Skip to main content

Full text of "Neglected factors in the study of the early progress of Christianity"

See other formats


This is a digital copy of a book that was preserved for generations on Hbrary shelves before it was carefully scanned by Google as part of a project 

to make the world's books discoverable online. 

It has survived long enough for the copyright to expire and the book to enter the public domain. A public domain book is one that was never subject 

to copyright or whose legal copyright term has expired. Whether a book is in the public domain may vary country to country. Public domain books 

are our gateways to the past, representing a wealth of history, culture and knowledge that's often difficult to discover. 

Marks, notations and other maiginalia present in the original volume will appear in this file - a reminder of this book's long journey from the 

publisher to a library and finally to you. 

Usage guidelines 

Google is proud to partner with libraries to digitize public domain materials and make them widely accessible. Public domain books belong to the 
public and we are merely their custodians. Nevertheless, this work is expensive, so in order to keep providing this resource, we liave taken steps to 
prevent abuse by commercial parties, including placing technical restrictions on automated querying. 
We also ask that you: 

+ Make non-commercial use of the files We designed Google Book Search for use by individuals, and we request that you use these files for 
personal, non-commercial purposes. 

+ Refrain fivm automated querying Do not send automated queries of any sort to Google's system: If you are conducting research on machine 
translation, optical character recognition or other areas where access to a large amount of text is helpful, please contact us. We encourage the 
use of public domain materials for these purposes and may be able to help. 

+ Maintain attributionTht GoogXt "watermark" you see on each file is essential for informing people about this project and helping them find 
additional materials through Google Book Search. Please do not remove it. 

+ Keep it legal Whatever your use, remember that you are responsible for ensuring that what you are doing is legal. Do not assume that just 
because we believe a book is in the public domain for users in the United States, that the work is also in the public domain for users in other 
countries. Whether a book is still in copyright varies from country to country, and we can't offer guidance on whether any specific use of 
any specific book is allowed. Please do not assume that a book's appearance in Google Book Search means it can be used in any manner 
anywhere in the world. Copyright infringement liabili^ can be quite severe. 

About Google Book Search 

Google's mission is to organize the world's information and to make it universally accessible and useful. Google Book Search helps readers 
discover the world's books while helping authors and publishers reach new audiences. You can search through the full text of this book on the web 

at |http : //books . google . com/| 







Neglected Factors in 
the Study of the Early 
Progress of Christianity 

STATE OF NEW YORK, 1897 4f fr 


r** ' 

eglected Factors in ' the 
Study of the Early Pro- 
gress of Christianity. By 

the Rev. James Orr, d.d., Professor of 

Church History in the United Presbyterian Theo- 
logical College, Edinburgh *■•«■•»»• 





t ( 




Tbe Bltfldiliaii Theology 

and the Evangelical Faith. 

A Volume of "The Theological Educator." 

Fcap 9vo, cloth, Ss. Sd., Second Edition. 

" This masterly exposition and criticism of 
the great German theologian. . . . Dr. Orr has 
done a work which will enhance his reputa- 
tion and make all his brethren grateful to 
him." — Aberdeen Free Press. 

"His volume is not a large one, but it is 
packed with matter, and it embodies the well- 
considered results of careful and extensive 
reading. It is the best English book we have 
on the subject. Nothing is left unnoticed 
that is necessary to a proper appreciation of 
this influential school of theology."— Cn'/^ca/ 


Through the liberality of Mr. Henry A. Morgan, N.Y.S. 

The three Lectures in this volume were 
originally prepared for the Mansfield Summer 
Schooly Oxford^ 1894. They were delivered 
as the Morgan Lecture Course^ in October^ 
1897, in the Theological Seminary of Auburn^ 
in the State of New York, They are now 
published by request of the Faculty, 





