him till his game was finished : whereupon Adamnan said 

* Rev. Celt., xx. 283. 

Fig. 338. 
Bone Chessman, King, full size ; found In a bog 
in Meath about 1817. (Drawn by Petrie : Book of 
Rights, page lxii.) 


he would wait, and that he would chant fifty psalms during 
the interval, in which fifty there was one psalm that would 
deprive the king's family of the kingdom for ever. The 
king finished his game however ; and played a second, 
during which fifty other psalms were chanted, one of 
which doomed him to shortness of life. But when he 
was threatened with deprivation of heaven by one of the 
third fifty, he yielded, and went to Adamnan.* 

That the Irish retained the tradition of the origin of 
chess as a mimic battle appears from the name given to 
the chessmen in the story of the Sick Bed of Cuculainn 
(p. 99) in the Book of the Dun Cow : — namely fianfidchella, 
i.e. as translated by O'Curry, ' chess-warriors ' ; flan, a 
champion or warrior : from which we may infer that the 
men represented soldiers. 

Another game called brannuighecht , or ' ftraww-playing,' 
as O'Donovan renders it, is often mentioned in connexion 
with chess ; and it was played with a brannabh, possibly 
something in the nature of a backgammon board. A party 
of Dedannans were on one occasion being entertained ; and 
a fidchell or set of chess furniture was provided for every 
six of them, and a brannabh for every five,t showing that 
chess-playing and 6nww-playing were different, and were 
played with different sets of appliances. Among the 
treasures of the old King Feradach are enumerated his 
brandaibh and his fithchella.% The Brehon Law prescribes 
fithchellacht and brannuidhecht (as two different things) 
with several other accomplishments, to be taught to the 
sons of chiefs when in fosterage.§ Notwithstanding that 
chess-playing and &ra?m-playing are so clearly distin- 
guished in the above and many other passages, modern 
writers very generally confound them : taking brannuigh- 
echt to be only another name for fitchellecht or chess- 
playing, which it is not. 

* Silva Gad., 422. J Three Fragments, 8, „. 

f Ibid., 250 : Ir. text, 220, so . § Br Laws, n. 155, 9; 157, bottom. 


There is still another game called buanbaig or buanfach, 
mentioned in connexion with chess and &ran«-playing, as 
played by kings and chiefs. When Lugaid mac Con and 
his companions were fugitives in Scotland, they were ad- 
mired for their accomplishments, among them being their 
skilful playing of chess, and brandabh, and buanbaig.* 
Nothing has been discovered to show the exact nature of 
those two last games. 

I have headed this short section with the name 
" Chess," and have all through translated fitchell by ' chess,' 
in accordance with the usage of O'Donovan, O'Curry, and 
Petrie. Dr. Stokes, on the other hand, uniformly renders 
it " draughts." But, so far as I am aware, there is no 
internal evidence in Irish literature sufficient to determine 
with certainty whether the game of fitchell was chess or 
draughts : for the descriptions would apply equally to 

9. Jesters, Jugglers, and Gleemen. 

From the most remote times in Ireland, kings kept 
fools, jesters, and jugglers in their courts, for the amuse- 
ment of their household and guests, like kings of England 
and other countries in much later times. In the tales we 
constantly read of such persons and their sayings and doings. 
They were often kept in small companies. Prince Cum- 
masgach, son of Aed mac Ainmirech (king of Ireland, a.d. 
572 to 598), had one chief jester named Glasdam, under 
whom were eight others ; and the whole nine constantly 
accompanied the prince, forming part of his retinue both 
at home and on his journeys. f 

The most common name for a jester or fool was druth 
(pron. droo : to be carefully distinguished from drui, a 
druid), which is explained in Cormac's Glossary (p. 59), 
by onmit, a fool, and in another place in the same Glossary 

* O'Grady, Silva Gad., 351, top: Ir. text, 312, 30 _ 3I: Revue Celt. 
xiii. 443 j Rev. Celt., in. 59 : Silva Gad., 409. 

I I 


(130) by midlach, an imbecile. Another word was mer, 
which generally means a mad person, but which, like onmit 
and midlach, appears to have been often used to denote 
simply an idiot or mental imbecile ; while druth was used 
as a more general term to include all sorts of fools, 
buffoons, jugglers, jesters, &c. Mer and druth are some- 
times used synonymously, as in the Book of Aicill.* The 
Law (in. 157, 23) fixed seven years as the age when it 
was to be decided whether a young person was or was 
not a druth, which here must be understood to have the 
restricted meaning of an idiot. In Cormac's Glossary 
(p. 81) jaindelach is given as still another name for an 

In the Senchus M6r the customary functions of a druth 
are indicated : — He is said to have the power of amusing, 
and that music was one of the means he employed to do 
so : for he was a minstrel (airfidig) ; and it also states that 
he had land — at least sometimes — no doubt in reward of 
his services : while a woman who was a mer or mad person 
was not a minstrel, and had no land.j From a legend 
given in Duald Mac Firbis's Irish Annals we learn that a 
druth was trained to give a particular kind of shout, which 
was easily distinguished from the shout of all other people. 
At the Battle of Allen in Kildare, fought a.d. 722, the 
rig-druth or royal clown O'Maileeny was taken prisoner 
and beheaded. But just before his execution he was asked 
to give a " druth' s shout," which he did : and it was so 
loud and melodious that its soft echoes were heard in the 
air for three days and three nights after his death. J And 
ever afterwards all druths learned to give " a clown's 
shout," like that of poor O'Maileeny. 

These half-witted dependents were always greatly 
attached to their masters, whose lives, as occasions arose, 
they sometimes saved by the sacrifice of their own, In 

* Br. Laws, in. 199, bottom ; 200, a . f Ibid., i. 137, bottom ; 139, top. 
t Three Fragments, 43 : and Rev. Celt., xxiv. 55. 


the year 598 a.d., Prince Cummascach — as the Story of 
the Boroma relates — went through Leinster on his " Free 
circuit of youth." His conduct was on the occasion so 
intolerably licentious that Branduff, king of Leinster, and 
his people, at last rose up in a fury and set fire to the 
house in which the prince and his retinue were feasting, 
standing all around in close array to prevent escape. 
Glasdam, the prince's jester, was one of the doomed 
company : but as he had been hospitably entertained a 
few days before by Branduff, he now cried out : — " Lo, I 
" have eaten thy meat : let not this deed of shame be now 
" wrought on me ! "* And Branduff answered : — " By no 
" means shall this be done : climb up to the ridge-pole and 
" leap out over the flames to the ground : we will let thee 
" pass, and thus shalt thou escape ! " But Glasdam refused 
to be saved without his master : and tearing off his fool's 
mantle and cap 4 while the flames were closing in, he said 
to Cummascach : — " Take these and escape in guise of 
me!" The prince put them on, and, leaping out, was 
allowed to pass, so that he escaped for the time, while the 
poor fool remained behind and was burned to death with 
the rest.f 

Fools when acting as professional clowns were dressed 
fantastically ; and they amused the people something in 
the same way as the court fools and buffoons of later 
times — by broad impudent remarks, jests, half witty, half 
absurd, and odd gestures and grimaces. Concobar's royal 
fool Roimid, who lived in Emania, is thus described in the 
Story of Mesca Ulad as he stood amusing a delighted 
crowd who surrounded him : — He had a black, pointed, 
thick head of hair ; his face [painted] a bluish-black, like 
an Ethiopian ; his eyes were large, wide open, and seemed 

* For the duty of host towards guest, see p. 168, supra. 

f O'Grady, Silva Gad., 410 : Stokes's version in Rev. Celt., xni. 33. 
Two other equally striking instances may be seen in the story of the 
jester of Fiachna, king of Ulster (Silva Gad., 427), and that of Mac Con's 
druth at the Battle of Cenn Febrat (Silva Gad., 349, 350). 


all white [on account of the blackened face] ; he wore a 
ribbed bratt or mantle all in folds, fastened with a brass 
clasp at his breast : at his side hung a melodious little bell 
(cluicin cedlbind) which he often struck with a bronze wand 
to procure attention, making such a sweet tinkle that it 
gave pleasure and delight to the arch-king and to the 
whole host. He was a laughable and amusing wight — adds, 
the old account — and there was no care, fatigue, or sorrow, 
however great, that a man would not forget for the time, 
while looking at this droll fellow and listening to his 
pleasantries.* King Conari's three jesters were such 
surpassingly funny fellows that, as we are told in the 
Bruden Da Derga (where they are called cuitbi, i.e. jeerers, 
gibers, mockers), no man could refrain from laughing at 
them, even though the dead body of his father or mother 
lay stretched out before him.f 

