There is indeed a total absence of evidence to suggest that there were

large episcopal dioceses, and this is perfectly consistent with the

tradition that Patrick ordained an immense number of bishops, variously

stated, with great exaggeration, at 450 and 350.[432] Tírechán, who

gives the former number, cannot mention the names of more than about

forty-five. But the fact that he can record so many shows that the

number of bishops under the Patrician system was not small. It does not

follow that they had no dioceses; it only follows that the dioceses were

comparatively small. _A priori_, this is what we might expect. The number

of small tribal territories in Ireland could not fail to affect and

largely determine the ecclesiastical organisation. And when we find in

the records of Tírechán, concerning the foundations of St. Patrick, that

in some churches he placed presbyters, while for others he consecrated

bishops, we may be sure that in this discrimination he was guided by

considerations of secular geography, and that he did not wish to multiply

bishops beyond necessity.

The evidence of the records and traditions preserved by Tírechán has the

more weight because it is quite undesigned; and, so far as it goes, it

seems to bear out the general view of the Patrician organisation which

has been indicated above. Here are the bishops who were ordained by

Patrick for certain churches in Connaught:—

Assicus: Elphin (313 Rolls ed.)

Sachellus: Baslick (304, 320₁₄; cp. _Vit. Trip_, 110)

Bronus: Caisselire (327)

Mucneus: Donaghmore, near Fochlad (326)

[Muirethachus: Killala? (327)][433]

Senachus: Aghagower (322)

Cainnech: Cellola Tog (324)[434]

Felartus: Saeoli (313)

To these may perhaps be added the _epis̃ prespiter bonus_ (whose name is

illegible in the MS.: fol. 13 rᵒ b, l. 12) who founded a church in Imgoe

Mair Cerrigi (in the barony of Costello, Mayo).

As many foundations in Connaught are recorded by Tírechán, with which

no bishop is connected, it is clear that, so far as this evidence goes,

Patrick proceeded on some principle in the distribution of episcopal

churches. There is one passage from which it might be hastily concluded

that he set two bishops in one place. He installed Assicus and his

nephew Bitteus at Elphin, and both Assicus and Bitteus are described as

bishops (313). The true interpretation obviously is that, when Assicus

left Elphin (as described in 313₂₉, _fecit profugam ad montem lapidis_),

he was succeeded by Bitteus, as was natural; and that Bitteus is here

(313₂₂) described as a bishop _proleptically_. (This accords with the

part he took in the ordination of Cairell at a later time, in connexion

with which Assicus is not mentioned: 314₂₆, see below.)

There is one interesting piece of evidence for the multiplication of

bishops after Patrick’s death. Under his arrangement there was no bishop

at Tamnach in Tirerrill; but

id cairellum

post haec autem posuerunt episcopos iuxta sanctam aeclessiam hi

Tamnuch quos ordinauerunt episcopi Patricii id est Bronus et

Bietheus.

The language is a little loose, but clearly means that afterwards

Cairellus was ordained a bishop for Tamnach by Bronus of Caisselire and

Bitteus (who had succeeded his uncle Assicus) of Elphin, and that he was

succeeded by a line of bishops. Why was this innovation made? Clearly

because the community of Tamnach, founded by Mathona, had grown in

importance and aspired to have a bishop for itself. Here we may recognise

a distinct bit of evidence for the early introduction of the system,

which very soon became prevalent, of bishops without sees attached to

religious houses.

There seems then to be no evidence in support of the improbable

supposition that Patrick’s bishops had not episcopal districts.

Such districts would naturally have been determined by the tribal

divisions of the country. Nor is there any difficulty in explaining the

subsequent development of this episcopal church into a church which was

predominantly monastic. The double position of the bishop is the clew

to the development. For the bishop was not only bishop for a certain

diocese, which would contain several churches and religious foundations,

but he was also head of a particular community. Thus Assicus presided

over a religious community at Elphin,[435] besides being episcopally

responsible for a certain district. This system was a natural consequence

of the condition of the country; there were no cities; monastic

establishments were the substitute. Hence Patrick’s bishops were probably

in most cases monks; we have references to _monachi Patricii_,[436] from

whom Patrick doubtless selected some of his bishops, just as Lérins

supplied some notable bishops for the sees of Gaul. It is easy to

conceive how in process of time other religious communities might aspire

to be self-sufficient and have bishops of their own. If the bishop, whose

jurisdiction included a number of communities, had not been himself the

head of a similar community, such a tendency would not have been likely

to arise. But when there were a number of monastic communities, A, B,

C, etc., in one district, which was under the episcopal jurisdiction

of a bishop who was the _princeps_ of A, one can understand how this

jurisdiction might have soon seemed to B, C, etc., an intolerable claim

to superiority on the part of a neighbouring community. The bishop came

to be regarded, and perhaps came to act, less as the bishop of the

_paruchia_ than as the _princeps_ of his community. The double position

of the bishop exposed him to a jealousy which would not have been aroused

if he had stood outside all communities alike. The consequence was the

multiplication of bishops, many communities wishing to be independent and

to have bishops, each for itself. Thus the diocesan system partly broke

down, and the instance of the ordination of Cairell for Tamnach seems to

be an early instance.

Was there any discriminating designation to distinguish those religious

settlements which were seats of bishops from those which were not? I

venture on the conjecture that the name _civitas_ was originally applied

only to the former communities. In Gaul, in Italy, in Roman Britain,

the bishop’s seat was in a true _civitas_; and we can understand that

in a cityless land, _civitas_ might have been used in the special

ecclesiastical sense of the settlement in which the bishop lived.[437]

It is worth noticing that in the Memoir of Tírechán two episcopal seats

in Connaught seem to be distinguished as places for the ordination

of _bishops_: Aghagower _in quo fiunt episcopi_ (322₇) and perhaps

Donaghmore in Fochlad.[438] Did Patrick specially mark out certain

episcopal residences as places where bishops might be consecrated?

The outcome of this discussion is that Patrick’s organisation was from

one point of view monastic, from another episcopal. It was monastic in

so far as many of the churches in the various regions were connected

with religious communities of a monastic character, and the clergy were

largely monks. But this did not prevent it being episcopal, in the sense

that there were episcopal districts or dioceses. There was not a body of

bishops without sees, who went round visiting churches promiscuously, but

each bishop had his own diocese.

There is another consideration which has not been noticed, and which

seems to me of great weight. It is the position which bishops hold in the

few laws in the _Senchus Mór_ which touch the Church. The bishop appears

in these laws as the dignitary in the Church who corresponds to the king

in the State. There is no mention of presbyter abbots as sharing the

privileges of the bishop, who has the same “dire” fine as a king.[439]

Clearly a state of things is contemplated in which the bishops are the

important administrators in the Church. Such rights could not have been

established from the sixth century forward, when the bishops were of

less account. This feature of the _Senchus Mór_ seems to me, therefore,

to have a double significance, in establishing both the antiquity of

the code itself and the eminence of the episcopal office in the fifth

century; the two things hang together and mutually support each other.

A corroboration is perhaps also supplied by the _De Abusionibus Seculi_,

a treatise of Irish origin, written before 700 A.D. (see App. A, i.

4), where the tenth Abuse is the _Episcopus negligens_, in which the

pastoral duties of the bishop are insisted upon, and his functions as a

_speculator_. It seems by no means improbable that the work is older than

600 A.D.

19. _The Place of Patrick’s Burial_

The decisive reasons for determining the vexed question of Patrick’s

burial-place in favour of Saul have been given in the text; but

something more may be said here in criticism of the evidence. Among

the miscellaneous notices appended to Tírechán’s Memoir (p. 332) by

the scribe Ferdomnach, occurs a comparison of Patrick to Moses in four

points, of which the fourth is: _ubi sunt ossa eius nemo nouit_. The

author of this similitude has simply taken advantage, for his purpose,

of the fact that there were two rival claimants, Saul and Downpatrick.

It does not imply that there was no distinct and universally accepted

tradition placing his burial in this district; for if there had been any

room for doubt about that, Armagh could not have failed to claim his

bones.

Another notice follows, professing to tell how the place of sepulture was

revealed:—

Colombcille spiritu sancto instigante ostendit sepulturam

Patricii, ubi est confirmat., id est hi Sabul Patricii, id est

in aeclessia iuxta mare proxima, ubi est conductio martirum

id est ossuum Coluimbcillae de Britannia et conductio omnium

sanctorum Hiberniae iudicii.

Whatever value we may attach to this passage as a record of an opinion

of Columba, it is good evidence for the existence of a tradition that

Patrick was buried at Saul. Todd (_St. Patrick_, p. 494) thinks that

Downpatrick is intended, and thereby reconciles this passage (which he

wrongly ascribes to Tírechán) with the statement of Muirchu that he was

buried there.[440] If it were a case for reconciliation, it would be a

more defensible hypothesis that Muirchu used loose language, and said

that Patrick was buried at Dún, whereas he was really buried in its

neighbourhood. But the two statements should have been protected against

any such arbitrary attempt to reconcile them by the existence of the

third statement _ubi sunt ossa eius nemo nouit_, for they afford us its

explanation. Muirchu meant Dúnlethglasse, as he said; and the writer of

the notice of Columba’s discovery meant Saul, as _he_ said. There were

two rival traditions, the Saul tradition and the Dún tradition; and it

was just the existence of these two traditions which could enable a man

to say, “After all, we know not where the saint was buried; in that too

he was like Moses.”

The whole story, or collection of stories, related by Muirchu (and

reproduced in the text) contain obvious marks of their growth, and

internal inconsistency; though, as put together by Muirchu, they

represent the account accepted at Dúnlethglasse. The first part,

containing the interview with the angel, and the death and _exequiae_

(295₁₈₋297₂₅), does not contemplate Dúnlethglasse at all. It only

contemplates Saul: _revertere ad locum unde uenis, hoc est Sabul_.

It reflects, as I have said, a conciliation between the claim, or

rather disappointment, of Armagh, and the actual burial at Saul; it is

designed to protect, if I may say so, the countenance of Armagh, and

the compromise is reflected in the two petitions representing the two

interests.

After this comes the passage relating the burial at Dúnlethglasse

(298₁₋₂₀). It is quite evident that it has a distinct and subsequent

origin; it was manufactured in a different ecclesiastical workshop. It

adds, like a sort of postscript, a new piece, a new command, to the

discourse of the angel, in a new interest, namely of Downpatrick. If

the whole story had been of one piece, this command would have formed

part of the angel’s original address; it would not be introduced as an

appendix. This criticism is in itself sufficient to exhibit the falsity

of Downpatrick’s claim. It is to be noted that this claim had the further

purpose of establishing an early date for the origin of the church of

Downpatrick. It could not claim to have been founded by the saint; it

alleged that it was founded in connexion with his burial.

The two stories, which Muirchu relates of the contention between the

Ulidians and the men of Orior (_orientales_), supply an instructive

illustration of the genesis of legends. When we have two stories of

this kind, one is generally subsequent to the other, and suggested by

it.[441] The common argument of both is to show how hostilities were

prevented. There can be no doubt which was the genuine and primitive

story. The inundation of the sea, a motive (as I pointed out in the text)

characteristic of the district, furnishes a presumption in favour of the

priority of the first story. The second story obviously arose out of

the Dúnlethglasse legend of the oxen;[442] in fact, its point depends

upon that legend. Thus was myth added to myth in the workshops of Irish

ecclesiastics. The ecclesiastical origin is seen not only in the incident

of the burning bush, but in the invention of the cart and two unyoked

kine (1 Samuel vi. 7 _sqq._).

20. _Legendary Date of Patrick’s Death_

The true year of Patrick’s death is furnished by our oldest document,

Tírechán (302₂₉, a passione Christi ... anni ccccxxxiii [MS. ccccxxxui]);

Ann. of Ulster, _s.a._ 461 (cp. Ann. Inisf. _s.a._ 493); Nennius, _Hist.

Brit._ 16 (pp. 158-9 ed. Mommsen). See the criticism of this material in

Bury, _Tírechán’s Memoir_, pp. 239 _sqq._

But side by side with the true tradition, we find in the Annals another

date, A.D. 493, which has been generally received and is the vulgar

era of the event. It is closely connected with the legendary age of

Patrick—120 years, which is as old as Muirchu’s Life (296₁₉). The

question arises, how came his age to be raised from somewhat more than 70

to 120 years, and why was A.D. 493 fixed on as the date of his death?