New spirit in Early Church studies — Baur and 
his successors — Influence of Pagan environ- 
ment on Christianity — Less attention given 
to the action outward of Christianity on 
Paganism — ^The spread of Christianity late- 
rally y i.e., in respect of mere numbers, greater 
than ordinarily recognised — Estimates on 
this subject — Difficulties arising from frag- 
mentariness of sources and unequal distribu- 
tion of Christianity — The Catacombs a new 
factor — Results from Catacomb discoveries 
— Comparison with New Testament and 
other data — Early progress of the Church — 
Christianity in Asia Minor — ^The Apologists, 
&c. — Carthage — Alexandria — Antioch — 
Gibbon's objections — Gaul and Spain — The 
final struggle — General result . . .13 







Influence of Christianity on the higher ranks of 
society under-estimated — New Testament 
evidence — Witness of the Catacombs — 
Pomponia Graecina — Flavins Clemens and 
Domitilla — Acilius Glabrio — Notices in 
Second Century — ^The wealth of the Church 
of Rome — The witness of the persecutions 
— Tertullian and Clement on luxury of 
Christians — Relations of Christianity with 
the Imperial Court in the Third Century — 
The Decian persecution and its effects — 
The Church before and under Diocletian — 
Social status of Church teachers — Result: 
membership of the Early Church not drawn 
mainly from the lowest, but from the inter- 
mediate classes, and embraced many of the 
wealthier and higher orders . . . -95 



The instreaming of Pagan influences on Chris- 
tianity has for its counterpart the out- 
streaming of Christian influences on Pagan 



society — These also ordinarily under-esti- 
mated — Silence of Pagan writers : what it 
means — Christianity and culture in the First 
Century — New Testament Epistles — Seneca 
and the Gospel — Rise and character of 
Apology in the Second Century — ^The literary 
attack on Christianity ; Celsus — Significance 
and spread of Gnosticism — The Pagan 
ethical revival in Second Century — Pagan 
preaching — Influence of Christianity on 
these — The Mysteries — The old Catholic 
Fathers — Rise of Neo-Platonism — Effects of 
Christianity on morals and legislation — Con- 
clusion 163 

Appendix 227 

Index 231 



New spirit in Early Church studies — Baur and his 
successors — Influence of Pagan environment on 
Christianity — Less attention given to the action 
outward of Christianity on Paganism — The 
spread of Christianity laterally, i.e., in respect of 
mere numbers, greater than ordinarily recog- 
nised — Estimates on this subject — Difficulties 
arising from fragmentariness of sources and 
unequal distribution of Christianity — The Cata- 
combs a new factor — Results from Catacomb 
discoveries — Comparison with New Testament 
and other data — Early progress of the Church — 
Christianity in Asia Minor — The Apologists, &c. 
— Carthage — Alexandria — Antioch — Gibbon's 
objections — Gaul and Spain — The final struggle 
— General result. 




IT is unnecessary at the commencement 
of these lectures to do more than refer 
to the changes which, within the last few 
decades, have taken place in the spirit and 
methods of the treatment of Church History. 
If there was a time within living memory 
when the charge could justly be brought 
against this branch of study of being the 

dreariest in the theological curriculum — a 



collection of dry bones and dead contro- 
versies — that time may confidently be said 
to have passed away ; and with it has disap- 
peared the idea that Church History must 
of necessity be an unprogressive science — the 
repetition of the old, unchanging story — 
seeing that the facts on which it is based 
must always remain precisely what they are. 
The changes referred to have come about not 
so much from the discovery of new materials 
— though of these also unremitting research 
has yielded an abundant supply — as from 
the new historical temper in which scholars 
have approached their task ; from the fresh 
power acquired of reading aright the mean- 
ing of the data already possessed, and of 
setting them in new lights and relations; 
from increased skill in colligating them, and 
in interpreting the significance of unnoticed 
details in their bearing on an entire situation 
— in which lies so much of the higher art of 


the historian. Just as the naturalist is re- 
puted to be able from a single bone to re- 
construct the form of some creature of the 
past, so our modern scholars aim at showing 
that the minutest fact is not isolated, but 
stands in organic relation with the all-per- 
vading life of the time ; and from comparison 
of the facts they seek to re-create for us a 
picture whose justification is its verisimili- 
tude, and its power of interpreting the sum- 
total of the phenomena. 