Professional gleemen travelled from place to place 
earning a livelihood by amusing the people like travelling 
showmen of the present day. To these the word druth is 
sometimes applied, though their more usual name was 
crossan, a word which glosses scurra in Stokes's " Glosses 
on Latin Declension," No. 14. When they had given an 
exhibition it was considered disgraceful to refuse them a 
contribution : and even St. Patrick, as we are told in the 
Tripartite Life (205), thought it necessary to comply with 
this native custom. When he was in Hy Fidgente he was 
entertained by a chief named Lomman, living beside 
Mullach-Cae, i.e. the mountain now called Knockea, and 
having his house on a spot lying, as the old record correctly 
states, to the south of Carn-Feradaig, which is now called 
Seefin Mountain, rising over the village of Glenosheen in 
Limerick. One day a number of gleemen came up and 
asked for some food. Patrick was troubled at this, for he 
had nothing to give them : but just at the moment a boy 

* Mesca Ulad, 35. f O'Curry, Man. & Cust., 11. 150 : Da Derga, p. 311. 


passed near with a cooked ram on his back, which the 
saint asked from him " to save his honour " — i.e. that he 
might not lie under the reproach of refusing the gleemen. 
The boy gave it gladly, and Patrick handed it over to 
them ; but before they had time to swallow it they were 
themselves swallowed up by the earth for their impudence. 
In the first part of this narrative the gleemen are called 
aes ceirdd, i.e. ' men of art ' : at the end they are designated 
by druth : and in another part of it they are called in 
Latin prcecones, i.e. ' public criers,' from their custom of 
bawling out their jokes. These " men of art " were evi- 
dently of the same class as those called crossans, who are 
constantly met with in the tales. 

A travelling band of crossans had a fuirseoir or obldire 
[furshore, oblaire], i.e. the chief buffoon and juggler of 
the company. When the sons of O'Corra were about to 
embark on their voyage of pilgrimage, a band of crossans 
came up and inquired among themselves who they were. 
" I know them well," said the fuirseoir ; " they are the sons 
" of O'Corra, the robbers and murderers, going on their 
" pilgrimage : and indeed they do not stand more in need 
"of it than we do." "I fancy " — said the leader of the 
band, scoffingly — " it is a long day till you go on a 
pilgrimage." " Never say so," replied the fuirseoir, " for 
I will certainly go with them " : and so he did. From the 
same narrative it is clear that they wore dresses of some 
special sort, which belonged, not to the individuals but 
to the company : and that if any member left the party 
he had to leave his dress behind him, as the fuirseoir had 
to do on the present occasion, for they stripped him and 
sent him off to the pilgrims stark naked.* We may form 
some idea of one part of the dress from the statement in 
Mac Conglinne (42) , that when he disguised himself as a 
fuirseoir, " he put on a short cloak and short garments : 

* Joyce, Old Celtic Romances, p. 409. 


" each upper garment being shorter with him, and each 
" lower one being longer. In this wise he began juggling." 
See also the description of Roimid's mantle, p. 484, supra. 

Cormac's Glossary (p. 141, explains the word reim 
by fuirseoir, i.e. buffoon, and does it in such a way as 
to give us a peep at one of the buffoon's tricks : — 
" Reim, the name for a fuirseoir, on account of the 
" distortions of face he makes towards people " [to make 
them laugh]. In the Crith Gabhlach the word is ex- 
plained similarly : — " Reimm, i.e. a fuirseoir or a druth, 
" a man who brings distortion (remmad) upon his body 
" and his face, is not entitled to dire fine (Br. Laws, vol. 1., 
" p. 208), because he goes out of his own shape before 
" hosts and crowds."* 

There was a druth of a different kind from all those 
noticed above, a hand-juggler — a person who performed 
sleight-of-hand tricks. Such a person was called a cless- 
amnach [classownagh] , i.e. a ' trick-performer,' from cless, 
a trick. In the Bruden Da Derga King Conari's cless- 
amnach and his trick of throwing up balls and other small 
articles, catching them one by one as they came down, 
and throwing them up again, are well described : — " The 
" blemish of baldness was on him, and the hair on the rest 
" of his head was whiter than canach-slebhe — canavan or 
" cotton grass ; he had clasps (unasca) of gold in his ears 
" (P- 2 59») supra) ; and he wore a speckled white cloak. He 
" had nine [short] swords, nine [small] silvern shields, and 
" nine balls of gold. [Taking up a certain number of 
" them] he flung them up one by one, and not one of them 
" does he let fall to the ground, and there is but one of 
" them at any one time in his hand. Like the buzzing- 
" whirl of bees on a beautiful day was their motion in 
" passing one another, "f This man seems to have been 

* Br. Laws, iv. 355, top : see also to the same effect, v. 109, h from 

t Mao. & Cust, 11. 144, 145 ; Da Derga, p. 286. 


the chief juggler ; for three others are afterwards described, 
having a number of small balls and of small darts, for 
performing similar tricks with.* 

The Brehon Law lays down fines for injuries to spec- 
tators by the careless performance of juggling tricks : and 
in assessing damages it distinguishes between dangerous 
articles such as knives, and non-dangerous ones such as 
balls. The fine for injury was heavier if inflicted by 
dangerous than if by non-dangerous articles — even though 
the injury might be the same : which of course was 
intended to ensure that due care should be taken during 
the performances, f 

It is worthy of remark that the gleemen continued 
till the time of Elizabeth. Spenser (View, 125) describes 
them : — " To these may be added another sort of like 
" loose fellowes, which doe passe up and downe amongst 
" gentlemen, by the name of Jesters " : and he goes on to 
say that they were great collectors and carriers of news. 

People of all the above classse, crossans, druths, jesters, 
tumblers, distortionists, and soforth, were looked upon 
as dishonoured and disreputable. This appears from the 
passage quoted on last page, where we see they were 
denied certain civil rights enjoyed by ordinary citizens ; 
and still more pointedly from an ordinance of the Senchus 
Mor, which, classifying banquets into godly, human, and 
demon banquets, defines demon banquets as those given 
to evil people, such as satirists, jesters, buffoons, mounte- 
banks, outlaws, heathens, harlots, and bad people in 
general. I And many other passages in Irish literature 
might be quoted to the same effect. 

* Man. & Cust., n. 147. f Br. Laws, in. 285. 

J Br. Laws, in. 25 

Ornament on top of DcreiiUh Round Tower. (From Petrie's Round Towers, 400.) 



Section i. Salutation. 

\OME of the modes of salutation and of showing 
respect practised by the ancient Irish indi- 
cate much gentleness and refinement of 
feeling, while others would now be considered 
degrading. When a distinguished visitor 
arrived it was usual to stand up as a mark 
of respect (ureirge, coimeirge, rising, or standing 
up). King Laegaire, sitting with his courtiers, in his 
council-hall at Tara to receive St. Patrick, ordered that no 
one should rise when he entered : but Dubthach Maccu 
Lugair, the king's chief poet, impressed by the saint's 
commanding presence, disobeyed the order and stood up 
as he entered. The narrative adds that this Dubthach 
became a convert, and thenceforward devoted his poetical 
talents to the service of God.* Mael Fothartaigh, son of 
Ronan king of Leinster, was so admired and loved, that 
at meetings of every kind " all would rise up before him."f 
St. Moiling, on a certain occasion, visited the house of 
Finaghta the Festive king of Ireland, but was received in 
a disrespectful manner by the company in general : — " He 
" found no uprising there [in honour of him] : and he was 

* Trip .Life, 283. f Kuno Meyer's Fingal Remain, Rev. Celt. ? xm. 372. 



" ashamed at not getting uprising."* It was not only a 
mark of respect, but — occasionally at least — something 
like an acknowledgment of superiority, as we may infer 
from a passage in the Irish Life of St. Finnchua. The 
saint having aided the king of Munster to defeat his 
enemies, stipulated that among other concessions " the king 
" of Munster should always stand up before Finchua's 
" successor "f : and still more clearly from the Crith 
Gabhlach extract given below, about a king and a bishop. 