In the analogy which was drawn between Patrick and Moses, one of the

items of resemblance is the same length of life. But the Mosaic motive

cannot have been sufficient to determine originally such a marked

perversion of fact. On the contrary, the age of 120 years must have been

otherwise suggested and have then, in its turn, contributed to suggest

the Mosaic analogy.

It is to be noted that in the _Liber Armachanus_ two divergent

subputations of Patrick’s age are found. One of these was originally

appended to the exemplar of Muirchu which was used by the Armagh scribe;

the other is among the notes which are appended to his transcription of

Tírechán.

In the first (300₂₁) it is stated that Patrick, having been captured in

his 20th year, was a slave for 15 years, studied for 40, and taught for

61. Hence it is inferred that his age was 111. There must be errors in

the figures.

In the second (331₂₂) we have the following statement:—baptized in 7th

year, captured in 10th, was a slave for 7 years, studied for 30 years,

and taught for 72. Hence his age at his death is inferred as 120 (10 + 7

+ 30 + 72 = 119), _ut Moyses_.

It is, I venture to think, of the greatest significance that in both

these cases the fictitious age is given as a total, and preceded by a

statement of the items which compose it. This fact furnishes the clue.

The Mosaic age, 120, was not handed down as a legend, nor was the true

age of somewhat over 70 audaciously raised to the Mosaic figure for

the purpose of the Mosaic comparison. The figure 120, or something

approximate, was reached by means of a computation of chronological

items, and it is these chronological items that we must examine in order

to discover the origin of the error.

In the two anonymous computations which I have quoted the items

disagree with each other, and also disagree with the data in Patrick’s

_Confession_. They cannot therefore form the basis of our examination,

because we may safely assume that the computation, by which 120 years

(exactly or approximately) was reached and was generally accepted in

Ireland, did not contravene the data of the _Confession_—data which

were rightly quoted in Tírechán (302₁₇) from the Book of Ultan. We may

safely assume that the computators, who succeeded in establishing the

Mosaic age, started with the incontrovertible fact supplied by the

_Confession_ that Patrick was either 22 years old, or in his 22nd year,

when he escaped from captivity. It is equally clear that the period which

they assigned to his teaching was 60 (rather 61 years)—for that is the

period from his arrival in Ireland, A.D. 433 (rather 432) to the alleged

date of his death, A.D. 493. This would leave 38 (rather 37) years to

be accounted for. Now Muirchu mentions two different records, 30 and 40

years, for the sojourn with Germanus (it is to be observed that, of the

two computations noticed above, one reckons with 30, the other with 40

years). If either of these, 30 years must have been the period accepted

in this computation for St. Patrick’s studies, and an interval of 8 (or

7) years left, which agrees remarkably with the notice of Tírechán (from

the _Liber apud Ultanum_, _ib._): uii. _aliis annis ambulauit_, etc. The

sum worked out thus—

22 + 7 + 30 + 61 = 120.

Of these items 22 had the best authority, and 30 and 7 were independent

records or traditions (as we know from Muirchu and Tírechán). The problem

is therefore reduced to discovering the origin of the 61 or 60 years

during which the apostle is wrongly supposed to have taught. That it was

due to a misinterpretation and not to a deliberate invention, there can,

I think, be no question.

A list of the Armagh succession in the Book of Leinster (printed in

Stokes, _Trip._ pp. 542 _sqq._) supplies a hint which suggests an

explanation of the error. The first entry is—

Patrick: 58 years _from the coming of Patrick to Ireland_ till

his death.

The fact that a period of 58 years is given here instead of the usual

number (493-433/2) is remarkable. Now if we count back 58 years from

the _true_ date of Patrick’s death, 461, we reach the year 403-404,

which was in the close vicinity of the year in which, as we saw, he was

probably taken into slavery. Hence it seems probable that in an early

record it was stated correctly that 58 years elapsed between the coming

of Patrick to Ireland, meaning his _first_ coming as a slave, and his

death, and that this notice was misinterpreted by subsequent computators

who referred it to his _second_ coming as a teacher. They computed: 22

+ 7 (_ambulauit_) + 30 (_legit_) + 58 (_docuit_) = 117. An age so close

to 120 could hardly fail to suggest Moses; the figures clamoured for

a slight manipulation, and the means adopted was to increase the last

period from 58 to 61. Thus the date of the death was determined: 432 + 61

= 493.

By this computation the coming to Ireland divided the whole life into two

almost equal parts (59 + 61), which would naturally come to be described

in round numbers as each 60 years. This probably led to the alternative

date for the obit given in the _Ann. Ult._ 492 (= 432 + 60).

The computation which established 120 years for the age and A.D. 493 for

the year of the death, thus involving 61 years for the Irish period, was

triumphant and became authoritative. But the computations quoted above

from the _Liber Armachanus_ show that other theories had been propounded

based on the addition of items. In the _Vita Tertia_ the age is given as

132. This figure seems to have been obtained by substituting 72 instead

of the (round) 60 for the last item in the sum; this is suggested by the

fact that in the second computation in the _Liber Armachanus_ the last

item is _septuaginta duo annos docuit_, though the total 120 (119) is

secured by erroneous dates for the captivity. What is the origin of this

number 72? It must have come down in some form, it must have represented

something, when, in order to do justice to it, one computator felt

compelled to raise the Mosaic 120 to 132, and another, though holding

fast to 120, modified the authentic dates of the captivity.

I suggest that this number 72 represents an old and correct record of

Patrick’s age at the time of his death.

21. _Professor Zimmer’s Theory_

The general obscurity which surrounds the early history of Ireland,

the difficulties which have been found in making out the period of

Patrick’s career from indefinite and contradictory data, the fact that

while his death fell on any theory in the fifth century, no mention of

him in literature is found before the seventh, these circumstances have

led to a variety of theories which beset and embarrass the student who

approaches Patrician literature. Patrick has been in turn eliminated and

reduplicated; and, in revenge for undue magnification, his rôle has been

reduced to something quite insignificant. We may say that his writings,

like the poems of Homer, have been taken away from him to be ascribed to

some one else of the same name. It is needless to notice here theories

which are fantastic or baseless, and have never gained any general or

wide acceptance, but the view which has been recently developed by the

brilliant Celtic philologist, Professor Zimmer,[443] cannot be passed

over without criticism.

In another note I have referred to Zimmer’s identification of Patrick

with Palladius. This is necessitated by, but would not necessitate, his

theory. If it were demonstrated to-morrow that Patrick and Palladius were

one and the same person, this would be quite as compatible with the view

of Patrick adopted in the foregoing pages as with the view of Zimmer.

Palladius may provisionally be left out of account for the purpose of the

present criticism, though I shall have a few words to say on the subject

at the end of this excursus.

Zimmer fully admits, though he had once denied, the genuineness of the

_Confession_ and the missive to the subjects of Coroticus; that is, he

admits that they were written by Patricius, a bishop in Ireland in the

fifth century. But he holds that the activity of this bishop was entirely

confined to south-eastern Ireland (Laigin), that he accomplished nothing

for the evangelisation or ecclesiastical organisation of the rest of

Ireland, that he died (A.D. 459: Zimmer) conscious of failure, and for

nearly two centuries after his death had a merely local reputation.

The rejection of the traditional Patrick is, so far as I understand,

based chiefly on two arguments: (1) Zimmer’s interpretation of the

_Confession_; and (2), if Patrick’s work had at all corresponded in

scope, magnitude, and import to the descriptions of it given by Irish

writers of the seventh century, it would have been noticed by Bede in

his _Ecclesiastical History_. As for the _Confession_ I have said enough

in the text, and have shown, I think, that its note is not, as Zimmer

holds, consciousness of failure. As for the argument from the silence

of Bede,[444] on which he lays much stress, I may quote what I said in

reviewing his book: The value of arguments from silence “depends entirely

on the cases; in some cases an argument from silence is conclusive.

But can it be said to weigh much here, if we reflect that a notice of

Patrick and his work in Bede’s book could have been simply a defensible

digression? We can place our finger on the unproven premiss in Zimmer’s

argument; he speaks of ‘Bede’s evidently keen interest in the early

beginnings of Christianity in the British isles’ (p. 11). Substitute

‘Britain’ for ‘the British isles’ and the cogency of the argument

disappears. Ninian and Columba are immediately relevant to his subject,

Patrick is not; and, assuming the common tradition of Patrick’s work (as

believed in Ireland _c._ A.D. 700) to be roughly true, it would be no

more surprising to find nothing about it in Bede than it would be to find

no mention of Augustine in an ecclesiastical history of Germany written

on the same lines as Bede’s.”—_English Hist. Review_, July 1903.[445]

Zimmer’s reason for restricting the sphere of work of his Patricius to

Laigin, or part of Laigin, seems to be the circumstance that the author

of the earliest biography, Muirchu, belonged to this part of Ireland,

being connected with Slébte. [It is possible, though he does not allege

it, that another reason may have been the tradition which associates

Palladius (whom he equates with Patrick) with the territory corresponding

to the county of Wicklow.] The suggestion is that in this neighbourhood,

at Slébte, for instance, were preserved traditions and writings of the

obscure Patricius, who was in the seventh century to be transformed into

the great apostle of Ireland. It is obvious that the argument, even if

it were based on a correct statement of facts, is quite insufficient to

prove the thesis.

But the statement of facts is not correct. There is another document

which, though also dating from the second half of the seventh century,

may claim (see above, p. 248) to have some slight advantage in point of

priority over the _Life_ of Muirchu. This is the _Memoir_ of Tírechán. He

had nothing whatever to do with Laigin; he was connected with Connaught

and Meath. His spiritual master, Ultan, was a bishop at Ardbraccan, and

had in his possession a book concerning the life of Patrick. We might

therefore, if we adopted Professor Zimmer’s method of argument, conclude

that the sphere of the true Patrick’s activity was confined to a part of

the kingdom of Meath. The mere existence of Tírechán’s work forbids us to

attach to the provenance of Muirchu’s work any significance of the kind

which Zimmer attributes to it.

It must also be observed in what a strange light Zimmer’s theory would

place the work of Muirchu. We should have to suppose that this writer,

taking upon himself to be Patrick’s biographer because he belonged to

the province in which Patrick had worked, ignores entirely Patrick’s

connexion with that province (merely mentioning that he landed on the

coast of Laigin), and devotes all his space to legendary adventures in

other parts of Ireland.[446] As a matter of fact, a critical analysis of

his work shows that a large amount of material depends on ancient local

traditions of the Island-Plain in Ulidia.

This brings us face to face with the great difficulty which Zimmer has to

meet. How, and why, and when was the obscure Patricius of fact transmuted

into the illustrious Patricius of tradition? Zimmer’s answer to the

question, when? has at least the merit of precision. He finds the motive

of the glorification of Patrick in the Roman controversy of the seventh

century, and he dates the appearance of the legend about A.D. 625. His

own words must be quoted (_Celtic Church_, p. 80):—

It would not require a long stretch of imagination if we assume

that, about 625, Ireland’s pious wish of having an apostle of

her own was realised by reviving the memory of this Patricius,

who had been forgotten everywhere except in the south-east.

It was in this way, I think, that the Patrick legend sprang

up with its two chief premisses: first, that Ireland was

entirely pagan in 432, as the lands of the Picts and of the

Saxons had been in 563 and 597 respectively; and secondly, that

Patrick converted Ireland within a short time and introduced

a Christian Church, overcoming all obstacles and winning the

favour of King Loigaire, incidents analogous to Columba’s

conversion of King Brude, or Augustine’s of Ethelbert of Kent.

Pointing out that the first mention of Patricius is in connexion with a

Paschal cycle, in Cummian’s letter to Segéne, Zimmer proceeds:—

Thus the Patrick legend is characterised on its first

appearance as serving the endeavours of the _Southern Irish_ to

enter into the _unitas Catholica_ by yielding to Rome on the

Easter question.

It is to be observed that Zimmer does not categorically allege that the

Patrick legend was invented by the Roman party in south Ireland. He

leaves this open. He says: “If this legend was not expressly invented

by an Irish member of the party in favour of conformity, it was, at

any rate, utilised at once by that party.” Let us take in turn the two

alternatives, (1) that it was invented expressly by the Roman party, (2)

that it was otherwise invented, but precisely in time for them to utilise.