These gains which have accrued to Church 
History from the combined philosophical, 
historical, and critical movement of the last 
half century, have been reaped nowhere more 
largely than in the study of the earliest age 
of Christianity. The initial impulse here 
belongs indisputably to the school of Baur, 
which, however ruled by false pre-supposi- 
tions, and open to challenge in its con- 
clusions, has left on this whole field of 


investigation its deep and abiding impress. 
IfBaur'sown criticism has gradually had to 
retract itself within comparatively narrow 
limits, it may claim, like the Nile waters, to 
have fertilised in the height of its overflow 
even the plains from which subsequently it 
had to retreat. From Baur's day a new life 
entered into Early Church History studies. 
Ritschl, at first a disciple, then an opponent, 
undertook an independent investigation into 
the origin of the Old Catholic Church ; Light- 
foot, not without aid from Ritschl, re-dis- 
cussed the question of the Ministry, and 
cognate problems of the Apostolic age, but 
revealed also the unrivalled strength of his 
own scholarship in his handling of the litera- 
ture of the age next succeeding ; Hatch, 
freshest of English minds in this department, 
sought to show how Church ideas and usages 
took shape under the action of forces in the 
Gentile world ; Harnack and the later Ritsch- 


Hans have carried out more systematically 
the idea of the rise of ecclesiastical dogma 
through the importation of the ideas and 
methods of Greek philosophy ; Neumann and 
Ramsay discuss the relations of the Chris- 
tians to the Roman State, and the latter 
scholar has instituted a series of researches 
of his own, which mark a new era in the 
discussion of Apostolic and sub-Apostolic 
history. Other names, as Weizsacker's, will 
readily occur. From this re-digging of the 
soil in all directions and microscopic scrutiny 
of every fibre and detail of the relevant 
material, it is impossible to doubt that 
enormous advantage will result. 

There is, however, one aspect of this note- 
worthy revival of interest in Early Church 
History which the purpose of these lectures 
requires that I should now more particularly 
notice. It must strike the observant student 

— at least can hardly fail to do so when 



attention is called to it — that all this move- 
ment of mind in the direction of a better 
comprehension of the early development of 
the Church — of the manner in which it 
gradually shaped itself in policy, in doctrine, 
and in usages — is governed mainly by the 
idea of tracing the influence on Christianity of 
its Pagan environment — of that intellectual, 
moral, political, and religious environment, 
which constituted the world into which 
Christianity entered, and which could not 
from its very nature but powerfully act upon 
and modify the new faith ; but that the same 
attention has not been given to a phenomenon 
which is the counterpart of this, viz., the action 
outwards of Christianity on that Pagan en- 
vironment, altering, re-shaping, modifying it. 
I am, of course, well aware that the action of 
Christianity on Pagan society — on its ideas, 
laws, institutions, morals — has in many of its 
aspects formed the subject of learned inves- 


tigation.^ But I do not find that it has been 
taken much account of in this most recent 
phase of the study of Early Church History 
which I have specially in view. There has 
been much investigation into the modes and 
the results of the inflow of Pagan ideas and 
associations into Christianity, but there has 
not been the same carefulness in inquiring 
whether the flow was all on one side, whether, 
as is antecedently probable, there was not a 
current outward corresponding to the current 
inward — to borrow a term from science, an 
exosmose corresponding to the endosmose — 
and what the strength of this outward current 
might be. It has not been sufficiently per- 
ceived — at least so I venture to think — that 
precisely in the proportion that the progress 
of investigation requires us to postulate a 

* Such books may be referred to as Troplong's 
De linfluence du Christianisme sur le droit civil des 
Romains, Schmidt's Social Results of Early Christianity, 
Leckys European Morals, Brace's Gesta Chrisii, &c 


greater influence of Paganism on Christian 
ideas and institutions than has formerly been 
assumed, there arises the probability, nay, the 
certainty, that Christianity likewise was a 
factor of greater importance in the world of 
Paganism than had previously been imagined, 
and that traces of this influence are also to be 
discovered, if they are as diligently looked 
for. Action and reaction, in this as in other 
spheres, may be presumed to be equal ; and 
if the action is proved to be greater than 
former representations allowed, it may be 
anticipated that the reaction, in the case of a 
force of such undoubted magnitude, will 
prove to be greater as well. 