Another mode of saluting to show respect was " raising 
the knee or thigh " while seated. On the occasion of the 
above-mentioned visit of St. Moiling, while no one actually 
stood up, Dermot the son of Colcu " raised his knee before 
him " — as did also his father Colcu — as a mark of reverence. 
St. Ruadan, when cursing Tara, says to King Dermot : — 
" The thigh [sliasait) that thou liftest not before me to 
stand up, be it mangled in pieces. "% Lifting the knee or 
thigh was not so great a mark of respect as standing up. 
In the Crith Gabhlach the question is asked, " Which is, 
higher, a king or a bishop ? " Answer : " The bishop is 
" higher, because the king stands up [to salute him] on 
" account of religion " : and the sentence concludes — " a 
bishop however raises his knee to a king."§ This lifting 
of the knee or thigh admits of an obvious explanation. A 
person sitting on a low seat — as the Irish generally sat — 
when about to stand up, naturally drew in one foot, which 
had the effect of raising the knee. So far this of itself 
was regarded as a mark of respect, being a preliminary to 
standing up : but to stand up altogether completed and 
increased the act of reverence. 

Giving a kiss, or more generally three kisses, on the 
cheek, was a very usual form of respectful and affectionate 

* Silva Gad., 420, bottom. See also, HyF, 143, 7> 
j Stokes, Lives of SS., 241, I7 . 
% See Silva Gad., 74, x (thigh) ; 83, 9 (knee). 
§ Br. Laws, iv. 339, bottom. 


salutation : it was indeed the most general of all. When 
St. Columba approached the assembly at Drum-ketta, 
" King Domnall rose immediately before him, and bade 
" him welcome, and kissed his cheek, and set him down 
" in his own place."* When Donnchad the son of King 
Concobar visited Conall Cernach, " Conall put his arms 
about his neck and gave him three kisses." Similar 
entries are found everywhere, both in the tales and in 
the ecclesiastical writings, showing that the practice was 
as prevalent among Christians as among pagans. 

A very pleasing way of showing respect and affection, 
which we often find noticed, was laying the head gently 
on the person's bosom. When Ere, King Concobar's 
grandson, came to him, " he placed his head on the 
breast of his grandfather, "f The old man Cailte, rising 
up in the morning, came to St. Patrick and laid his 
head on his bosom : and Patrick blessed him and said : 
" In whatsoever place God shall lay His hand on thee, 
heaven is in store for thee. "J Adamnan (pp. 35, 36) 
notices this custom, where he relates that King Aedan's 
son, Ochy Boy, saluted St. Columba respectfully by laying 
his head on his bosom. Sometimes persons bent the head 
and went on one knee to salute a superior§ : and in case 
of eminent saints, laymen often prostrated themselves on 
the two knees before them : to show both respect and 

It seems an odd way of manifesting respect for a man, 
to carry him on one's back for some distance : yet it was 
in this manner Muiredach, king of Leinster in the sixth 
century, acted towards St. Finnen of Clonard. Finnen 
had spent some time in Britain, and returning to Ireland, 
landed on the Leinster coast ; whereupon the king went 
down to meet him, and to show reverence carried him on 

* Adamnan, 38, note, 1st column J Silva Gad., 137, middle. 

t Rossnaree, 55. § Ibid., 298. 

|| Stokes, Lives of SS., lines 381, 2929, 4348, 4693. 


his back three several times across three fields near the 
harbour. Yet this over-condescension on the part of the 
king was evidently resented by some of his household ; 
for one of them made a sharp remark to the saint : 
" Thou art oppressive, O cleric, on the king " : on which 
Finnen replied : " The number of times that I have been 
" taken on his back will be the number of kings of his race 
" over the province " ; after which he pronounced a blessing 
on him.* It somewhat mitigates the humiliation of the 
king's action here when we know that carrying on the 
back was not unusual. Some of the early saints had 
in their household a fer-imchuir (' man of carrying ') or 
trein-fher (' strong-man ') for carrying them over fords or 
rough places. St. Patrick's " strong-man " has been already 
mentioned : and St. Ciaran's fer-imchuir was Mailoran. 
When St. Patrick was a boy his foster-father carried 
him home on his back ; and a man brings his sick 
mother on his back to St. Brigit to be healed. 

In the Life of St. Cormac in the Book of Lecan we are 
told that a certain chief named Dai " came and put his 
mouth to the floor out of humility to Cormac."f When 
St. Patrick visited Moylurg in Connaught, Bishop Maine 
" drove the [saint's] horses into a meadow, and cleansed 
their hoofs in honour of Patrick."! 

2. Pledging, Lending, and Borrowing. 

Although there were no such institutions in ancient 
Ireland as pawn-offices, pledging articles for a temporary 
loan was common enough. The practice was such a 
general feature of society that the Brehon Law took 
cognisance of it, and stepped in to prevent abuses. The 
Law did not prohibit charging interest on loans, though 
laws with this object existed in England, in supposed con- 
formity with the Mosaic dispensation (Exod., xxii. 25). 

* Stokes, Lives of SS., p. 224. f Stokes, Lives of SS., p. 343. 

J Trip. Life, 145. 


The Brehon Law, though laying down many precautions, 
did not contain any provision against usurious or excessive 
rates of interest : neither did early English law : for the 
British usury laws are of late introduction. 

Portable articles of any kind — including animals — 
might be pledged for a loan, or as security for the repay- 
ment of a debt ; and the law furnishes a long list of 
pledgable articles. The person holding the pledge might 
put it to its proper use while in his possession, unless 
there was express contract against it. But he was to use 
reasonably — he was not to injure articles of household or 
of ornament by rough usage, or a horse by overwork. He 
was moreover obliged to return the pledge on receiving a 
day's notice, provided the borrower tendered the sum 
borrowed, or the debt, with its interest : and if he failed to 
do so he was liable to fine. Suppose it was an ornament 
of gold or silver, such as a brooch ; and that it was not 
returned at the time required, so that the owner had to 
appear at any public function or celebration, such as a 
fair, without it : in this case the fine for withholding it was 
increased by a personal or honour fine for the indignity he 
suffered by being obliged to appear in public without his 
ornament.* If the pledged article was lost, or not returned, 
the value — and in some cases double the value — had to be 
made good. 

Sometimes a person borrowed an article on hire for the 
use of it, engaging to return it by a certain day. If he 
failed to return it at the proper time, besides what he gave 
for the use of it, he had to pay interest {tairgilli) on it for 
every day he withheld it. When a man lent an article 
without any charge — for mere kindness — fixing a day for 
its return, if it was not returned on the stipulated day, the 
borrower might be charged interest for it. 

There were distinct terms for all these transactions. 
A loan for kindness was called din : a loan for interest was 

* Br. Laws, v. 397, bottom. 


airhead : a loan in general was iasacht. Interest or usury 
was called tairgilli, and also fogaibthetu, a very old term, 
for it occurs in Zeuss, as noticed below. Borrowing or 
lending, on pledge, was a very common transaction among 
neighbours ; and it was not looked upon as in any sense 
a thing to be ashamed of, as pawning articles is at the 
present day. The practice moreover continued in use in 
Ireland down to comparatively recent times : of which 
an instance, occurring in the beginning of the eighteenth 
century, may be seen in the Kilk. Archaeol. Journal for 
1856-7, pp. 168, 169. 

It may be observed that the existence in ancient 
Ireland of the practice of pledging and lending for 
interest, the designation of the several functions by 
different terms, and the recognition of all by the Brehon 
Law, may be classed, among numerous other customs and 
institutions noticed throughout this book, as indicating a 
very advanced stage of civilisation. At what an early 
period this stage — of lending for interest — was reached 
may be seen from the fact that the MS. from which Zeuss 
took fogbaidetu, ' usura,' is the Wurzburg copy of St. Paul's 
Epistles which was glossed by some Irish monk in the 
eighth century.* 

3. Provision for Old Age and Destitution. 

Old age was greatly honoured, and provision was made 
for the maintenance of old persons who were not able to 
support themselves. The Brehon Law says : — " Age is 
rewarded by the Feine " : and " where there are two chiefs 
" of the same family who are of equal dignity and property, 
" the senior shall take precedence. The old man is entitled 

* Fogbaidetu, for fogaibthetu, is in the 1st edition of Zeuss, p. 844, last 
line : but Ebel has omitted it in his 2nd edition. Nearly all the regula- 
tions regarding borrowing and pledging given in this section, as well as 
many others, will be found in Br. Laws, in. 492, and note 3 ; 493, and 
495 : and in O'Curry, Man. & Cust., 11. 62, lit, 112, 113. 