(1) This alternative implies that the legend was manufactured in

south-eastern Ireland (the only place where, _ex hypothesi_, Patricius

was remembered) for the purpose of bringing about the conformity

of the northern Irish. But the legend itself, as developed in the

seventh-century sources, as developed in Muirchu’s work, on which

Zimmer lays great stress, repudiates this origin. Southern Ireland,

south-eastern Ireland, do not appear in the legend at all (except the

single reference to Fíacc and the mention of the landing in Wicklow); and

it would require the strongest direct evidence to prove that, in spite

of this radical fact, the legendary story originated there. The theory

implies that the south-eastern Irish, while they resuscitated Patricius,

made him over entirely, and transferred all their rights in him, to the

north of the island—severed him entirely from themselves. So far as the

single circumstance of the conversion of Loigaire and the incidents of

the first Easter is concerned, that might pass; for in Loigaire, as High

King, the south as well as the north of Ireland had part. But the point

is that the whole setting connects Patrick with the north and not with

the south. Zimmer’s hesitation shows, I suspect, that he was to some

extent conscious of this difficulty; and so he leaves open

(2) The other possibility that the rise of the Patrick legend may have

had another origin. The only motive he suggests is “Ireland’s pious

wish of having an apostle of her own.” Now if the origin of the Patrick

legend was independent of the Roman controversy, why need it be placed

in A.D. 625? How are we to determine its date? The answer, which seems

to be implied by Zimmer, is that it must have been subsequent to the

times of St. Columba and St. Augustine, the object being to set up a

primitive apostle, to be for Ireland what Columba and Augustine were for

Pictland and for England. This is, of course, the purest speculation;

but setting aside the question of date, we should have to suppose—if we

are not to fall into the same difficulty as in the case of the first

alternative—that the idea arose in north Ireland, and that the legend

was invented there. This would imply that the northern ecclesiastical

mytho-poets had recourse to south-eastern Ireland to find an obscure

ecclesiastic to glorify; that they detached him ruthlessly from his

home and appropriated him entirely. We need not, however, consider this

improbability, as Zimmer himself does not contemplate it. His view is

that Patrick was resuscitated “in the district of his special activity,”

“with the help of his own writings and of documents about him.”

It would seem far more natural that if the Irish were in search of a

founder, they should seize on Palladius, whose mission was recorded by

Prosper. Zimmer therefore identifies Palladius with Patricius; but the

Irish had no idea of such an identity.[447] For in the oldest _Life_,

Patricius and Palladius are distinguished. If, as Zimmer thinks, the

creation of an apostle was due to a “specific tendency,” namely,

approximation to Rome, it would be strange if Palladius, for whose direct

mission from Rome there existed the record of Prosper, were not chosen.

The legend of the conversion, the story of the first Paschal celebration,

could as easily have been spun round him as round the author of the

_Confession_. Zimmer avoids the difficulty by identifying them.

For his combination of the rise of the Patrick legend with the Paschal

question, Zimmer lays much stress on Muirchu. I have pointed out above

(Appendix A, ii. 3) that a certain indirect connexion between Muirchu’s

work and events connected with the Roman controversy may be fairly

inferred. But this inference does not furnish any support for Zimmer’s

daring theory. And it is important to notice in this connexion that

Muirchu did not believe that Patrick went to Rome; he admits that he

wished to do so, but denies that he passed beyond Gaul. This in itself

would make us hesitate to believe that the story of Patrick, as expounded

by Muirchu, was a recent fabrication in the interests of the Roman cause.

But we may waive all particular criticisms of Zimmer’s reconstruction,

and state the general and decisive objection to his or any similar

theory. It is this. The nature of the traditions which are preserved in

the two seventh-century compilations written by Tírechán and Muirchu

forbids the hypothesis of recent fabrication. In the first place a

critical examination of the texts of these works enables us to conclude

that they were largely based on older _written_ material. In the second

place, it is perfectly inconceivable that all the detailed traditions

which Tírechán collected both from written and from oral resources

concerning Patrick’s work in Connaught should have been deliberately

invented, between 625 and 660, in a region where Patrick’s name was never

known. In the third place, the really characteristic Patrician stories,

the death of Miliucc, the events of the first Easter, the story of

Daire, are not of the kind which are fabricated, generations after the

life of the hero, for a deliberate purpose. They belong to the legends

that spring up soon after the death of their hero, or even during his

lifetime. I may refer to what I have said in the text (p. 111). If we had

no other evidence, the tale of the first Easter at Slane and Tara would

be in itself a guarantee that the “Patrick legend” could not have been

deliberately invented in the seventh century.

One more observation. It would be difficult to explain how it came that,

if the author of the _Confession_ spent his life in Leinster, and his

name was sufficiently well remembered to make his fortune in the seventh

century, no particular church in that region claimed him. Zimmer may

get out of the difficulty through his identification of Palladius with

Patrick; he may say that Patrick’s church was the Palladian Cell Fine,

where memorials of Palladius were preserved. But this explanation would

only serve to emphasise the improbability of the theory of identity.

It is impossible to understand how Cell Fine remembered its founder as

Palladius and not as Patricius, seeing that (_ex hypoth._) the name

Patricius (as used by the author of the _Confession_) had so completely

superseded the name Palladius, that the bishop was not only glorified

as Patricius in the seventh century, but even distinguished definitely

from Palladius. Again: either it was remembered in south Ireland or it

was not remembered at the beginning of that century, when the Patrician

legend is alleged to have taken shape, that Palladius and Patricius

were the same person. If it _was_ remembered, then how came it that

Palladius-Patricius was differentiated into two, when it was assuredly

more in the interests of the Roman cause to glorify an apostle sent

by Celestine than to discriminate a successful missionary who was not

sent by a Pope, from an unsuccessful missionary who was? If it _was

not_ remembered, then the only passage which Zimmer can quote for the

identification (_Lib. Arm._ p. 332, see above, p. 389, note) can be at

once eliminated from the discussion, as resting on mere conjecture.

[The argument, which Zimmer adduces for his theory, from the statement

that Patrick’s burial-place was unknown, falls to the ground when the

evidence in regard to his burial-place is criticised as a whole. See

above, Appendix C, 19.]

FOOTNOTES

[1] I may be permitted to remark that in vindicating the claims of

history to be regarded as a science or _Wissenchaft_, I never meant to

suggest a proposition so indefensible as that the presentation of the

results of historical research is not an art, requiring the tact and

skill in selection and arrangement which belong to the literary faculty.

The friendly criticisms of Mr. John Morley in the _Nineteenth Century and

After_, October 1904, and of Mr. S. H. Butcher in _Harvard Lectures on

Greek Subjects_ (1904), Lecture VI., show me that I did not sufficiently

guard against this misapprehension.

[2] For the expansion of Christianity in the first three centuries see

Harnack’s invaluable work _Die Mission und Ausbreitung des Christentums

in den ersten drei Jahrhunderten_ (1902).

[3] Rufinus, _Hist. ecc._ ii. 7. For the Georgian legend of Nino see

_Life of St. Nino_, translated by Marjory and J. O. Wardrop, in Oxford

_Studia Biblica et Ecclesiastica_, vol. v. (1900).

[4] _De Vocatione Gentium_, ii. 32.

[5] On the other hand it may be questioned whether the army itself did

anything to diffuse Christianity within the Empire. In the west certainly

its chief significance in the history of religion was what it did to

spread the solar, Mithraic worship. Cp. Harnack, _op. cit._ 268, 388.

[6] Rufinus, _Hist. ecc._ i. 9.

[7] Armenia was already Christian at the beginning of the fourth century

in the days of Maximin.—Eusebius, _Hist. ecc._ ix. 8. 2.

[8] Socrates, _Hist. ecc._ vii. 30.

[9] The island of Man is indeed another exception. The Scottic

colonisation of north-western Britain (Argyle, etc.) was comparatively

late, but before the middle of the fifth century (see below, chap. ix. p.

192).

[10] Professor Rhŷs thinks that it was to Ireland, more than to Britain,

that the Gallic Druids went to learn their art, and that Caesar (in _B.

G._ vi. 13) was badly informed; and he has recently stated this view in

_Studies in Early Irish History_ (_Proceedings of British Academy_, vol.

i.), p. 35. It is remarkable that, apart from Caesar’s assertion, the

only evidence for Druidism in southern Britain pertains to the island of

Anglesey (Tacitus, _Ann._ xiv. 29), and Professor Rhŷs holds that in the

first century A.D. Anglesey (Mona) was not yet Brythonic. Druidism in

the Isle of Man is attested by a stone inscribed _Dovaidona maqi Droata_

“(the burial-place) of Dovaido, son of (the) Druid.” See Professor Rhŷs

in the _Academy_ for August 15, 1890.

[11] Tacitus, _Agricola_, c. 24, medio inter Britanniam atque Hispaniam

sita. Cp. Caesar, _B.G._ v. 13. The notice in Orosius (_Hist._ i. 2, §

72) of the lighthouse at Brigantia in north-western Spain as built _ad

speculam Britanniae_ is noteworthy. Compare the remarks of Professor

Rhŷs, _op. cit._ p. 47.

[12] Tacitus, _ib._ aditus portusque per commercia et negotiatores

cogniti.

[13] Tacitus, _ib._ The policy recommended by Agricola, who considered

one legion sufficient to hold the island, was based partly on the ground

of political expedience. The conquest of Ireland, he thought, would have

a similar wholesome effect on Britain to that which the conquest of

Britain had on Gaul, by removing the spectacle of liberty (_si Romance

ubique arma et uelut e conspectu libertas tolleretur_).

[14] The baronies of Upper and Lower Deece, in Co. Meath.

[15] Decies within Drum, and Decies without Drum, in Co. Waterford.

[16] See note, Appendix B.

[17] _Ib._

[18] It seems probable that Pelagius sprang from these Gaelic settlers in

Britain. See below, p. 43.

[19] See below, cap. viii. _ad fin._

[20] Un peuple n’emprunte pas l’alphabet des voisins s’il n’a pas à

correspondre avec eux.... Qui donc constate un emprunt de monnaie et

d’alphabet, en tous temps et en tous lieux, peut affirmer un échange de

produits et d’idées (V. Bérard, _Les Phéniciens et l’Odyssée_, i. p. 20).

[21] See Appendix C, 1.

[22] See note, Appendix B.

[23] See Appendix C, 2, for the following account of the invasions of

Britain.

[24] _Circa_ A.D. 389; see Appendix C, 3.

[25] See note, Appendix B.

[26] There is no evidence, and no probability, that the name Paul was

adopted on his conversion, or that it had anything to do with Sergius

Paullus.

[27] Frigeridus is Gothic Frigairêths.

[28] For date see Appendix C, 3.

[29] See Appendix C, 4.

[30] Croagh Patrick, close to Westport.

[31] Dalriada = north Antrim; Dalaradia = south Antrim and Down. The

Latin form, Ulidia, is used in this book for Ulaid in the narrower

meaning.

[32] It has been conjectured that Miliucc’s dwelling was on the hill of

Skerry, on the northern side of the Braid; see below, p. 86.

[33] Another possible theory is mentioned in Appendix C, 4.

[34] See note, Appendix B.

[35] The association of Saint Elias with the sun was due to the

resemblance of the name to the Greek ἥλιος.

[36] See Appendix C, 6.

[37] St. Honorat.—Lero is Ste. Marguerite.

[38] A.D. 426.

[39] A.D. 433.

[40] His own expression “as a son” shows that _parentes_ here means

kinsfolk, not parents, and justifies the inference that his parents were

dead.

[41] Pelagius, _Letter to Demetrias_, Migne, _P.L._, xxxii. 1100.

[42] Prosper has an epigram on the thesis that the whole life of

non-Christians is sin:

Perque omnes calles errat sapientia mundi

Et tenebris addit quae sine luce gerit.

(_Epig._ 83, ed. Migne, 51, p. 524.)

[43] Compare Celestine, _Ep._ iv. (Migne, _P.L._, 1. 434), nullus inuitis

detur episcopus.

[44] The conjecture is due to Professor Zimmer.

[45] The old kingdom of Leinster, or Laigin, was south of the Liffey,

and in this book “Leinster” is used in this sense (not equivalent to the

modern province, which includes the old kingdom of Meath). See below,

chap. iv.

[46] See Appendix C, 9, on Patrick’s consecration.