I am now in a position to explain with 
some definiteness the character of the thesis 
I propose to defend in these lectures. I 
think facts do exist — and many of them — to 
show that there really was this current out- 
wards of which I speak,. and that Christianity 


was actually a much more prominent factor 
in Pagan society than the ordinary repre- 
sentations would lead us to believe ; in other 
words, that just as the trend of investigation 
has been to show that there was a much 
greater influence of its Pagan environment 
upon the Church than has generally been 
conceded ; so, correspondingly, the direction 
of recent evidence has been to establish that 
the effects of Christianity on Pagan society, 
both extensively and intensively, were like- 
wise greater than has been admitted. I am 
fully conscious that in treating this subject 
I can say nothing that is new to scholars — 
little, perhaps, that is new to any one. The 
facts to which I am to refer are, most of 
them, sufficiently familiar — are, at the least, 
readily accessible ; but we have hourly evi- 
dence that it is possible for a fact to be 
familiar, and yet not to receive its due weight 
in the study of a subject. It may help to 


disarm criticism if I say that, in what I 
advance, I desire to disclaim anything like 
dogmatism. I put forth these ideas tenta- 
tively, and rather with the view of their 
being canvassed and checked by others, than 
as definitive conclusions of my own mind. 
Their end will be gained if they are in any 
degree provocative of reflection in those 
who may honour them with their attention. 
My treatment, which I should wish to be 
taken in its entirety, will be directed to 
show : — 

I. That Christianity had a larger exten- 
sion laterally^ t,e,^ in point of mere numbers, 
in the Roman Empire, than the ordinary 
representations allow. 

II. That it had a much larger extension 
vertically^ j>., as respects the different strata 
of society, than is commonly believed ; 
and — 

III. That it had a much greater influence 

iia«v aaw^ 


intensively ox penetratively y i,e,y in its effects 
on the thought and life of the age, than is 
generally acknowledged. 

The remaining part of this lecture will 
be devoted to the first of these topics. 


The extension of Christianity laterally 
or numerically in the Roman Empire. 

The attitude of mind of most historians 
on this question of the numerical extension 
of Christianity in the Roman world may 
be described as highly conservative. It is 
difficult to understand why this should be 
so, except that a prepossession in favour of 
a very moderate rate of increase having 
been engendered by the authority of certain 
great names, the feeling has established 
itself that this traditionally-received opinion 
ought not to be lightly disturbed. Whatever 
changes are assumed to be necessary in our 



conceptions of the relations of Christianity 
to Paganism in other respects, it is taken 
for granted with wonderful unanimity that 
there is neither room nor call for any revision 
of opinion here. Every one is familiar with 
Gibbon's estimate that the Christians in the 
time of Constantine constituted at most one- 
twentieth part of the population of Rome, 
and a like proportion of the whole subjects 
of the Empire.' Friedlander accepts and 
endorses this computation. ^ Chastel, a 
French writer, without, however, giving data, 

' Decline and Fall, ch. xv. Gibbon estimates the 
population of Rome at about 1,000,000, and gives the 
Christians one-twentieth of these, or about 50,000. 
The population of the Empire he takes (ch. ii.) to be 
about 120,000,000, which would give about 6,000,000 
Christians for the whole Empire. For other estimates 
of the population of Rome and the Empire, see 
V. Schultze's work referred to below, Untergang des 
Heidenihums, I. p. 9. Schultze computes 100,000,000 
for the Empire, and, "with greatest probability," 
600-810,000 for the Capital. 