" to good maintenance, and the senior is entitled to noble 
" election."* The scribe who copied this part of the Law 
prefixes a remark of his own which expresses still more 
strongly the veneration for age : — [I swear] " by this book 
" that so far as I can, I will, in the name of God, bring the 
" senior before the junior in every case, as these laws down 
" here state." 

When the head of a family became too old to manage 
his affairs, it was an arrangement sanctioned by the Law 
that he might retire, and give up both headship and land 
to his son, on condition of being maintained for the rest of 
his life. In this case, if he did not choose to live with his 
son, a separate house — commonly called a house of inchis 
— was built for him, the dimensions and furniture of which, 
as well as the dimensions of the little kitchen-garden, 
are set forth in the law. Three items of maintenance are 
distinguished and carefully specified : — food, milk, and 
attendance. As to attendance : among other things it is 
worth while to mention that he was to have a bath once 
at least every twentieth night, and his head was to be 
washed every Saturday. His supply of firewood is also 
specified, f 

If the old man had no children he might make over 
his property to a stranger on the same condition of due 
maintenance. Or he might purchase from the neigh- 
bouring monastery the right to lodge on the premises 
and board with the inmates : an arrangement common in 
England to a late period, where the purchased privilege of 
boarding and lodging in a monastery was called " Corrody."| 
This plan for providing for helpless old age, which was 
something like the present practice of purchasing an 
annuity, continued to prevail in Ireland down to at least 
the sixteenth century. In the seventh volume of the Proc. 

* Br. Laws, iv. 373 and note. 

t O'Curry, Man. & Cust., 11. 30, 31, 479, and note 515 : Br. Laws, iv. 
305, note 3. J Br. Laws, in., Richey, in Introduction, Ixv. lxvi. 


Roy. Ir. Academy (p. 18) Dr. Todd has published and 
translated a deed in the Irish language, dated 1522, by 
which an old lady gave up her lands to a man who had a 
mortage on them, on condition that he should support 
her for the rest of her life. 

As to old persons who had no means, the duty of main- 
taining them fell primarily of course on the children : or 
failing children, on the foster-child. A son who supported 
his father in old age had a special distinguishing term 
applied to him in the Senchus Mor — mac-gor or gormac : 
but this last term was also applied to a sister's son.* A 
son or daughter who was able to support parents but who 
evaded the duty was punished by having his or her dire- 
fine lessened (which meant loss of status : 1. 208, supra), 
or in some other way.f The general recognition of the 
son's duty to support his parents is noticed in a passage 
in Adamnan's " Life of Columba " (p. 159), where certain 
brothers who had been supporting their father and mother 
forced another brother who had been long absent to under- . 
take the duty. 

If an old person who had no children became destitute 
the tribe was bound to take care of him ; for in the words 
of the Senchus Mor : — " It is one of the duties of the fine 
(circle of relations, tribe) to support every tribesman " : 
and the Gloss adds : — " They do this by duties which are 
required of them according to justice. "J A usual plan was 
to send the old person to live with some family willing to 
undertake the duty, who had an allowance from the tribe 
for the cost of support. 

In some cases destitute persons dependent on the tribe 
who did not choose to live with a strange family, but pre- 
ferred to have their own little house, received what we now 

* Br. Laws, i. 207, , 9 . 111. 56, 20 . iv. 291, I0 . 43, note 1 ; and v. 71, 
line 4 from bottom. 

j Br. Laws, III. 53, 57, bottom ; iv. 185, 307, a? . see also IV., Richey, 
Introduction, lxvi. lxvii. } Br. Laws, in. 55, 2; 57, 9 . 


call outdoor relief. According to the Sequel to the Crith 
Gabhlach, there was a special officer called uaithne [oohina : 
lit. a ' pillar '] whose business it was to look after them : or, 
in the words of the tract, to " oversee the wretched and the 
poor," and make sure that they received the proper allow- 
ance : something like the relieving officer of our present 
poor laws. He was of course paid for this duty ; and it is 
added that he should bear " attacks on his honour " without 
his family or himself needing to take any action in the 
matter — referring to the abuse and insult he was likely to 
receive from the peevish and querulous class he had in 
charge. He was permitted to bear from them insults, 
which, if coming from ordinary members of the com- 
munity, should be resented in the recognised way to wipe 
off the disgrace.* 

It is plain from some expressions in the Senchus Mor, 
as well as from the words of the Glossator, that in those 
times, as at the present day, there were poor persons who 
preferred the free and easy life of the wandering beggar- 
man| : as reflected in one verse of a once popular modern 
Anglo-Irish song : — 

" Of all trades a-going begging it is my delight ; 
My rent it is paid, and I lay down my bags every night : 
I'll throw away care, and take a long staff in my hand ; 
And I'll flourish each day courageously looking for chance." 

They carried a bag for contributions ; and the Senchus 
Mor mentions giving them alms as a commendable deed, 
here transmitting a record of a popular feeling that has 
continued to prevail to our own day. Shane O'Neill — 
John the Proud — prince of Tyrone in the time of 
Elizabeth, always put aside the first dish of food for the 
poor : — " to serve Christ first " — as he said. More than 
half a century ago I saw the same custom carried out — 

* Br. Laws, iv. 351 : Sullivan, Introduction, 251, bottom, 
f Br. Laws, in. 19, 2Si 21, top. 


though in less regal style — in my grandfather's house, 
where, just before the family sat down to the noonday 
dinner, a big dish of laughing potatoes was always laid 
aside for wandering beggars : and the potatoes rarely 
survived till night. 

From the provisions here described it will be seen that 
the most important features of our modern poor-laws were 
anticipated in Ireland a thousand years ago. 

4. Irish Poetry and Prosody : Love of Nature and of 
Natural Beauty. 

In very early times, not only poetry proper, but 
histories, biographies, laws, genealogies, and such like, 
were often written in verse as an aid to the memory. 
Among all peoples there were — as there are still — certain 
laws or rules, commonly known as Prosody, which poets 
had to observe in the construction of their verse : of which 
the main object was harmony of numbers. The classifi- 
cation and the laws of Irish versification were probably 
the most complicated that were ever invented : indicating 
on the part of the ancient Irish people, both learned and 
unlearned, a delicate appreciation of harmonious combi- 
nations of sounds. The following statement will give the 
reader an idea of this. There are in Irish three principal 
kinds of verse. Of the first kind, which is called " Direct 
Metre " (Dan Direch), there are five species, all equally 
complicated. The first of these required the observance 
of the following rules : — (1) Each stanza to consist of 
four lines making complete sense ; (2) In each line seven 
syllables ; (3) Alliteration in at least two principal words 
of each line ; (4) The lines to rhyme, the rhymes being 
greatly varied, and occurring very often ; (5) The last 
word of the second line to have one syllable more than 
the last word of the first line ; a like relation between the 
last words of the fourth and third lines. 

K 1 


In Irish poetry of all kinds the rhymes were very 
frequent, occurring, not only at the ends of the lines, but 
also within them, once, twice, or even three times. The 
rhymes were either between vowels — i.e. assonances — or 
Detween consonants. For this last purpose the consonants 
were scientifically divided into six classes, " soft," " hard," 
" rough," " strong," " light, and " the queen," i.e. the letter 
s which formed the sixth class : the letters of each of the 
first five corresponding and rhyming with each other, but 
not with those of any other class. One-syllable, two- 
syllable, and three-syllable rhymes were used with equal 

That the old writers of verse were able to comply with 
these numerous difficult prosodial rules we have positive 
proof in our manuscripts ; and the result is marvellous. 
No poetry of any European language, ancient or modern, 
could compare with that of Irish for richness of melody. 
Well might Dr. Atkinson exclaim (in his Lecture on Irish 
Metric, p. 4) : — " I believe Irish verse to have been about 
" the most perfectly harmonious combination of sounds 
" that the world has ever known. I know of nothing in 
" the world's literature like it." 