[47] No better illustration of this can be found than Pope Gregory’s

provision for the mission of Augustine to England, as recorded in Bede,

_Hist. ecc._ i. 29; he sent, besides fellow-workers, “uniuersa quae ad

cultum erant ac ministerium ecclesiae necessaria, uasa videlicet sacra,

et vestimenta altarium, ornamenta quoque ecclesiarum, et sacerdotalia uel

clericilia indumenta, sanctorum etiam apostolorum ac martyrum reliquias,

necnon et codices plurimos.”

[48] It has recently been held, more plausibly but erroneously, that

Patrick was on his way to Rome when the news of the death of Palladius

overtook him. See Appendix C, 8.

[49] Celestine probably died July 27, and Xystus succeeded July 31, 432.

These dates have been determined by M. Duchesne, _Liber Pontificalis_, i.

pp. ccli.-ii.

[50] It is probable that excommunication by a Roman bishop was also

recognised as universally binding. The question whether the popes had

the right of annulling sentences pronounced by provincial councils on

bishops, depends on the question of the authenticity of the Council of

Sardica. See J. Friedrich, _Sitzungsber._ of the Bavarian Academy, 1901,

417 _sqq._; E. Babut, _Le concile de Turin_, 75.

[51] A.D. 366-384.

[52] See Babut, _Le concile de Turin_ (1904), a valuable work.

[53] This has been well brought out by M. Babut.

[54] _Novella_, xvi.

[55] _Commonitorium_, ii. 33, 34.

[56] The chief source for the social and economic conditions of ancient

Ireland is the collection of the _Ancient Laws of Ireland_ (6 vols.,

1865-1901). A clear account of the general framework of society, with

interesting details and illustrations, will be found in Dr. Joyce’s

_Social History of Ireland_, vol. i.

[57] _Tuath_ = people, tribe, tribal district.

[58] _Flaith_ = noble.

[59] The _bó-aires_.

[60] The tributes and presents which are due from the under-kings to the

over-kings, the donations which the over-kings owe to the under-kings,

the privileges which the various kings possess, are the subject of the

_Book of Rights_ (edited and translated by O’Donovan, 1847), which still

awaits a critical investigation. It is easy to see that it was compiled

in Munster in the tenth century, but it was based on older material of

high antiquity, and clearly reproduces the general character of the

mutual relations which theoretically bound together the Irish kingdoms.

[61] The king of Aileach was so called because his palace was at Aileach,

near Londonderry. His territory was north Ulster to the Bann. Ulaid was

east Ulster; Oriel, south Ulster.

[62] This is clearly to be inferred from the _Book of Rights_, where

no relations or mutual obligations are mentioned as existing between

the three Ulster kings. Nor was there, since the destruction of the old

Ulidian kingdom in the third century, any name to designate the whole

province, for Ulaid was confined to the kingdom in the east of Ulster.

The use of Ultonia to describe the province, as distinguished from Ulidia

= Ulaid, is of course merely a literary convention.

[63] See Petrie, _Tara Hill_, 135.

[64] His date, according to the Annals, was A.D. 358-366; Niall reigned

A.D. 379-405; his nephew, Dathi, 405-428; and then his son, Loigaire,

428-463. For Amolngaid (Dathi’s brother), king of Connaught, see Appendix

C, 14.

[65] Tyrconnell.

[66] The derivation of the word _druid_ (nom. _drui_, gen. _druad_)

is uncertain. Perhaps, as Professor Rhŷs holds, Druidism was not of

Celtic origin, and the word “was adopted by the Celts from some earlier

population conquered by them” (see his “Studies in Early Irish History,”

in _Proceedings of the British Academy_, vol. i. p. 8). _Druidecht_ is

the Irish for magic. For the functions and powers of the Druids some

excellent pages in Dr. Joyce’s _Social History_, I. c. ix., may be

recommended; illustrations and references will be found there.

[67] Mug Ruith, servant of the wheel, was the name of a mythical Druid.

[68] The _Feth Fiada_.

[69] For these superstitious ceremonies at baptism cp. Duchesne,

_Origines du culte chrétien_, pp. 296-7 (the exorcism of salt), 299, 317;

cp. 349.

[70] In the remarkable ancient Irish Christian incantation, the Lorica,

ascribed to St. Patrick (see Appendix A, 5), the Trinity, Angels,

Prophets, and other Christian powers are invoked, but also “might of

heaven, brightness of sun, brilliance of moon, splendour of fire, speed

of light, swiftness of wind, depth of sea, stability of earth, firmness

of rock,” to intervene between him who repeats the spell when he arises

in the morning and “every fierce merciless force that may come upon my

body and soul; against incantations of false prophets, against black laws

of paganism, against false laws of heresy, against deceit of idolatry,

_against spells of women and smiths and druids_, against all knowledge

that is forbidden [so Atkinson] the human soul.”

[71] M. Réville, dealing with the third century, puts this very well.

“Chacun croit sans le moindre difficulté à toutes les merveilles et à

toutes les folies. On dirait même que plus une pratique est merveilleuse,

plus elle a de chance d’être admise sans contestation. Chose singulière!

les adeptes des religions opposées ne contestent pas la réalité des

miracles allégués par leurs adversaires: Celse admet les miracles des

chrétiens, et ceux-ci ne se refusent pas à admettre les miracles païens;

des deux parts on attribue aux mauvais esprits les merveilles invoquées

par les adversaires” (_La Religion à Rome sous les Sévères_, p. 131).

[72] See note, Appendix B.

[73] The _Memoir_ by Tírechán.

[74] The _Life_ by Muirchu.

[75] Cuchullin of legend.

[76] Then called Brene Strait.

[77] The Slaney (see Appendix B, note). It flows from L. Money past

Raholp.

[78] It has been conjectured that the stronghold of Miliucc was on the

hill of Skerry, north of Slemish, on the other side of the Braid valley.

Muirchu says that Patrick saw the conflagration from the south side of

Slemish. We may interpret south to mean south-west. A cross, mentioned

by Muirchu, was erected on the spot where the legend supposed Patrick to

have stood, and the memory of this is still preserved in the name of the

townland of Cross, on a hill to the west of Slemish.

[79] Dr. W. Stokes, taking the story literally, suggests that Miliucc

committed self-destruction as “a mode of vengeance” (_Book of Lismore_,

p. 295).

[80] _Mag-inis_, later known as Lecale (_Leath Cathail_), now the

baronies of Lower and Upper Lecale. It is accurately described as a

peninsular plain.

[81] But meaning _barn_.

[82] There is a second story (also recorded by Muirchu), clearly inspired

by the same motive. Patrick was resting near Druimbo (in the north of

Mag Inis, and close to a salt marsh), and he heard the noise of pagans

who were busily engaged in making an earthwork. It was Sunday and he

commanded them to cease from work. When they refused he cursed them:

“Mudebrod! may your work not profit you!” and the sea rushed in, as in

the other story, and the work was destroyed. The curse _mudebrod_ (or

_mudebroth_) has not been explained.

[83] See above, chap. i. p. 14.

[84] See Appendix C, 11.

[85] See above, cap. iv. p. 72.

[86] Compare what has been said above in chap. i. p. 9.

[87] This name is the same as the British Vortigern (Welsh Gwrtheyrn),

and the original Goidelic form was similar. It occurs in Ogam

inscriptions, thus: ... Maqi Vorrtigern<i>, on a stone of Ballyhank (near

Cork), now in the Dublin Museum (Rhŷs, _Proc. of R.S.A.I._, pt. i. vol.

xxxii. p. 9, 1902).

[88] The distance of Tara from Slane is about ten miles.

[89] Yet more remote from the Paschal season was the feast of Samhain at

the close of autumn (November 1), when on the hill of Tlachtga, not far

from Trim, a fire was kindled, from which, tradition says, all hearths in

Ireland were lit. It was at Samhain too, according to tradition, that the

High Kings used to hold such high festivals at Tara as are designated in

the story. See note, Appendix B.

[90] This incident is obviously suggested by St. John xx. 19, 26. When

St. Columba went to the palace of King Brude the closed gates opened of

their own accord (Adamnan, _V. Col._ ii. 35).

[91] See note, Appendix B.

[92] See note, Appendix B.

[93] Tírechán, p. 308, perrexitque ad civitatem Temro ad Loigairium

filium Neill iterum quia apud illum foedus pepigit ut non occideretur in

regno illius.

[94] See Appendix C, 12.

[95] At Crag, in Co. Kerry.—Macalister, _Studies in Irish Epigraphy_, ii.

p. 52.

[96] Cp. Appendix A, ii. 5.

[97] It should be Donagh-shaughlin, for Donagh is _domnach_, a church,

whereas _dún_ is a fort. There is no doubt that Dún here is a corruption,

as we get the form Donnaclsacheling in a document of A.D. 1216 (Reeves,

_Eccl. Ant._ p. 128).

[98] See Appendix B, note on cap. ii. p. 23.

[99] See Appendix A, i. 6. One of the best quatrains is the fourth:

Dominus illum elegit ut doceret barbaras

nationes, ut piscaret per doctrinae retia,

ut de seculo credentes traheret ad gratiam

Dominumque sequerentur sedem ad etheriam.

[100] The _Confession_ shows that this comparison was sometimes in

Patrick’s mind.

[101] Perfectam vitam, _Hymn_ v. 4. Secundinus died A.D. 447, acc. to

_Ann. Ult._

[102] Telltown comes by popular etymology from the genitive Taillteann.

The site is marked by a round rath. O’Donovan said in 1856 that it

had been in recent times a resort for the men of Meath for hurling,

wrestling, and other sports (_Four Masters_, i. p. 22).

[103] Herbord, _Vit. Ott._ ii. 14. The silence of early authorities is

decisive against the isolated statement that Patrick preached at Taillte

against the “burning of the first-born offspring.” (See Appendix B, note.)

[104] Mac Fechach.—Tírechán, 310₂₄.

[105] Áth Brón.—Tír. 307₂₈.

[106] St. Colomb’s House. For its description and measurements see Petrie

(_Round Towers_, 430-31), who compares it with St. Kevin’s House at

Glendalough, and Dunraven’s _Notes on Irish Architecture_, vol. ii. p. 50

(plans and photograph).

[107] Tírechán, 310-11.

[108] See Appendix C, 4, _ad fin._

[109] See note, Appendix B.

[110] It may be observed that if the idol of Mag Slecht had been

eminently important for all Ireland, and had been destroyed at a period

subsequent to St. Patrick, there could hardly fail to be a Christian

record of its fall. In the _Annals of the Four Masters_, s.a. 464, it

is said that Conall, son of Niall, ancestor of the lords of Tyrconnell,

was done to death by the “old Folks” of Mag Slecht, who caught him

unprotected. The thought occurs that Conall had supported the attack on

the worship of Cenn Cruaich, and that his death was an act of vengeance

wreaked by people of the plain who still clung to the old faith.

[111] Perhaps A.D. 444-5. See Appendix C, 14.

[112] See Appendix C, 13, on Patrick in Connaught.

[113] At Duma Graid, close to Lake Kilglass. See Tírechán, 313, and _Vit.

Trip._ p. 94.

[114] Between Sligo and Leitrim.

[115] May the name be the same as that of the tribe of the Anghaile

(Annaly), who extended their power subsequently into Tethbia (cp.

O’Donovan, _Book of Rights_, p. 11, _note_)?

[116] Tamnach.

[117] “Church of Bishop Brón.”

[118] In Mag Airthic. See Appendix C, 13.

[119] Ciarrigi. Through the baronies of Costello, Clanmorris, and

Kilmaine. Possibly Aghamore, south of Kilkelly, may lie on the supposed

route. It has been conjectured that the church _in campo Nairniu_

(Tírechán, 321) was there.

[120] In quo fiunt episcopi.

[121] Muiriscc (_Muir_ = sea) Aigli. (The promontory dominated by

Knocknaree in Sligo Bay was also called Muiriscc, Tír. 327.) The

promontory was also known as Umail. This name is preserved in the

_Owles_, designating the regions on both sides of Clew Bay, now the

baronies of Murrisk and Burris-_hoole_; the latter word also contains the

name Umail.

[122] Its height is 2510 feet. Mount Nephin, close to Lake Conn, is

higher.

[123] Carrick-on-Shannon.

[124] He first went to a place called Duma Graid, and ordained there

the arch-presbyter Ailbe, who resided at Shancoe (as mentioned above).