"> Sittengeschichte Roms, III. p. 531. 



reckons the Christians at about one-twelfth 
of the population,' and this, or one-tenth, 
perhaps, represents the average opinion. 
Victor Schultze, one of the best informed 
of recent investigators, estimates the pro- 
portion at one-tenth, but with important 
qualifications which practically nullify his 
verdict. " This reckoning," he says, " remains 
at all events far behind the actual number. 
. . . The investigator assuredly gains from 
the testing of the sources in detail the clear 
impression that, in the beginning of the 
fourth century, the Church on the great 
world-theatre of over 103,000 geographical 
square miles numbered more than 10,000,000. 
It is hardly credible that the number of Jews 
at that time should have exceeded that of 

' One-fifteenth in the West, and one-tenth in the 
East. — Hist, de la destrucL du PaganismCf pp. 35-6. 
Chastel rejects Gibbon's computation as too low, and 
those of Staiidlin (one-half) and of Matter (one-fifth) 
as too high. 


the Christians." ' Others wisely decline to 
commit themselves to a precise estimate, 
still, however, usually with the presumption 
that the proportion was exceedingly small. 
Thus Uhlhorn scouts what he represents as 
Tertullian's statement that the Christians in 
a single province were more numerous than 
the whole Roman army, which, he says, as 
if it were an idea not for a moment to be 
entertained, would make about 9,000,000 
Christians in the Empire! 2 

In face of so weighty a consensus of 
authorities, I feel that it requires some 
courage to defend a different opinion. I 
am emboldened, however, by the considera- 

' Untergangdes Griesch.'Rom, Heidenthums, I. p. 23. 
Schultze is professor atGreifswald. 

' Conflict of Christianity (E.T.), p. 264. TertuUian, 
however, does not quite put the matter in the way 
stated {Apol, 37). Uhlhorn says elsewhere : " It is 
generally assumed that they formed about one- 
twelfth of the whole population in the East, and 
in the West about one-fifteenth " (p. 402). 


tion that in pleading for a much larger 
influence of Christianity numerically than 
these estimates allow, I do not stand abso- 
lutely alone. A few of the older writers, as 
Matter, put in a plea for one-fifth, or even 
a higher proportion, but their voices have 
scarcely been heard in the general chorus 
for a more moderate view. Still a tendency 
is beginning to manifest itself to a revision 
of the traditional estimate. Canon Robert- 
son, among recent historians, apparently 
leans to a proportion between one-tenth and 
one-fifth.^ And Keim, in his posthumous 
work, Rom und das Ckrtstentkum, expresses 
the belief that even at the close of the second 
century, the Christians were one-sixth of 
the population of the Empire.* G. Boissier, 
in his spirited book. La Fin du Paganisme^ 

* Hisi. of Churchy bk. i, ch. viii. 

» p. 419. " It is not sapng too much," he writes, 
"to name a siicth part of the Roman Empire 


speaks even more strongly on the arbitrari- 
ness of modern scholars, and their unwarrant- 
able rejection of evidence on this subject. 
After quoting the well-known passages from 
Tertullian, Pliny, and Tacitus on the wide 
diffusion of Christianity, he says : " This is 
precisely what they (the objectors) refuse to 
admit. In the first place, they will take no 
account of the affirmations of Tertullian. He 
was, they say, a rhetorician and a sectary, 
facts which ought to render him doubly 
suspected. It would be ridiculous to take 
seriously his fine phrases, and give his 
rhetorical amplifications the force of argu- 
ment. As for the letter of Pliny, and the 
passage in Tacitus, we have seen above that 
some do not believe them to be authentic, 
and the statements which they contain on 
the subject of the numbers of the Christians 
are one of the chief reasons alleged for 
rejecting them. There is found in them an 


exaggeration which betrays the forger, and 
appears altogether incredible. ... It is pro- 
claimed, finally, as a principle which needs 
no demonstration, that it is impossible that 
a religion should make such progress in so 
short a time. I confess that this confidence 
confounds me. Is it reasonable to settle in 
a word questions so obscure, so little under- 
stood P"^ Even V. Schultze, as we saw 
above, is not very sure of his ground, and 
declares that the reckoning he gives remains 
far behind the actual numbers. Elsewhere, 
indeed, he uses language which would imply 
that the Christians, at the beginning of the 
fourth century, might be one-fifth, or even 
more, of the population.^ 

' I. pp. 445-46. 

' Thus he speaks of the heathenism of the time as 
"over two-thirds of the population of the Empire " j 
again " as sixty or eighty millions out of one hundred 
millions " ; and again of the Christians soon after the 
Edict of Toleration as "at most one-fifth of the 
population of the Empire" (I. pp. 39, 59). 