Of each principal kind or measure of verse there were 
many divisions and subdivisions, comprising altogether 
several hundred different metrical varieties, all instantly 
distinguishable by the trained ears of poet and audience.* 
We have seen that there were seven grades of " Poets." 
Of the lower class, called " Bards," there were also a 
number of grades. Each of the grades of both had certain 
metres allotted to them ; and each individual was allowed 

* See Irish Metric by Dr. Atkinson : the Irish Treatise on Irish Metre 
from the Book of Ballymote, translated and annotated by the Rev. B. 
M'Carthy, d.d., in the Todd Lecture Series of the Roy. Ir. Acad., vol. in., 
pp. 98-141 : Stokes's Feilire, Introduction, 12-15 : Hyde, Lit. Hist., 
chapters xxxvi. to xxxviii. ; and his little book, Irish Poetry (1903), in 
which will be found a most useful survey of Irish poets and poetry, 
with their manifold subdivisions. 


to compose only in his own special measure, or in those 
belonging to the inferior grades ; but he was not permitted 
to compose in the measure of any grade above him. 

Those of the Irish poets — whether clerical or lay — who 
learned to write Latin, imported many of the Irish pro- 
sodial rules into their Latin poetry, using accent instead 
of quantity, so as to imitate exactly the metres, assonantal 
rhymes, alliterations, consonantal harmonies, and all the 
other ornaments common in Irish poetry.* 

Some of the greatest Celtic scholars that ever lived — 
among them Zeuss and Nigra — maintain that rhyme, now 
so common in all European languages, originated with the 
old Irish poets, and that from the Irish language it was 
adopted into Latin, from which it gradually penetrated to 
other languages, till it finally spread over all Europe, f 
But other eminent men, including the German scholar 
R. Thurneysen, think that rhyme was mainly borrowed 
by the Gaels from the Romans. The preponderance of 
learned opinion is certainly on the side of the Gaels ; but 
the subject requires to be further investigated before 
final judgment can be pronounced. One thing is quite 
certain, that rhyme — as we have already said — was 
brought to far greater perfection in Irish than in any 
other language. | 

It has been stated (p. 497) that laws, histories, genea- 
logies, and such like, were often written in verse to make 
them more easily remembered. Several pieces of this kind 
have been published, of which the following — all or most 
mentioned and described elsewhere in this book — are the 

* See the Irish Liber Hymnorum, by Drs. Atkinson and Bernard, 11. xv. 

f All that can be said at present in favour of this view will be found 
in Dr. Sigerson's Introduction to his Bards of the Gael and Gall, and in 
Dr. Hyde's Irish Poetry, pp. 45-57. See also Hyde's Lit. Hist., chap, xxxvi. 

J Dr. Douglas Hyde has given, in his Literary History of Ireland, and 
in his Irish Poetry, English translations of many old Irish poems, in 
which the rhymes, metres, and alliterations of the originals are exactly 
imitated. Dr. Sigerson, in his Bards of the Gael and Gall, also often 
imitates the old rhymes and alliterations. 


principal. The Feilire of Aengus ; the Book of Rights ; 
a portion of the Tribes and Customs of Hy Fiachrach ; 
the Duan Eirenneach and the Duan Albanach, two long 
pieces of versified history published by Dr. Todd in his 
Irish version of Nennius ; and the Topographical Poems 
of O'Dugan and O'Heeren. This last is an enumeration 
of the principal tribes of Ireland at the time of the English 
invasion, with the districts they occupied and the chiefs 
who ruled over them. The part relating to Leth Conn 
(North of Ireland) was written by John O'Dugan, who 
died in 1372 ; and that relating to Leth Mow (South) by 
Gilla-na-neeve O'Heeren, who died in 1420. The whole 
poem has been translated and annotated by O'Donovan, 
and published by the Irish Archaeological and Celtic 
Society. In such compositions as these we could hardly 
expect to find true poetry, for they are little more than 
mere catalogues in verse. 

The complicated restrictions of their prosody must 
have greatly hampered the play of the old Irish poets' 
imagination, so much of their energies were concentrated 
on overcoming mere mechanical difficulties. Yet in spite 
of this, they produced a great body of very beautiful poetry. 
Sufficient materials are not yet available to enable us to 
pass a general judgment on the character of early Irish 
poetry: for, as has been remarked, such pieces as the 
Feilire of Aengus are not, and never were intended to be, 
poetry, any more than the versified lists of kings and 
events we often see in modern English school-books. Yet 
these have been almost the only ancient Irish metrical 
compositions that have hitherto been brought within reach 
of the public, and their prominence is due to their historical 

The great majority of Irish poetical pieces— poetry 
in the true sense of the word— are still hidden away in 
manuscripts scattered through the libraries of all Europe. 
The few that have been brought to light, through the 


investigations of scholars of taste, show that many of the 
ancient Irish poets were inspired with true poetical genius. 
Most of these pieces are characterised by one prevailing 
note-a close observation and an intense love of nature in 
all its aspects. As favourable specimens may be instanced 
King and Hermit," lately translated and published (in 
pamphlet form) by Professor Kuno Meyer (iqoi) "a 
singularly beautiful poem," as he designates it, written in 
he tenth century : Deirdre's farewell to Alban, of which a 
translation is given below; the poetical description of 
the pagan heaven in vol. i., p. 2g4> su p ra> and another and 
different one with text and translation by Kuno Meyer 
m Mr Nutt's "Voyage of Bran," vol. l t p. 4 . Good 
examples of Irish poetry also are the series of verse-pieces 
interspersed through the prose of » The Courtship of Ferb " 
all of which are very simple, and most of them pathetic 
and picturesque. Many other very beautiful examples 

SeanoLh." " P ° etiCal *"* ° f ^ A ^ l ™ h na 

On this subject I cannot do better than quote a few 

weisn poetry, I know nothing to place by their side. 

to 'and Xse b eivenTerI red ** a " *" « »I*— both those referred 
inasmneh a^thrrfch me^rT*.' ^ disad ™">«= * translation, 

— * ^c sx»rsr x a o A t u a„r y 


The poem here published [" King and Hermit "] affords a good example 
of that marvellous descriptive art of Irish poets, which they share 
with the Welsh bards."* 

The poet's adage, " A thing of beauty is a joy for ever," 
found real and concrete application among the ancient Irish. 
Their poetry, their tales, and even their proper names, bear 
testimony to their intense love of nature and their apprecia- 
tion of natural beauty. Keats, in the opening of "En- 
dymion," enumerates various natural features and artificial 
creations as " things of beauty," among others, the sun, 
the moon, " trees old and new," clear rills, " the mid-forest 
brake," " all lovely tales that we have heard or read." 
These and many other features of nature and art, not men- 
tioned by Keats — the boom and dash of the waves, the cry 
of the sea-birds, the murmur of the wind among the trees, 
the howling of the storm, the sad desolation of the land- 
scape in winter, the ever-varying beauty of the clouds, the 
cry of the hounds in full career among the glens, f the 
beauty of the native music, tender, sad, or joyous, and 
soforth in endless variety — all these are noticed and dwelt 
upon by those observant old Irish writers — especially in 
their poetry — in words as minutely descriptive and as 
intensely appreciative as the poetry of Wordsworth. 

It would be easy to multiply instances of this bent of 
mind, but we must be contented with a few illustrations 
here. An excellent example is Midir's address to the 
lady Befinn, already given (vol. I., p. 294). In the Life of 
St. Senan — a plain prose narrative — it is related that a 
child, playing beside its mother near a cliff in the west 
of Clare, fell " over the edge of Ireland " into the sea, but 
was preserved from injury by the intercession of St. Senan. 

* Many similar testimonies might be adduced from native Irish writers : 
but I prefer — for obvious reasons — to quote from this scholarly and dis- 
cerning foreigner. 

t The delight of the Irish, both ancient and modern, in hunting and 
its accessories has been already noticed (pp. 467, 468, supra). 


A person who heard the shrieks of the mother went down 
to look for him, and found him sitting quite safe in the 
trough of the sea where he had fallen, " playing with the 
" waves. For the waves would reach up to him and laugh 
" round him, and he was laughing at the waves, and putting 
" the palm of his hand to the foam of the crests, and he 
" used to lick it like the foam of new milk."* 

They loved the music of the wind and the waves. The 
sound of the breeze rustling through the foliage so struck 
the imagination of those spiritual people, that in Cormac's 
Glossary (p. 132) the word omna, one of the Irish names 
for the oak-tree, is derived from fuamna, " sounding " : 
" because " — in the words of the Glossary — " great is the 
sound of the wind blowing against it." In the Life of 
St. Columkille it is stated that while residing in Iona, he 
wrote a poem in the Irish language, a tender reminiscence 
of his beloved native land, in which he expresses himself 
in this manner : — 

St. Columkille's Remembrance of Erin. 