It may be suspected that the name Duma Graid (for which we expect a

modern Doogary) is preserved in Dockery’s Island, near the mouth of Lake

Kilglass.

[125] See note, Appendix B.

[126] Tuatha De Danann, people of the goddess Danann. They are said, in

the mythical history of Ireland, to have colonised the country and to

have been conquered by the Milesians.

[127] Fountain of Clebach.

[128] See note, Appendix B.

[129] See Appendix A, ii. 1.

[130] See Appendix A, i. 4, on the tonsure question.

[131] Selce has not been identified.

[132] Kill-araght. From here Patrick may have revisited Mag Airthic and

the Kerries.

[133] _Irrus_ Domnand, “the peninsula of Domnu” = barony of _Erris_ in

Mayo. Cp. Rhŷs, “Studies in Early Irish History,” p. 38.

[134] Ballina.

[135] It was one of the many Donaghmores, “great churches,” which Patrick

is said to have founded. He consigned it to the care of Mucneus.

[136] The name of a townland, in which there is an old churchyard and

traces of ruins, to the right of the road from Ballina to Killala, a mile

south of Killala. For Donaghmore and Mullaghfarry (_farry_ = _forrach_

= _foirrgea_, Tír. 327) see O’Donovan, _Hy Fiachrach_, pp. 466 and 467,

notes.

[137] See Appendix C, 8.

[138] _De laude sanctorum_ (Migne, _Patr. Lat._ xx.).

[139] Jerome, _Adversus Vigilantium_, c. 5.

[140] A.D. 440.

[141] See above, chap. iii. p. 64.

[142] For the evidence see Appendix C, 15.

[143] It may be Ptolemy’s _Regia_ (Ῥηγία). Cp. Rhŷs, “Studies in Early

Irish History,” p. 49 (_Proc. of British Acad._ vol. i.).

[144] See note, Appendix B.

[145] The dimensions of these houses are given, _Vit. Trip._ p. 226:—“27

feet in the Great House, 17 feet in the kitchen, 7 feet in the oratory

[_aregal_, supposed to be derived from _oraculum_]; and it was thus that

he used always to found the _congbala_” [_i.e._ the sacred enclosures,

or cloisters]. If these houses were circular, the numbers represent the

diameters. For the topography of Armagh see the paper of Reeves, _The

Ancient Churches of Armagh_ (Lusk, 1860), with a plan. The locality

of the first settlement, _ubi nunc est Fertae martyrum_, “the grave

of the relics” (Muirchu, 290), he fixes, by means of the monastery

of Temple-fertagh, which existed at the beginning of the seventeenth

century, to the land south of Scotch St., near Scotch St. river (p. 10).

[146] The two stages, first below, and then on the hill, are doubtless

historical. We may conjecture that the second and final foundation is

that which is recorded in the Annals, and that the first settlement had

been made before the visit to Rome.

[147] This is expressed by _quantum habeo_, “so far as it is mine,” in

Muirchu, 292₃₁.

[148] See note, Appendix B.

[149] _Ib._

[150] There can be little question that the (contemporary) expression _in

provincia nostra_ in _Ann. Ult._, A.D. 443, means “in Ireland,” conceived

as a single ecclesiastical province, like the province of a metropolitan.

[151] Láthrach Patricc (_Trip._ 349₈). Cp. Reeves, _Antiquities of Down

and Connor_, pp. 47 and 236; for Glore, _ib._ 87, 338; for Dunseveric,

_ib._ 286. For Clogher and Ard-Patrick (Louth) see note, Appendix B.

[152] _Ep. against Corot._ 375.

[153] _Ann. Ult._, A.D. 439.

[154] Or Killishea.

[155] See note, Appendix B.

[156] Áth Fithot, south of Tallow.

[157] Old Kilcullen, south of (new) Kilcullen, in Co. Kildare.

[158] See note, Appendix B.

[159] _Ib._

[160] In barony of Slievemargy, in Queen’s County, a mile or so

north-west of the town of Carlow.

[161] The _Life_ by Muirchu, see Appendix A, ii. 3.

[162] See note, Appendix B.

[163] Generally described inaccurately as the Acts of a Synod. The

genuineness of the document is vindicated in Appendix A, 4.

[164] For this sphere of Christian activity in the early Church see

Harnack, _Mission und Ausbreitung des Christentums_, p. 120.

[165] A Christian who believes in a supernatural female form (_lamia quae

interpretatur striga_) seen in a mirror is to be anathematised. One is

reminded of

Was seh’ ich? Welch ein himmlisch Bild

Zeigt sich in diesem Zauberspiegel!

(Goethe, _Faust_, Part I.)

[166] See Appendix A, i. 4.

[167] Chap. iii. pp. 61 _sqq._

[168] Collection of Irish Canons, 20. 5. b (ed.² Wasserschleben, p. 61).

For the possible date of the canon, and for some further illustration of

the subject, see Appendix C, 16.

[169] See Appendix C, 17.

[170] _Confession_, 368₉.

[171] _Ib._ 372₁₇; cp. 367₁₃.

[172] _Ib._ 368₂₆.

[173] Otto of Bamberg is said to have baptized 22,156 converts in

Pomerania during his first journey! Mon. Prieflingensis, _V. Ott._ ii.

20; Ebbo, _V. Ott._ ii. 11.

[174] _Confession_, 369₂₂.

[175] _Confession_, 367₁₆.

[176] _Ib._ 372. It may be conjectured, from the context, that this

happened in Connaught.

[177] So Otto of Bamberg used to distribute presents in Pomerania as a

means of propagating Christianity, Herbord, _Dial._ 2. 7.

[178] The question arises, Where did Patrick get his money? Did he

inherit from his father? It is useless to ask.

[179] _Confession_, 371₂₅.

[180] See the anecdote in Tírechán, p. 303.

[181] Epistles of Gregory, vi. 10 (A.D. 595), _M.G.H._ vol. i. p. 389.

[182] Todd, _St. Patrick_, p. 154.

[183] The early abbots of Hi (Iona) were almost entirely chosen from

a branch of the family of Tirconnell (Reeves, _Adamnan_, genealogical

table, p. 342).

[184] See the bequest of Fith Fio in _Lib. Arm._ (_Trip._ 338). It is

added that if there be no suitable person in the community of Drumlease,

some one from Patrick’s community (Armagh, or any Patrician community?)

should be chosen.

[185] _Corus Bescna_, p. 73 (_Ancient Laws of Ireland_, vol. iii.).

[186] Tírechán, 330₂₉, _fecit alteram (aeclessiam) hi Tortena orientali

in qua gens oThig Cirpani, sed libere semper_. Cp. 321₇.

[187] _Additional Notices_ in _Lib. Arm._ (338₄, _liberauit rex Deo et

Patricio_). The exact boundaries of the land are given, as if from the

original document. Two interests were concerned here, that of Caichán and

that of MacCairthin, and the land is described as “Caichán’s Fifth.” The

two men are designated as _flaith_ (lord) and _aithech_ (tenant-farmer?),

and they jointly devoted the land to ecclesiastical use.

[188] _Corus Bescna_, p. 73.

[189] _Ib._ p. 71.

[190] Cp. _Ancient Laws_, iii., Introd. p. lxxii.

[191] _Ib._ luii., _Corus B._ pp. 41, 43.

[192] _Ib._ pp. 39-43.

[193] Cp. Introd. pp. luiii. _sqq._

[194] See Appendix C, 18.

[195] Todd, _St. Patrick_, 51 _sqq._

[196] See Appendix C, 17.

[197] See above, chap. vii. p. 143.

[198] It has twenty-one letters, a b c d e f g h i l m n o p q r s t u v,

and ng (a guttural nasal, which occurs in the name _Amolngaid_; cp. the

Greek double gamma). If the Goidels had originally invented an alphabet

to suit their own language they would never have constructed this. They

had to resort to various devices to represent their sounds by its means.

See further note, Appendix B.

[199] More strictly, a new letter was added, and _u_ was differentiated

into two, to represent its two sounds. It is as well to say that in

describing the ogams as a cipher it is not intended to imply that they

were cryptic, but only that they were not an independent alphabet.

[200] For the Iberian alphabet see Hübner’s _Monumenta linguae Ibericae_

(1893). Cp. Strabo, 3. 1. 6.

[201] _B.G._ vi. 14.

[202] Desjardins, _Géographie de la Gaule_, ii. 214, note 3.

[203] Ail Clúade.

[204] _Milites._

[205] _Conf._ 360₈.

[206] Cp. _Letter, ad init., inter barbaras itaque gentes habito

proselitus et profuga._

[207] _Conf._ 374₂₉. Compare 357₁₅.

[208] _Conf._ 370₂. The passage 373₅₋₉ also supports the view in the

text. In that passage the oldest MS. has _ab aliquo uestro_; and we

should probably read _uestrum_ with the later MSS.

[209] _Ib._ 359₂.

[210] _Ib._ 372₃₁.

[211] See above, chap. iii. p. 53.

[212] But see note, Appendix B.

[213] See Appendix A, 5.

[214] This is the theory of Professor Zimmer.

[215] The Second Letter to the Corinthians seems to have been especially

before him. This was natural. In it Paul was vindicating his character.

[216] The legend will be found in _Vit. Trip._ pp. 112 _sqq._

[217] The old lists of the Armagh succession agree in assigning to

Benignus ten years as bishop, so that, as Benignus died in 467 (_Ann.

Ult., sub anno_), he would have succeeded in 457.

[218] March 17.

[219] Oirthir, not to be confounded with the kingdom of Oriel (Oirgéill),

of which it formed the eastern portion.

[220] Inundations are a recurring motive in the legends of the

Island-plain. See the salt-marsh stories, above, p. 91.

[221] This second incident can be shown to be a subsequent invention. See

Appendix C, 19.

[222] This story is also told by Muirchu, but not in immediate connexion

with the story of the waggon and oxen seized by the men of Orior. It

seems probable that the latter was suggested by the former. We meet

the duplicate waggon and oxen in the Life of St. Abban (Colgan, _Acta

Sanctorum_, i. March 16, cc. 41 _sqq._), where the account of that

saint’s death and burial and the struggle between the north and the south

Leinster men is obviously borrowed from the stories about St. Patrick.

Another story of wild bulls drawing a saint’s body to its tomb will be

found in the Life of St. Melorus of Cornwall, _Acta Sanctorum_ (Boll.),

Jan. 1, vol. i. p. 136.

[223] It is to be seen in the National Museum at Dublin. For the evidence

as to the bell and the staff, see notes, Appendix B. For the copy of the

gospels, which used falsely to be supposed to be his, see note, Appendix

B, on chap. viii. p. 162.

[224] This theory of Professor Zimmer is examined at length in Appendix

C, 21.

[225] Except in regard to Britain, and the British Church was similarly

isolated.

[226] _De Cons. Stil. Lib._ iii. l. 151.

[227] Zimmer put forward the theory that the original _Confession_

contained more biographical details than our texts (_Celtic Church_, p.

50). See my criticism showing that his argument has no basis (_Eng. Hist.

Review_, xviii., July 1903, pp. 544-6).

[228] _St. Patrick_, p. 347.

[229] _Councils_, ii. p. 296, note _a._

[230] _Celtic Church_, _ib._

[231] Bury, _ib._

[232] Huc usque uolumen quod Patricius manu conscripsit sua: septima

decima Martii die translatus est Patricius ad caelos.

[233] Sixth century or not much later, because the writings of Muirchu

and Tírechán attest its existence in the second half of the seventh

century.

[234] The attempt of Pflugk-Harttung (_Die Schriften S. Patricks_,

in _Neue Heidelberger Jahrbücher_, p. 71 _sqq._, 1893) to prove the

_Confession_ and _Letter_ spurious is a piece of extraordinarily

bad criticism. He designates the _Liber Armachanus_ as “Irlands

pseudoisidorische Fälschung.”

[235] Now fully admitted by Zimmer, who formerly doubted it.

[236] iteneris A.

[237] terreno A.

[238] requissistis A (with sign of query, Z, in margin).

[239] paradissum A.

[240] aeclessia A.

[241] Curie A.

[242] = ἐλέησον.

[243] It may be well to translate this sentence. “Church of the Scots,

nay of the Romans, in order that ye may be Christians as well as Romans,

it behoves that there should be chanted in your churches (_uobiscum_) at

every hour of prayer the _Kyrie eleison, Christe eleison_.” Compare Mr.