" How delightful to be on Ben-Edarf before embarking on the foam- 
white sea ; how pleasant to row one's little curragh round it, to look 
upward at its bare steep border, and to hear the waves dashing against 
its rocky cliffs. 

" A grey eye looks back towards Erin ; a grey eye full of tears. 

" While I traverse Alban of the ravens, I think on my little oak 
grove in Derry. If the tributes and the riches of Alban were mine, 
from the centre to the uttermost borders, I would prefer to them all 
one little house in Derry. The reason I love Derry is for its quietness, 
for its purity, for its crowds of white angels. 

" How sweet it is to think of Durrow : how delightful would it be 
to hear the music of the breeze rustling through its groves. 

" Plentiful is the fruit in the Western Island — beloved Erin of many 
waterfalls : plentiful her noble groves of oak. Many are her kings 
and princes ; sweet-voiced her clerics ; her birds warble joyously in 
the woods ; gentle are her youths ; wise her seniors ; comely and grace- 

* Stokes, Lives of SS., p. 212. 

| Ben Edar, Howth, a rocky headland near Dublin. 


ful her women, of spotless virtue ; illustrious her men, of noble 

" There is a grey eye that fills with tears when it looks back to- 
wards Erin. While I stand on the oaken deck of my bark I stretch my 
vision westwards over the briny sea towards Erin."* 

When the sons of Usna returned from Scotland to 
Ireland at the treacherous invitation of King Concobar, 
Deirdre, the wife of Naisi one of the brothers, seating 
herself on a sea-cliff of the present county Antrim, looked 
sadly over the waters at the blue headlands of Scotland, 
with gloomy forebodings for the future, and uttered this 
farewell : — 

Deirdre's Farewell to Alban. 


" Dear to me is yon eastern land : Alban with its wonders. Beloved 

is Alban with its bright harbours and its pleasant hills of the green 

slopes. From that land I would never depart- except to be with 


" Kil-Cuan, O Kil-Cuan, f whither Ainnli was wont to resort : short 
seemed the time to me while I sojourned there with Naisi on the margins 
of its streams and waterfalls. 

" Glen-Lee, O Glen-Lee, where I slept happy under soft coverlets : 
fish and fowl, and the flesh of red deer and badgers ; these were our 
fare in Glen-Lee. 


" Glen-Masan, O Glen-Masan : tall its cresses of white stalks : often 
were we rocked to sleep in our eurragh in the grassy harbour of Glen- 


" Glen-Orchy, O Glen-Orchy : over thy straight glen rises the smooth 
ridge that oft echoed to the voices of our hounds. No man of the 

* The two Irish poems, with translations, from which the above is 
extracted, will be found in Reeves's Adamnan, pp. 275 and 285. They 
are very ancient ; and they illustrate our theme equally well, whether 
they were written by St. Columkille or by any other Irishman. 

■f This and the other places named in Deirdre's Farewell are all in the 
west of Scotland. 


clan was more light-hearted than my Naisi when following the chase 
in Glen-Orchy. 


" Glen-Ettive, O Glen-Ettive : there it was that my first house was 
raised for me : lovely its woods in the smile of the early morn : the sun 
loves to shine on Glen-Ettive. 


" Glsn-da-Roy, O Glen-da-Roy : the memory of its people is dear 
to me 5 sweet is the cuckoo's note from the bending bough on the peak 
over Glen-da-Roy. 


" Dear to me is Dreenagh over the resounding shore : dear to me 
its crystal waters over the speckled sand. From those sweet places I 
would never depart, but only to be with my beloved Naisi."* 

The singing of birds had a special charm for the old 
Irish people. Comgan, otherwise called Mac da Cherda 
(seventh century: vol. 1., page 224, supra), standing on 
the great rath of Cnoc-Rafann (now Knockgraffon in 
Tipperary), which was in his time surrounded with woods, 
uttered the following verse, as we find it preserved in 
Cormac's Glossary (p. 7) : — 

" This great rath on which I stand 
Wherein is a little well with a bright silver drinking-cup : 
Sweet was the voice of the wood of blackbirds 
Round this rath of Fiacha son of Moinche."f 

Among the numerous examples of Metre given in a 
treatise on Prosody in the Book of Ballymote is the 

* From Joyce's " Old Celtic Romances," where the whole tale is 
given. The original Irish text of this beautiful poem, which was first 
published, with translation, by Theophilus O'Flanagan, in his version of 
the Story of the Fate of the Sons of Usna, in Trans. Gael. Soc, 1808, has 
been lately republished, with the tale itself, by the Society for the 
Preservation of the Irish language. For a metrical version, see Ferguson, 
Lays of the Western Gael. 

f That is Fiacha Muillethan, king of Munster in the third century, 
who resided at Cnoc-Rafann : see p. 103, supra. 

506 Social And domestic life [part Hi 

following verse, selected there merely for a grammatical 
purpose : — 

" The bird that calls within the sallow- tree, 
Beautiful his beak and clear his voice ; 

The tip of the bill of the glossy jet-black bird is a lovely yellow; 
The note that the merle warbles is a trilling lay."* 

It would be hard to find a more striking or a prettier 
conception of the power of music in the shape of a bird- 
song, than the account of Blanid's three cows. When the 
Isle of Man was invaded by the Irish Red Branch heroes, 
they took the king's castle and brought away all the jewels ; 
among them — as the best jewel of all — the lady Blanid, 
daughter of the king, together with her great brazen 
caldron, her three cows, and also three little birds that 
used to sing for them while milking. These cows were 
always milked into the caldron, but submitted reluctantly 
and gave little milk till the birds came to their usual perch 
— on the cow's ears — and sang for them : then they gave 
their milk freely till the caldron was filled. f See also 
vol. I., p. 591, supra. 

Even the place-names scattered over the country — 
names that remain in hundreds to this day — bear testi- 
mony to this pleasing feature of the Irish character : 
for we have numerous places still called by names with 
such significations as " delightful wood," " silvery stream," 
" cluster of nuts " (for a hazel wood), " prattling rivulet," 
" crystal well," " the recess of the bird- warbling," " melo- 
dious little hill," " the fragrant bush-cluster," and soforth 
in endless variety. J 

* Mac Carthy, Cod. Pal. -Vat. 131. Other examples in Silva Gad. 109 : 
Stokes and Strachan's Thes. 11. 290 : and Hyde, Lit. Hist., 275. 

| Kilk. Arch. Journ., 1870-1, p. 411. 

% For the originals of all the above names, and for numerous others of 
a like kind, see Irish Names of Places, vol. 11., chap, iv., on " Poetical 
and Fancy Names." 


Many students of our ancient literature have noticed 
these characteristics. " Another poem " — writes Mr. Alfred 
Nutt in his " Studies on Ossian and Ossianic Literature " 
(p. 9) — " strikes a note which remains dominant throughout 
" the entire range of Ossianic Literature : the note of keen 
" and vivid feeling for certain natural conditions. It is a 
" brief description of winter : — 

" ' A tale here for you : oxen lowing : winter snowing : summer passed 
away : wind from the north, high and cold : low the sun and short 
his course : wildly tossing the wave of the sea. The fern burns deep red. 
Men wrap themselves closely : the wild goose raises her wonted cry : 
cold seizes the wing of the bird : 'tis the season of ice : sad my tale.' " 

This is a bald literal translation : what would it be if 
dressed in the diction of Scott, Goldsmith, or Wordsworth ! 
Dr. Whitley Stokes, speaking of the " Colloquy of the 
Ancient Men " — an Ossianic composition consisting of a 
series of short narratives framed in one connecting story — 
notices the genuine feeling for natural beauty and the 
passion for music that pervade the whole of them.* In 
this connexion I may again also point to the beautiful 
poem of " King and Hermit," already mentioned, and 
Professor Kuno Meyer's Introduction to it.f 

So far we have had under consideration the Irish 
poetry of the early centuries. In later times the Irish 
poets broke away from their ancient prosodial trammels 
and produced much excellent poetry. Among the remains 
of these times — from the fifteenth century down — we have 
many pieces of great beauty — odes, ballads, elegies, songs, 
&c. — the products of true poetical inspiration. Spenser, 
a supreme judge of poetry, but in general a prejudiced 

* Preface to the Acallamh : Ir. Texte, iv. xii. 