Jenkinson in the _Academy_, Aug. 11, 1888.

[244] aeclessia A.

[245] See Tírechán, p. 302.

[246] The assumption that all these details are taken from the book is

confirmed by the one explicit exception. The sojourn in the _insula

Aralanensis_ is given on the oral authority of Bishop Ultan (302₂₄). This

was evidently Ultan’s explanatory comment on the text _in insolis_, etc.

[247] In a paper on Muirchu in the _Guardian_, Nov. 27, 1901.

[248] It may be pointed out that the small number of the _dicta_—three,

or more probably two—is in favour of their genuineness.

[249] Since writing this, I observe that the same thing struck Loofs (_De

ant. Brit. Scot.que eccl._ p. 50). He held the _Dicta_ to be genuine,

admitting the possibility of later additions. So too B. Robert, _Étude

crit. sur la vie et l’œuvre de St-Patrick_ (1883), p. 74.

[250] This is also shown by the addition of _Christe eleison_, as Mr.

Brightman has pointed out to me. Cp. Gregory the Great, _Ep._ ix. 12.

Milan is also excluded; the Milanese only use _Kyrie_. I have had the

advantage of communicating with Mr. Brightman on the subject; otherwise

I should hardly have ventured to deal with it, as I have no liturgical

knowledge.

[251] So the missionary Boniface insists on the necessity of synods

and _canonica iura_ in a letter to Pope Zacharias (Ep. 50, p. 299, ed.

Duemmler in M.G.H. _Epp._ iii.).

[252] Bradshaw has clearly distinguished two recensions of the

collection, which he designates as the A-text and the B-text. Theodore’s

_Penitential_ is the latest work quoted in the A-text, Adamnan’s _Canons_

the latest in the B-text. See Bradshaw’s letter to Wasserschleben in

Wasserschleben’s edition of the Canons, p. lxx.

[253] Spelman i. 59 _sq._, Haddan and Stubbs, ii. 333 _sqq._

[254] The canons which are cited in the Valicellane only are marked by

square brackets. The list of correspondences in Haddan and Stubbs, ii.

333, _a_, is incomplete.

[255] See Haddan and Stubbs, ii. 333, _a_.

[256] 1. 8. b; 5. 2; 6. 2. b (two MSS. give _Sin. Rom._, one _Sin. Rom.

siue Kartagin._, the rest _Sin. Kartagin._); 7. 3. a; 20. 3. b; 40. 13.

a; 46. 35. c; 46. 38. a, b; 47. 12. b; 47. 12. c; 47. 20; 66. 19. a.

There is another case of _Syn. Rom. uel Kart._ in one MS.; 9. 1. a. The

quotation from Pope Symmachus, _Ep. ad Caes. ep. Arel._ c. 1, under the

title _Regula canonica Romana_ in 17. 8 may stand on a different footing.

[257] 14. 2. c; 17. 7. b; 17. 9. b; 18. 2. a; 20. 3. a; 20. 3. c; 20.

5. a; 21. 2; 33. 1. f; 35. 4. c; 41. 6. a and b; 42. 7; 42. 25. a; 45.

13; 45. 14; 46. 29; 52. 2; 52. 3; 52. 6; 56. 4. a; 66. 16. To these may

be added three other items: 20. 6. a, _institutio Romana_; 28. 5. b

and 33. 4, _disputatio Romana_. Also 42. 23 _Sinodus Romana_, but the

chapter is found only in one MS.; and in 3. 4 one MS. has an additional

quotation from _Synodus Romanorum_. I do not include 42. 24, because the

heading _eadem sinodus_ may be referred to the heading of c. 22 _Sinodus

Hibernensis_, and not to the heading of 23 _Sinodus Romana_, which, as I

have mentioned, is found in only one MS. (Sangallensis).

[258] Here are the two sections:—

e. _Sinodus Romana_: Omnis qui fraudat debitum fratris ritu

gentilium excommunis sit donec reddiderit. f. _Item_: Qua

fronte rogas a Deo debitum tibi dimitti cum debitum proximi tui

non reddidisti?

[259] It may be observed that in the Valicellane MS. we find some

instances of _Hibernensis uel Romana_; 33. 4, 6, and 9.

[260] 11. 1. b; 20. 5. b; 21. 12; 21. 6. b; 25. 3; 25. 4; 29. 7; 37. 27;

37. 29; 42. 26. b; 44. 9; 46. 32. b; 47. 11. b; 67. 2. d. Of these 37.

29 has the curious heading _Sinodus totius mundi et Patricius decreuit_

(with the variants _Sinodus Hibernensis et Pat. decr._, and simply

_Sinodus Hibernensis_).

[261] 21. 25 occurs only in one MS., 37. 6 in two.

[262] 25. 3 and 4.

[263] In Ware’s _S. Patricio ... adscripta Opuscula_ (London, 1656, a

rare little volume), pp. 85-7. See below, p. 245.

[264] “Brit. ecc. ant.,” in _Opera_, vol. vi. p. 491. The same view was

urged by Varin. But neither Ussher nor Varin gave positive proof.

[265] It is officially recognised in the 40th canon of the Fourth Council

of Toledo, A.D. 633.

[266] Cp. the 15th canon of Nicaea, Mansi, ii. 200.

[267] So too in the _Hibernensis_, i. 22, a and c.

[268] _Fastes épiscopaux de l’ancienne Gaule_, vol. i. p. 41.

[269] Cp. also canon 27 of the Council of Hippo, A.D. 393 (Mansi, iii.

923): ut episcopi non proficiscantur trans mare nisi consulto primae

sedis episcopo suae cuiusque prouinciae, ut ab eo praecipue possint

formatas sumere.

[270] Printed by Pamelius in his edition of Cyprian; by Ware in his

_Opuscula_ of St. Patrick (1656); by Migne, _P.L._ 40, 649 _sqq._

[271] For his alleged relationship see below, p. 292.

[272] To this rule the MSS. present two exceptions, which should be

corrected: v. 70, _praeuidit_, which has been corrected to _praeuidet_

(cp. Atkinson, _Lib. Hymn._ ii. 13), and v. 66, _qui ornatur uestimento

nuptiale indūtus_, where we ought evidently to read _inclŭtus_.

[273] 311₂₉, Rolls ed.

[274] _Ann. Ult. s.a._

[275] 314₂₈; _Ann. Ult. s.aa._

[276] See Bury, _Tírechán’s Memoir of St. Patrick_ (_Eng. Hist. Rev._

April 1902), p. 255.

[277] _Ib._ pp. 237, 238, 260.

[278] 311₂₃₋₂₅. Cp. Bury, _ib._ 261.

[279] 311-312.

[280] Bury, _ib._ 258. “His whole book is a practical service to the

cause of the claims of Armagh. It is virtually a list of the churches

which claimed to have been founded by Patrick. If it had been completed,

it would have exhibited the full extent of the _paruchia Patricii_.”

[281] 313₂₇, 307₇, 313₂₈, 323₂₉, 319₅, 318₂₅ × 301₉.

[282] 311₂₈.

[283] _In libro apud Ultanum_, 302₃.

[284] Bury, _Tírechán’s Memoir_, 248-250.

[285] Bury, _Tírechán’s Memoir_, 239.

[286] 319₄. Bury, “Supplementary Notes” (_Eng. Hist. Rev._ Oct. 1902),

702-703.

[287] It stops at p. 331, l. 9, in the Rolls ed. See my paper in _E.H.R._

_ut cit._ p. 237.

[288] _Proc. of R.I.A._ (xxiv. sect. C, 3), 1903, p. 164 _sqq._

[289] This has been fully recognised by Dr. Gwynn, _loc. cit._

[290] _Machia_ (330₂₂) probably means Domnach Maigen, not Armagh (Gwynn,

_loc. cit._).

[291] 348₁₈, d.g. [= _Duma Graid_, Reeves, but this is far from certain];

Ailbe i Senchui altáre; and Machet Cetchen Rodán Mathona. Compare also

350₈ with 331₄.

[292] The credibility of the Genealogy, as an independent record, is

particularly strong; the Ballymote scribe was acquainted with the

_Tripartite_, and quotes from it _à propos_ of the sons of Forat,

notwithstanding the contradiction. The discrepancy with the Patrician

tradition is, in fact, a guarantee that the record is trustworthy.

[293] P. 269₁₃ Rolls ed.: _patris mei Coguitosi_, the brilliant

correction of Bishop Graves for the corrupt _cognito si_ in A. On the

passage, and on Cogitosus, see his paper in the _Proceedings R.I.A._

viii. 269 _sqq._

[294] See Graves, _ib._ The conjecture is accepted by Dr. Stokes (_Trip.

Life_, 269, note 2).

[295] Colgan, _Acta SS._ p. 465 and _n._ 31.

[296] _Ann. Ult. s.a._

[297] See Reeves, _Adamnan_, pp. l. li. Professor Kuno Meyer has just

published an old Irish treatise on the “Law of Adamnan” passed at this

synod (“Cáin Adamnáin,” in _Anecdota Oxoniensia_, 1905). The document

contains a list of the bishops, abbots, and kings present at the

synod which was held at Birr. Muirchu appears (p. 18): Murchu maccúi

Machthéine. Muirchu appears in the Martyrologies under June 8 (see

_Calendar of Oengus_, ed. Stokes, p. xciii.).

[298] I suggested this in the _Guardian_, Nov. 20, 1901, p. 1615, c. 2.

[299] Muirchu does not name his father’s work, but his expression

_ingenioli mei_ (269₁₄) may be an echo of the _rusticus sermo ingenioli

mei_ in the prologue to the _Vita Brigidae_.

[300] 271₁₇.

[301] 269₁₉.

[302] This is suggested by the use of the third person. In the Preface

Muirchu writes in the first person. The note is similar to the note which

is prefixed to the memoir of Tírechán and is obviously due to a copyist.

[303] See 495₃ Rolls ed. (ad Britanias nauigauit), and 495₂₆₋ (the second

captivity).

[304] Some mistakes have occurred in the course of compilation and

transmission: see below, p. 348.

[305] See Bury, _Tírechán’s Memoir of St. Patrick_, p. 16; but I did not

see then that the source was probably Irish.

[306] Tírechán, 330₁₅₋₁₉: Muirchu, 276₁₁₋₁₄, and 300₁₀₋₁₃: Bury, _ib._ p.

14.

[307] “The Tradition of Muirchu’s Text,” in _Hermathena_, xxviii. pp. 199

_sqq._

[308] On account of the notice of Auxilius (of Killossy) and Iserninus

(of Kilcullen). It seems very probable that the notice of Iserninus in

the _Liber Arm._ (f. 18) may have been derived from information furnished

by Bishop Aed on the occasion of his visit to Armagh. See above p. 253.

[309] In the Table of Contents to Book II. this is the title of the first

and the last section alike; but the last item in the table was wrongly

taken to be a heading of sect. 1 (though there are no other headings to

the sections), until the true explanation was pointed out by Dr. Gwynn.

[310] _Ann. Ult. s.a._

[311] This is the conjecture of Zimmer, _Celtic Church_, p. 81.

[312] These parts were first published by Rev. E. Hogan, _Anal. Boll._

vol. i. I have had the advantage of using a photograph of the MS., kindly

given me by Dr. Gwynn.

[313] See also Todd, _St. Patrick_, 489; Stokes, Intr. to _Tripartite_,

cxi. _sq._; Bernard and Atkinson, _Liber Hymnorum_, ii. 175-6.

[314] Criticised by Thurneysen, _Revue celtique_, 6, 326 _sqq._, who

rejects the theory of interpolation except in the case of stanza 17. So

too Stokes and Strachan.

[315] The stanzas which are abnormal, or defective, in metre, assonance,

etc., are—2, 7, 9, 12, 14, 15, 17, 24, 26, 27, 29, 30, 31, 33 (for

criticisms on their subject-matter see Atkinson, _ib._ xliii. _sqq._).

Stanza 16 has a “glossatorial” character (_ib._ xlviii.). The ejection

of 10 on ground of subject-matter may be confirmed by the abnormal

endings (_nua_ and _tua_, cp. Atkinson, xlii.). 18 (rejected by Zimmer

and Atkinson) is clearly an imitation of 18, and this is indicated by

the repetition of the rhymes. The rejection of 19 and 20 depends on the

subject-matter, and 21 repeats 19. The irrelevance of 22 is obvious.