•f Excellent renderings into English rhymed metre — many of them very 
beautiful — of about 140 Irish poems, of various ages, from the most 
ancient to modern times, have been given by Dr. Sigerson in his Bards 
of the Gael and Gall. 


witness for Ireland, and not disposed to praise things Irish, 
has given the following testimony regarding this poetry of 
the later Irish bards (View, 124) : — " Yea, truly I have 
" caused divers of them to be translated unto me, that I 
" might understand them, and surely they savoured of 
" sweet wit and good invention, but skilled not of the 
" goodly ornaments of poetry [i.e. they wanted the qualities 
" that go to form great poetry] ; yet were they sprinkled 
" with some pretty flowers of their natural device, which 
" gave good grace and comeliness unto them."* 

In modern Irish poetry the old prosodial rules are 
almost wholly disregarded. The rhymes are assonantal, 
and very frequent : they occur not only at the ends of the 
lines but within them — sometimes once, sometimes twice ; 
and not unfrequently the same rhyme runs through several 
stanzas. In other respects modern Irish poetry generally 
follows the metrical construction of English verse, f 

* The question of the change from the old order of things in Irish 
poetry, and its effects on the productions of the bards, is too extensive to 
be dealt with adequately here. The reader may consult Dr. Hyde's very 
full treatment of this subject in his Lit. Hist, of Ireland, pp. 479 et seq. 

f For the metrical laws observed by the Irish poets in the eighteenth 
century, see the " Introductions " of the Rev. Patrick S. Dinneen, m.a., 
to his two volumes, the Poems of Egan O'Rahilly, and Amhrain Eoghain 
Ruaidh Ui Shuilledbhdin (The Poems of Owen Roe O'Sullivan). 

In regard to rhyming, Shelley's poem, " The Cloud," resembles some of 
these modern Irish pieces : but his rhymes are what are called in English 
prosody " perfect," not assonantal. This specimen will show what is 
meant : — 

" I bring fresh showers for the thirsting flowers 
From the seas and the streams ; 
I bear light shade for the leaves when laid 
In their noonday dreams.' 

The Cork poet, Fitzjames O'Brien, directly imitated the native poetry 
in his poem on " Lough Hyne " (also with perfect, not assonantal, rhyme) : 

" I know a lake where the cool waves break, 
And softly fall on the silver sand ; 
And no steps intrude on that solitude 

And no voice save mine disturbs the strand." 

See also, for other instances, Father Prout's " Bells of Shandon," 
Moore's song, " Wreathe the Bowl," and Curran's " If Sadly Thinking." 

The peasant-poets of a century ago, or more, when composing 
their doggerel songs in English — with which language they were only 


5. Six Stages of Life. 

Shakespeare's seven stages of human life are repre- 
sented by : — 1. the infant ; 2. the schoolboy ; 3. the lover ; 
4. the soldier ; 5. the justice ; 6. old age ; 7. decrepitude. 
The ancient Irish, long before the time of Shakespeare, 
divided life into six stages — called " Columns of Age " 
(colomna dis) — which are simply enumerated in Cormac's 
Glossary (p. 41), without any poetical setting : — 1. infancy 
{ndidendacht) ; 2. boyhood (macdacht) ; 3. youth or puberty 
(gillacht) ; 4. adolescence or manhood (hoclachus) ; 5. old 
age (sendacht) ; 6. decrepitude (diblidecht or dimligdetu). 
Here two of the Irish stages — youth and manhood — are 
represented by three — lover, soldier, and justice — in 

6. Human Temperaments. 

Dr. Whitley Stokes, in the Preface to his Three Irish 
Glossaries (p. xl.), gives a curious myth, from an old Irish 
manuscript in the British Museum, to account for the 
different dispositions and temperaments of mankind, 
which, as he remarks, corresponds with similar legends 
of Teutonic and Indian origin. The following are the 
opening words of his translation : — " It is worth knowing 
" what Adam was made of, namely of eight parts : — the 
" first part, of earth ; the second part, of sea ; the third 
" part, of sun ; the fourth part, of clouds ; the fifth part, 
" of wind : . . . ; the seventh part, of the Holy Ghost ; 
" the eighth part, of the light of the world." 

The old account then goes on to explain the tempera- 
ments from these components : — The earth part is man's 

inperfectly acquainted — imitated, by a sort of hereditary instinct, the 
assonantal rhymes of their native language. The following is a good 
specimen the rhyming syllables are italicised : — 

" The grand improvements they would ajnuse you, 
The trees are drooj>'vng with fruit all kind ; 
The bees per/»ming the fields with music 
Which yields more beauty to CastleAyde." 


body, and if this predominates in a man he will be slothful. 
The part of the sea is the blood, and this in excess makes 
a man changeful. The part of the sun is his face and 
countenance, which will be lively and beautiful if that part 
is prominent. The part of the clouds, . . . The part of 
the wind is the breath, and if that part prevails he will be 
of a strong character. The part of the Holy Ghost is the 
soul, the predominance of which will make him lively, of 
good countenance, full of grace and of the divine Scripture. 
The part that was made of the light of the word is piety, 
and if this prevails he will be a loving, sensible man.* 

7. Blood-Covenant. 
Giraldus Cambrensis, in his " Topography of Ireland " 
(in., xxii), says that, when the Irish make a very solemn 
league, they ratify it " by drinking each other's blood, which 
they shed for this purpose. This custom " — he goes on 
to say — " has been handed down to them from the rites 
" of the pagans, who were wont to confirm their treaties 
" with blood " : after which he proceeds to add some 
comments, in his usual bitter style. This statement has 
been considered as one of Giraldus's libels ; and it has 
been warmly denied by Keating in the Preface to his 
" History of Ireland " (p. xxvii), by Lynch in his " Cambren- 
sis Eversus " (in., 217, 221), and by Lanigan in his " Ecclesi- 
astical History " (iv., 285). But here Giraldus, though he 
distorts, and pours forth a volume of rancorous abuse and 
exaggeration, does not bear absolutely false witness. That 
the Irish had a cro-cotaig, or ' blood-covenant ' (cro, blood, 
cognate with Latin cruor), admits of no doubt. In the 
year 598 A.D., when Branduff, king of Leinster, was pre- 
paring to fight the king of Ireland in the Battle of Dunbolg, 

* Dr. Stokes has given a poetical paraphrase of this in English (" Man 
Octipartite "), which may be seen in Stopford Brooke and Rolleston's 
Treasury of Irish Poetry, p. 348. In the Preface to Three Irish Glossaries, 
Dr. Stokes refers, for corresponding myths among other ancient nations, 
to several authorities — German, Sanskrit, Latin, and Old English. See 
also Rev. Celt., 1. 261. 


as told in vol. I., p. 141, supra, he met, on the side of a 
mountain at some distance from both camps, a party of 
Ulidians (from East Ulster), who had formed part of the 
king of Ireland's army ; and induced them to abandon the 
monarch and join his own standard. The king of Ulidia 
then said : — " A blood-covenant (cro-codaig) and an agree- 
ment shall be made between us ": whereupon — in the 
words of the old narrative — " they seated themselves on the 
" mountain and made a cotach or bond of fellowship that 
" should never be broken." The king of Ireland was 
defeated and slain in that great battle : and the old 
history adds that the mountain, which had been called 
before that time Slieve Nechtain, was afterwards called 
Slieve Codaig, the ' mountain of the covenant.' This moun- 
tain lies on the left of the road as you go from Hollywood 
to Donard in Wicklow ; and it retains the name to this 
day in the slightly altered form of Slieve Gadoe.* 

Here there is no hint — further than the name — as to 
the nature of the covenant, or what exactly was done : for 
the good reason that it was so well understood in the time 
of the writer, that he needed only to name it. But we get 
the details from Martin, f as he found the custom practised 
down to his time, eleven centuries later, among our cousins, 
the people of the Western Islands of Scotland, who derived 
it from Ireland. His testimony is moreover independent, 
as he knew nothing of the early Irish custom. He says of 
the islanders : — " Their ancient leagues of friendship were 
" ratified by drinking a drop of each other's blood, which was 
" commonly drawn out of the little finger. This bond was 
" religiously observed as a religious bond." And he goes 
on to say that whoever violated it utterly lost character, so 
that all people avoided him. The ancient Irish custom 
was something like this : and the blood that each person 
drank consisted of portion of a single drop, mixed with 

* Silva Gad., 413 : Rev. Celt., XIII. 73 : also Irish Names of Places, 
11. 463. t Western Isles, 109. 


water. So that Giraldus's " drinking each other's blood " 
must be toned down to this rather mild formality. 