I leave the second stanza as doubtful, for though there is a metrical

anomaly (_daec_ a disyllable), there is no objection on the ground of the

subject-matter; but it could be dispensed with.

[316] Cp. Muirchu, 296₁₂.

[317] See Bury, _Guardian_, Nov. 27, 1901, p. 1647.

[318] There is no other edition.

[319] I have shown, from misunderstandings in V₄ that its author was

ignorant of Irish, while the author of V₂ was an Irishman (_op. cit._

197).

[320] Dr. W. Levison of Bonn kindly called my attention to a _Vita_

preserved at St. Omer which proves to be a copy of the _Vita Secunda_

different from that used by Colgan. It is contained in Cod. 716

(_Legendarium beatae Mariae de Claromarisco_), a book of the thirteenth

century, vol. ii. ff. 155-9. For the text of _Vita Quarta_, the Stowe MS.

105A (Brit. Mus.) is important (see my _Tradition_, etc., p. 186 _note_).

[321] Except so far as to show that neither of the two existing MSS. was

used by Colgan. The text is based on Rawlinson B. 512, but it is not

explained why this was chosen as the basis in preference to Egerton 93

(which—I speak under correction—does not seem inferior).

[322] _Ann. Ult. s.a._

[323] _Ib._

[324] _The Tripartite Life of St. Patrick_, in _Trans. of R.I.A._, xxix.

Pt. vi. 1889.

[325] A scribe of Armagh, _ob._ 725 (_Ann. Ult._).

[326] _Ann. Ult. s.a._

[327] The narrative here (c. 14) is very confused, and perhaps there is

interpolation.

[328] I observe that Lanigan wished to derive _Hermon_ from _her_,

“great,” and _maen_, “rock.”—Todd, _St. Patrick_, 337 note.

[329] The literature which I have used in working through the Nennian

problem is as follows:—Zimmer, _Nennius Vindicatus_, a brilliant and

indispensable book, but too ingenious, and full of wiredrawn arguments;

many of the conclusions have been upset by the Chartres text (Mommsen’s

Z) which Zimmer left out of consideration. This text was published by

Duchesne, _Nennius retractatus_, in _Revue celtique_, xv. 173 _sqq._;

and was used by Mommsen for his authoritative edition of the work in

_Chronica Minora_, vol. iii. (it is much to be regretted that he did

not devote a separate column to printing the text of Z in full). New

light was then thrown on the problem by Thurneysen, _Ztsch. f. deutsche

Philologie_, xxviii. 80 _sqq._ His interpretation of _exberta_ in the

title in the Chartres MS. as a mistake for _excerpta_ (_Incipiunt

excerpta filii Urbagen_) seems probable (Dr. Traube’s emendation

_experta_ has not convinced me); his identification of this son of

Urbagen with Run map Urbgen, who baptized the Northumbrians in 627

(_Hist. Britt._ c. 63), though plausible, cannot be considered certain.

Duchesne, in a judicious and instructive criticism with reference to

Mommsen’s edition and Thurneysen’s article, has summed up the conclusions

which may safely be drawn from the data: _Revue celtique_, xvii. 1 _sqq._

Mr. E. W. B. Nicholson, who had reached several of Professor Thurneysen’s

conclusions independently, published his views in _Ztsch. f. celtische

Philologie_, iii. 104 _sqq._ The most important point in this paper is

that the true reading of the important words in the title of the Chartres

MS. is: _exberta fili vrba gen_. See also L. Traube in _Neues Archiv_,

xxiv. 721 _sqq._

[330] Nennius, _Preface_, ed. Momms., p. 143; Elbodug died A.D. 809,

Zimmer, _Nenn. Vind._ 51.

[331] (1) Noteworthy is the explanation of _sed prohibuit illum Deus quia

nemo_ etc. (Muirchu 272₂₀) by the insertion of _per quasdam tempestates_

after _Deus_. In the context this is incongruous, and it can hardly have

been originated by Nennius. Had he a MS. of Muirchu containing additions

inserted from Muirchu’s source? [_Deus_ is in the Bruxellensis, but

omitted in the Armachanus.] (2) He changes Muirchu’s _Victoricus_ into

_Victor angelus Dei_. (3) He says that Patrick’s first name was _Maun_

(_Magonus_, Tírechán, 302₅). (4) It is to be observed that while Muirchu

mentions two views as to the duration of Patrick’s sojourn with Germanus,

namely, forty or thirty years, Nennius gives a much smaller period, _per

annos septem_. Thus Muirchu’s _Life_ does not explain Nennius, c. 52; he

had some additional material.

Nennius agrees with Brux, and V₂ V₄ in recording that Palladius died _in

terra Pictorum_.

[332] Todd Lecture, Series iii. The _Codex Palatino-Vaticanus_, 830, p.

354 _sqq._, cp. 252 _sqq._

[333] _Ib._ 247 _sqq._

[334] Cp. Columbanus, _Epist._ (M.G.H., Epp. iii.) 157, and the notice

in the Würzburg MS. of St. Matthew, quoted by Zimmer, _Nenn. Vind._

252, note (Scheps, _Die ältesten Evangelienhandschriften der Würzburger

Bibliothek_, 27).

[335] A genealogy of Brito is ascribed by Gilla Coemgin to _senior

nobilis Guanach_, and Todd pointed out that the reference was to the

_Liber Cuanach_ (Zimmer, _Nenn. Vind._ 250-1). Calling attention to the

notice in _Ann. Ult. s.a._ 616, _usque hunc annum scripsit Isidorus

cronicon suum_, Zimmer observes that the old recension (up to 616) of

Isidore’s chronicle was known in Ireland, and conjectures that its

arrival may have been the stimulus which prompted the work of Cuana.

[336] An older authority, Maucteus, was quoted by Cuana (_Ann. Ult. s.a._

471).

[337] Dr. MacCarthy quotes appropriately the 20th canon of the Council of

Milevi, A.D. 416.

[338] See Migne, P.L. 87, 969.

[339] For such entries in the blank spaces of a Paschal Table, compare,

_e.g._ the _Paschale Campanum_ (_Chron. Min._, ed. Mommsen, i. 745

_sqq._).

[340] Mr. Phillimore’s suggestion that _cum_ is a misrendering of the

Old-Welsh _cant_ = by, seems improbable, as the notice is not likely to

be a translation. I should say that _cum_ is simply a dittogram of the

last syllable of _dominicum_, and has ousted _a_.

[341] We are indeed enabled to infer that before the tenth century A.D.

457 had been maintained by some to be the date of Patrick’s death.

[342] An examination of the dates in the sixth century suggests that

the entries of contemporary events did not begin before the seventh.

Certainly the erroneous date of the battle of Mons Badonis was a late

insertion.

[343] For references see Tillemont, _Mémoires_, xv. 769. Leo had taken

the step of writing to the Emperor Marcian on the matter in 453.

[344] _Vit. Hilarii. Arel._ 16. See Levison, _Neues Archiv_, xxix. p. 99.

[345] Levison, _loc. cit._ pp. 125 _sqq._

[346] Compare Zimmer’s criticism, _Celtic Church_, 64-5.

[347] Stokes (_Urkeltischer Sprachschatz_, 198) and Rhŷs seek a Celtic

etymology for Magonus. Rhŷs treats it as a derivation from Goidelic

_magus_ (whence the Irish _mug_, “servant”), meaning perhaps originally a

“boy.”

[348] His day was Nov. 27; _Mart. of Donegal_, p. 319.

[349] Dichu of Saul appears in the _Martyrology_ of Donegal under April

29 (p. 114).

[350] Instances are collected by Professor Rhŷs in _Proceedings of

R.S.A.I._ Pt. i. vol. xxxii. p. 5; to which add the Donard stone (_ib._

Pt. ii. vol. xxxiii. p. 114).

[351] This comes from Muirchu’s Irish source for the legend. See above,

App. A, ii. 3.

[352] For _scriptum erat_ I would read _scriptum quod scriptum erat_.

[353] _sine honore_, supposed to mean “without recognition of my

episcopal title.”

[354] _male uidimus faciem designati nudato nomine_ (365₂₉). This gives

much better sense than the ordinary rendering, which refers _designati_

to Patrick, and _nud. nom._ to the suppression of the episcopal title. I

observe that Mr. White, though he does not adopt it in his translation,

gives it as an alternative interpretation in his note.

[355] Another petition (which in _V. Trip._, p. 116, appears in connexion

with Mount Crochan) is added in V₃, c. 88. See Bury, _Trans. of R.I.A._

xxxii. C. Part iii. p. 223.

[356] I retract the date “_c._ A.D. 850” in my note †, _op. cit._ p.

218. I have abandoned a view of the relations of the Nennian MSS. which

prompted me to assign this date.

[357] The Armagh tradition, connecting Columba with St. Patrick’s tomb,

is referred to in the _Additions to Tírechán_ (see Appendix C, 19).

[358] I have shown that this was possibly the original corruption,

“Tradition of Muirchu’s Text,” p. 196.

[359] _haut_ has been restored for -cha _ut_ of the Brussels MS. The

restoration, obviously right, is borne out by Probus (Colgan, _Trias

Thaum._ 47), who here transcribes Muirchu. See Stokes, _ad loc._

[360] Probus here, using another but related MS. of Muirchu, gives

_Nentriae prouinciae_.

[361] Mr. Nicholson’s explanation that it is a corruption of _Britanniae_

will hardly be accepted. Palaeographically it has no probability; it was

hardly necessary, and we should rather expect _Britanniarum_.

[362] It is obvious how readily the corruption rose from

_bannauētabernie_ or -_urnie_.

[363] Stokes, _Tripartite_, Introduction, p. cxxxvii. He refers to

Gregory of Tours, _Hist. Franc._ x. 28, _in uico Nemptudoro_. Skene

(_Celtic Scotland_, ii. 436, n.) identifies Nemthur with _Neutur_ in the

Black Book of Caermarthen (_Four Ancient Books of Wales_, vol. ii. p. 3).

[364] It is perhaps hardly necessary to observe that Ammianus when he

says that Theodosius _instaurabat urbes_ does not refer to Valentia

(which he has not yet mentioned) in particular, but to Britain in general.

[365] In A.D. 383 he was sent on an embassy to the Persians (whom Keller

calls “Parthians,” p. 15), and married Serena soon after (384 or 385). In

A.D. 386 he was engaged in the campaign with the Gruthungi.

[366] This seems to me a just remark of Mr. Hodgkin, _Italy and her

Invaders_, vol. i. (2nd ed.), p. 716.

[367] “Tyrants of Britain, Gaul, and Spain,” _Eng. Hist. Review_, 1886,

i. 55, note 3. This is the chief study on the subject (republished in

_Western Europe in the Fifth Century_, 1905, chaps. ii., iii.). Cp. App.

20 to Bury’s edition of Gibbon, vol. iii.

[368] II. k. Jan. ed. Mommsen, p. 465.

[369] Prosper’s notice, _sub ann._ 407, agrees with the implication in

Olympiodorus.

[370] Preface to edition of Gildas (_Chron. Min._ iii. 1), p. 7.

[371] The Picts are described as a _gens transmarina_ (the Picts of

Dalaradia?).

[372] _Ann. Ult. sub a._ 461.

[373] _Ann. Inisf._ ed. O’Conor _sub a._ (“488” =) 493: _anno ccccxxxii a

passione Domini_.

[374] Ed. Mommsen, p. 158-9.

[375] Rolls ed. p. 365₁₅, _nescio, Deus scit, si habebam tunc annos

quindecim_.

[376] _Ib._₁₁.

[377] 366₁₀.

[378] 365₁₀.

[379] So it is interpreted by Todd, _St. Patrick_, p. 392; by Neander,

_Allg. Geschichte der christlichen Religion u. Kirche_, iii. 185, note;

by Zimmer, _Celtic Church_, p. 43.

[380] I adopt the admirable correction of Dr. N. J. D. White for

_occasionum—inuenerunt me_ (MSS.).

[381] The truth is that the length of time which elapsed since the

wrongdoing is not pertinent. The point is the interval between the

confession of it to his friend, who offered no obstacle to his ordination

as deacon (nor later to his ordination as bishop), and the occasion on

which it was used against him. It is this lapse of 30 years which makes

his friend’s conduct so unaccountable.

[382] It is interesting to note that this name is found on three ogam

stones, in the genitive forms _Gosoctas_, _Gossucttias_, and _Gosocteas_

(see Rhŷs, _Ogam-inscribed Stones_, in reprint from _Proc. of Royal Soc.

of Antiquaries of Ireland_, Part i. vol. xxxii. 1902, p. 24).

[383] The force of _adhuc_ here (not synonymous with _iterum_) is exactly

like its force in _adhuc capturam dedi_. See above, Appendix B, note on

p. 34.

[384] Muirchu’s _et erat annorum triginta_ (496₁) must not be alleged; it

is based on a misconception of the _annos triginta_ of the _Confession_

(365₁₀), and influenced by the Scriptural parallel to which Muirchu

refers.

[385] Jerome, Ep. cxxiii. (_ad Ageruchiam_), Migne’s ed. vol. i. 1057-8.

[386] After A.D. 439.

[387] vii. § 50. Orosius, vii. 38 and 40, gives no details of the

devastations.

[388] Edited by Brandes in the _Corp. Script. Eccl._ 1889.

[389] Compare the sketch of Professor Dill, based on these poems, in

_Roman Society_, bk. iv. cap. 2 (p. 263, ed. 1).

[390] So I would amend the corrupt _metuendis_. Ellis reads _tetricis_,

and suggests _mediis_. But there is greater point in _nudis_; it implies

“without the crops, etc., which would naturally attract an enemy.”

[391] P. 18.

[392] Pp. 16-18, and Appendix B.

[393] Cp. Arrian’s _Cynegeticus_, esp. chaps. 1, 2, 3.

[394] The nature of the cargo is another argument against the view that

Britain was the destination.

[395] _Annis_ xxx. Tír. _loc. cit._ The numeral is less probably a

mistake for iii. than the result of a mistaken attempt to account for

the chronology of Patrick’s life on the hypothesis that he lived to the

age of 120. Other schemes assigned 30 or 40 years to the Auxerre period

(Muirchu, 271₂₂).

[396] For the traditions and legends connecting Palladius with Scotland

see Skene, _Celtic Scotland_, ii. 29 _sqq._

[397] Schoell, _De ecclesiasticae Britonum Scotorumque historiae

fontibus_, 1851, p. 77; Loofs, _De antiqua Britonum Scotorumque

ecclesia_, p. 51 (1882). Compare B. Robert’s criticisms on their

arguments, _Étude critique_, etc., pp. 28 _sqq._

[398] Professor Zimmer (_Early Celtic Church_, p. 38) regards Palladius

as a Roman rendering of Patrick’s name Sucatus (warlike), following a

suggestion of Mr. O’Brien (_Irish. Eccl. Record_, 1887, pp. 723 _sqq._).

This is quite unconvincing, in the absence of any evidence that Palladius

was a Briton. Why should he not have belonged to the _stirps Palladiorum_

of Bourges (Apoll. Sidon. _Epp._ vii. 9, 24)? Cp. Duchesne, _Fastes

épisc._ ii. p. 26, _note_.

[399] P. cxxv. ed. Stokes (1871). This calendar, Mr. Stokes has shown,

cannot be earlier than the end of the tenth century.

[400] The name of Palladius does not appear in any Irish calendar of

saints. It appears under July 6 in the Breviary of Aberdeen. See Todd,

_St. Patrick_, p. 299.

[401] Cp. _Chron. Scot._ pp. 24-25; Cormac’s _Glossary_ sub _Mogheime_.

[402] M. d’Arbois de Jubainville has a note on Sen Patrick in _Revue

celtique_, ix. p. 111 _sqq._ He thinks a second Patricius came over to

Ireland as a mere name in a copy of the Hieronymian Martyrology towards

end of sixth century; and that of him were made the two Patricks of

August 24 in _Mart. of Tallaght_.

[403] See Bury, “Tradition of Muirchu’s Text,” p. 205.

[404] The name of this place, whether in the form _Ebmoria_ (_Lib. Arm._)

or in the variant form _Euboria_ (for the variants see Bury, “Tradition

of Muirchu’s Text,” p. 185) is unknown. The context shows that it lay

between Auxerre and the north coast of Gaul. Ebroica = Evreux would be

my guess. (It probably became the seat of a bishop before the end of the

fifth century, but probably not long before. Cp. Duchesne, _Fastes ép._

ii. 226).

[405] On the evidence for this date see _note_ 15 in my “Sources of

Early Patrician Documents” (_E.H.R._, July 1904), where I refer to the

discussion of Dr. W. Levison in his monograph cited above, Appendix A, i.

7.

[406] _Nennius Vindicatus_, p. 123, _note_. Later compilers divided

_Amatho rege_. In Irish conditions a _rex episcopus_ would not sound so

strange as in Gaul, though I do not know of an instance before Cormac of

Cashel at the beginning of the tenth century. In _Tripartite Life_ (p.

34) we get a further step in the evolution; the bishop appears as Amatho

rí Románach (King of the Romans); and the final stage is reached when

Patrick is ordained _coram Teodosio imperatore_.

[407] I pointed out this inference in “Sources of Early Patrician

Documents” (_cit. sup._).

[408] _Ann. Ult. s.a._ 439. For the separate coming of Iserninus see

account in _Lib. Arm._ f. 18 rᵒ a (342, Rolls ed.). It is there indeed

supposed that they started at the same time from Gaul, but were severed

by storms, and so arrived separately in Ireland. The motive for this tale

is evidently the genuine record that they did not come together.

[409] There are three cases (Zimmer, _ib._ 24-26):

1. Long ō instead of long ā: _e.g._ _trindōit_ (= _trinitatem)_; _altōir_

(= _altāre)_; _caindlōir_ (_candelarius_); _notlaic_ (_nātalicia_);

_popa_ (_papa_). If the Latin forms had come directly to Ireland, the ā

would have been preserved; ō for ā is characteristically British.

2. _c_ for _p_: _casc_ (_pascha_); _crubthir_ (_presbyter_). Words of

this kind must have been “interpreted to the Irish by British mouths,”

for if they had been borrowed from Latin, _p_ would have been preserved.

The motive for the change of _British p_ to _c_ in the loan-words was the

observed fact that in native words Gaelic _c_ corresponded to British

_p_ (representing the velar _q_). The treatment of Patrick’s name is a

significant instance. From its British form we get in Irish _Coithrige_,

from its Latin form _Patraicc_. _Caille_, a veil (_caillech_, a nun), is

supposed to be another case (= _pallium_). Mr. Nicholson suggests that it

is Celtic, = _capillia_ (_Keltic Researches_, p. 104).

3. _sr_ for _fr_, _sl_ for _fl_: _e.g._ _slechtan_ (_flectionem_),

_srogell_ (_flagellum_). In case of a direct borrowing, _sr_, _sl_ would

have been kept; but coming through British they were treated on the

analogy of Irish _sruth_ = Br. _frut_.

[410] _Tracia_ being an error for _Frācia_.

[411] The prose list in the _Book of Ballymote_ describes both Duachs as

Tenga Uma (and as slain by Muirchertach), but in the case of the first

_Duach Galach_ is written above.

[412] Dai Galach is mentioned as Brian’s youngest son, and destined to

reign in Connaught, in the text printed, with translation, by Dr. W.

Stokes, from the _Yellow Book of Lecan_ (col. 898), concerning the sons

of Eochaid Muigmedoin, in _Rev. Celt._ xxiv. (1903), p. 182.

[413] The notice is discussed by Todd, _St. Patrick_, pp. 469-70, but he

does not attempt any explanation of _probatus est_, and by omitting the

first part of the notice altogether he shows that he did not apprehend

the significant association of the _probatio_ with Leo.

[414] _Nostra provincia_ is ecclesiastical Ireland, or north Ireland—not

a particular district or kingdom.

[415] The _Vit. Trip._ (Part 3, p. 238) places the visit to Rome _after_

the foundation of Armagh. This order seems to be simply an inference from

the fact that the relics which Patrick got at Rome were brought to Armagh.

[416] The references are to Gundlach’s ed., _Epp. Mer. et Kar. Aeui_,

vol. i.

[417] Mr. Warren rightly rejects the pregnant meaning found in this

passage by Döllinger (_Liturgy and Ritual_, p. 39).

[418] _difficiles_ is in three MSS., including the Coloniensis (eighth

century): it is clearly wanted, and (if there was not some stronger and

fuller phrase in the original canon) it has a very pregnant meaning.

[419] I believe that the silence of Cummian in his letter to Segéne

(Migne, _P.L._ 87, 970) has been urged as an objection to the

authenticity of the canon. But the argument from silence in such a case

as this has no cogency. It is quite possible that he was not aware of the

canon.

[420] See B. Krusch, _Der 84jährige Ostercyclus und seine Quellen_, 1880.

[421] Columbanus, _Epp._ ed. Gundlach, Ep. 4, p. 162.

[422] _Ib._ Ep. 1 (to Gregory I.), p. 157; _Catal. Sanct. Hib., unum

Pascha, quarta decima luna post aequinoctium vernale_ (H. and S.,

_Councils_, ii. 292, 293).

[423] Dr. MacCarthy’s explanation (Introd. to _Annals of Ulster_, iv. p.

lxxix) of an entry in the Chronicle of Marius Aventicensis (ad ann. A.D.

560), as implying a survival of the old 84 cycle in Gaul, does not seem

tenable (see A. Anscombe, _Z. f. Celtische Philologie_, iv. p. 333).

[424] I have omitted all reference to the abstruse technique of the

computations, and the question as to the difference between the two

cycles of 84 (differing in the number of the _saltus_), for which see

MacCarthy, _op. cit._ pp. lxv. _sqq._; Krusch, _op. cit._

[425] For dating, see MacCarthy, _op. cit._ cxl. cxli.

[426] Migne, _Patrol._ 87, 969.

[427] Krusch, 302 _sqq._ For dating, see MacCarthy, _op. cit._ cxvii.,

where the proof of fabrication will be found.

[428] So MacCarthy, cxxxvii., note 2.

[429] Dr. MacCarthy shows that Cummian was uncritical (cxl.), but I

cannot go quite so far as he. He accepts, as genuine tradition, the

statement that Patrick brought a cycle with him (rightly observing

that the addition of _et fecit_ cannot be pressed), but holds that it

was identical with the Celtic cycle of later times. It is with much

diffidence that I venture to differ from him on this point; but I owe

it to his investigations that I have been able to reach a definite

conclusion.

[430] The learned Appendix A, pp. 123 _sqq._, in Reeves, _Eccl. Ant._, is

valuable still.

[431] Ed. Gundlach, _M.G.H., Epp. Mer. et Kar. Aeui_, i. p. 163. See

Jerome, _Ep._ 58 (Migne, 22, 583).

[432] 450 in Tírechán and one MS. of _Catal. Sanctorum_; 350 in other

MSS. of _Cat. Sanct._

[433] It is not, of course, certain that Bishop Muirethachus _qui fuit

super flumen Bratho_, for whom Patrick and Bronus wrote an alphabet,

is identical with the _epscop Muiredaig_, “an aged man of Patrick’s

household,” who is said in _Vit. Trip._, p. 134, to have been left by

Patrick in Cell Alaid. The foundation of Cell Alaid is not mentioned by

Tírechán.

[434] The text of Tírechán leaves it doubtful whether the foundation

of Cellola Tog was laid under Patrick’s auspices, and whether Patrick

ordained Cainnech bishop. He is not included in the author’s catalogue of

bishops ordained by Patrick (p. 304).

[435] For the _monachi_ of Assicus, see 313₃₃ and 328₃₀.

[436] Cainnech was a _monachus Patricii_, 324₁₃. Another _mon. P._ is

mentioned 329₉. The deacon Coimanus of Ardd Licce seems to have been

another, 317₂₆. Gengen and Sannuch are described as _mon. P._ 305₁₇; and

for the foreigners _Conleng and Ercleng_, see 313₁₂.

[437] Thus Slébte, the seat of Bishop Fíacc, would have been a _civitas_.

It is called _Slebtiensis civitas_ by Muirchu (271₁₉). See also _Lib.

Angeli_, 355, _omnis aeclessia libera et civitas <quae> ab episcopali

gradu uidetur esse fundata_. Cp. an Irish canon, ascribed to Patrick,

in the _Hibernensis_, 29, c. 7 (p. 101 ed. Wasserschl.), _aut a sancta

ecclesia aut in civitate intus_.