Sometimes the Irish ratified covenants by having them 
written in blood. When St. Cairnech brought about a 
league between the Hy Neill and the Cianachta, i.e. the 
people of Keenaght in Meath, as we are told in the 
old tale of the Death of Murkertagh mac Neill, king of 
Ireland (a.d. 512-533), he mixed the blood of both tribes 
in one vessel, and wrote with it the treaty : thus rendering 
it inviolable.* This, as we see, was done under Christian 

The blood-covenant was not peculiar to Ireland and 
Scotland. It came down from primitive times : it was 
once prevalent in many countries, and exists among some 
people to this day. Herodotus, in his account of King 
Croesus, records that the Medes and Lydians made oaths 
in treaties like the Greeks, but added this ceremony : — 
they scratched their arms till they drew a little blood, and 
each licked the blood off the other's arm. Professor Max 
Mullerf writes : — 

" Another widely-spread custom is the drinking of blood, as the 
highest sanction of a promise or a treaty. Herodotus (in. 8) alludes 
to this custom as existing in Arabia ; and how long it prevailed and 
how firmly it was established we may gather from the fact that Moham- 
med had to forbid it as one of the heavy sins — idolatry, neglect of duties 
towards parents, murder, and the blood-oath." 

8. Cremation-ashes thrown into water. 

In some cases the body of an animal or of a human 
being was burned, and the ashes thrown into water — 
generally a stream — with the idea of removing some 
malign influence, or obliterating the memory or the effects 
of a crime. According to an ancient legend, which is fully 
given by Keating (pp. 336, 338), a series of calamities fell 
on Munster in the third century of the Christian era. And 

* Petrie's Tara, 121, note f Anthropological Religion, p. 191 


when the nobles inquired into the matter, they were told 
that all the misfortunes were caused by two sons who were 
born to their king — the offspring of incest. Whereupon 
they demanded that the boys should be given up to them, 
" that they might consume them with fire and cast their 
ashes into the running stream," with the object of putting 
an end to the trouble. Whether the design was carried 
out is not stated. 

When Diancecht the Dedannan leech-god (vol. I., p. 261, 
supra) slew Meichi the son of the war-witch Morrigan, he 
found in the body three hearts in the shape of serpents' 
heads, which, had they been allowed to grow, would have 
destroyed all the animals in Ireland. Diancecht burned 
these three venomous snake-heads, and threw the ashes 
into the current of the river Barrow, which heated the 
water and caused it to seethe up so violently that it killed 
and boiled to rags all the fish along the whole course of 
the river.* We have already seen (p. 456, supra) that the 
body of a mad dog should be burned and the ashes thrown 
into a running stream. 

Maildune and his companions landed from their currach 
on an island, where they found a palace inhabited by only 
one little cat. A great collection of torques and other 
precious jewels hung round the walls, which Maildune told 
his men not to interfere with. His foster-brother however, 
disregarding the injunction, took down one of the torques 
and brought it away. But the cat followed him and over- 
took him in the middle of the court, and, springing on him 
like a blazing fiery arrow, went through his body and 
reduced it in a moment to a heap of ashes. He then 
returned to the room, and, leaping up on a low pillar, 
sat upon it. Maildune turned back, bringing the torque 
with him, and, approaching the cat, spoke some soothing 
words ; after which he put the torque back to the place 
from which it had been taken. Having done this, he 

* O'Grady, Silva Gad., 524, top 



collected the ashes of his foster-brother, and, bringing 
them to the shore, cast them into the sea.* The practice 
of burning the dead, and throwing the ashes into the sea, 
prevailed also among the Scandinavian nations. 

9. Something further about Animals. 

There are not, and never have been, any venomous 
reptiles in Ireland. There are small lizards, five or six 
inches long, commonly called in Irish, art- or arc-luachra, 
' lizard of the rushes,' but they are quite harmless. 
St. Patrick is credited in legend with freeing the island 
from venomous and demoniac reptiles ; but two centuries 
before his time, Solinus wrote regarding Ireland, illic 
nullus anguis, " there is no snake." Giraldus Cambrensis, 
in the twelfth century, testified that there were no snakes 
or adders, toads, or frogs ; and in an Irish ms., quoted by 
Stokes, f Ireland is compared to Paradise, which has no 
venomous reptiles and no frogs. According to Giraldus, 
the first frog ever seen in Ireland was found in his own 
time in a meadow near Waterford : but recently our 
naturalists have discovered a native frog, or rather a 
small species of toad, in a remote district in Kerry, which 
must have cunningly eluded the eye of St. Patrick, for 
they have been in the place from the beginning. 

But though we have no great reptiles in nature, we are 
amply compensated by legends, both ancient and modern, 
according to which there lives at the bottom of many of 
the Irish lakes a monstrous hairy serpent or dragon, 
usually called piast or heist, i.e. " beast," from Latin bestia ; 
and sometimes nathir, i.e. " serpent " ; chained there by a 
superior power — commonly credited to St. Patrick — till 
the day of judgment. Our most ancient literature, pagan 
as well as Christian and ecclesiastical, abounds in legends 
of those frightful reptiles. Sometimes they guard a liss 

* Old Celtic Romances, 133. f ^ n Trip. Life, Introduction, xxix. 


or fort. The legend is as prevalent to-day as it was a 
thousand years ago : and very many lakes have now, as 
the people say, a frightful monster with a great hairy 
mane, at the bottom.* 

But we had a much more gigantic and much more 
deadly sea-monster than any of these — the Rosualt — a 
mighty animal that cut a great figure in Irish tales of the 
olden time. There is a well-known plain at the foot of 
Croagh Patrick mountain, called Murrisk, where the body 
of this sea-monster was cast ashore, from which the name 
is derived. For according to the legendary account in the 
Book of Leinster, Murrisk is only a shortened form of 
Muir-iasc [Murreesk], meaning ' sea-fish.' When the 
Rosualt was alive — which was in the time ot St. Columkille 
— he was able to vomit in three different ways three years 
in succession. One year he turned up his tail, and with 
his head buried deep down, he spewed the contents of his 
stomach into the water, in consequence of which all the 
fish died in that part of the sea, and currachs and ships 
were wrecked and swamped. Next year he sank his tail 
into the water, and rearing his head high up in the air, 
belched out such noisome fumes that all the birds fell 
dead. In the third year he turned his head shoreward and 
vomited towards the land, causing a pestilential vapour to 
creep over the country, that killed men and four-footed 
animals, f 

St. Brendan, during his celebrated voyage in the 
Atlantic, once landed on the back of a huge fish, think- 
ing it was an island, and his companions made a fire and 
began to cook their supper ; but the beast, no doubt 
feeling the heat inconvenient, suddenly sank into the sea, 
and they barely escaped with their lives into their boat. 

* For more about these supernatural aquatic monsters see Adamnan, 
11. xxvii : IarC, p. 19 : O'Cl. Cal., 145 : Tain Bo Fr., 149, 157 1 Silva 
Gad., 283 : Joyce, Irish Names of Places, vol. 1. 198. 

t LL, 167, b, 46 ! Silva Gad., 480, a . 527, 9 . Rev. Celt., vol. 1. 258. 


The name of this great fish was Jasconius,* from which 
we infer that it was an immense eel, for " Jascon-ius " — 
pron. Yasconius — is merely the Latin way of writing the 
Irish easconn (pron. ascon or yascon), an eel. This monster 
was in fact a sea-serpent. An incident similar to that of 
Brendan is related in the Story of the First Voyage 
of Sindbad the Sailor. All this will remind the reader 
of the great Norse sea-animal called the Kraken : the 
monstrous fish mentioned in the First Book of " Paradise 
Lost " :— 

" Him, haply, slumbering on the Norway foam, 
The pilot of some small night-foundered skiff, 
Deeming some island, oft, as seamen tell, 
With fixed anchor in his scaly rind, 
Moors by his side under the lee, while night 
Invests the sea, and wished morn delays." 

But the Norse Kraken was a dull, listless sort of a