Two things specially make it difficult to 
arrive at exact conclusions as to the number 
of Christians in the Roman Empire in this 
early period. One is the exceeding paucity * 
and fragmentariness of our sources of inform- 
ation ; the other is that the rate of progress 
in the different parts of the Empire was very 
unequal — much higher, e.g.^ in the East than 
in the West ; in Italy and North Africa than 
in a province like Gaul. " The imperfection 
of the record," as geologists would say, must 
ever be remembered. We shall find as we 
proceed abundant illustration of the danger 
of drawing wide inferences from isolated 
data, or of supposing that because nothing 
happens to be said of the progress of Chris- 
tianity in a particular district, therefore pro- 
gress was not being made. The second 
century, for instance, is already approaching 
its close before we get even a glimpse of the 
large and flourishing Church of Carthage, 


which, with the Church of Alexandria, then 
suddenly starts into visibility. ^ On the other 
hand, the rate of progress was undoubtedly 
very unequal, and even more instructive than 
the inequality of progress is the fact which 
furnishes the principal explanation of it It 
is characteristic of the advance of Chris- 
tianity that all through it struck at the great 
centres, and followed the great lines of inter- 
communication in the Roman world ; that its 
chief victories were won where Greek and 
Roman culture had prepared fhe way for it ; 
and that its posts of strength and influence 
were chiefly in the wealthy and populous 

' " Of the African Church before the close of the 
second century, when a flood of light is suddenly 
thrown up by the writings of Tertullian, we know 
absolutely nothing" (Lightfoot, Philippians, p. 224). 
Another example is Cyrene, where the size and 
adornment of the graves show the existence of a 
numerous and well-to-do community, of which we 
do not hear otherwise (Cf. V. Schultze, I. p. 21). 
In the troubles of the times this church afterwards 
fell into decay. 


cities — Rome, Corinth, Antioch, Alexandria, 
Carthage, Lyons, and the like — from which it 
could spread into, and best dominate, the sur- 
rounding districts.' Its method — the same fol- 
lowed by Paul in his missionary work — was to 
seize and occupy the leading vantage-points, 
with a view to an ultimate wider diffusion. 
Numbers, then, in a case of this kind, are 
assuredly not everything. As important as 
numbers was the way in which the numbers 
were distributed, and the spirit that animated 
them. It is ftot overlooked by the writers 
from whose opinions we shall have to dissent, 
that, though numerically so feeble, — as they 
regard the matter, — Christianity had yet, 
through its inherent spiritual energy, and 
ever-strengthening organisation, early made 
itself a factor of the first importance in the 
Roman Empire, — that, as Merivale says, 

' Cf . V. Schultze, I. p. 15 ; Ramsay, Church in 
Roman Empire, p. 147 (ist edit), &c. 


"The active and growing strength of the 
Roman world was truly theirs — theirs was 
the future of all civilised society." ' But the 
question is pertinent whether this acknow- 
ledged power of Christianity could have been 
exerted by the mere fraction of the popula- 
tion which they suppose the Christian Church 
to have been ; or whether the immense moral 
energy which, at the end of three centuries, 
and on the back of a prolonged and deadly 
persecution, raised the Church to a place 
of undisputed political supremacy in the 
Empire, does not of itself point to some 
fault in the numerical estimate. I cannot, of 
course, in a brief lecture, go into all the 
evidence. I can only take test cases, which 
fairly represent large areas, and may serve to 
illustrate principles. 

Now that there is need for some revisal 
of currently received notions on the rate 

' Epochs of Early Church History, p. 2. 



of progress of early Christianity is shown, I 
think, very convincingly by one branch of 
evidence, the full bearings of which on our 
subject seem as yet to be very imperfectly 
appreciated. I refer to the remarkable Cata- 
comb explorations of De Rossi and others in 
the present century. It is customary to dis- 
count the glowing testimonies of the second 
century Apologists, and of early Christian 
writers generally, on the score of rhetorical 
exaggeration ; but here, opened to us within 
recent years, is another book of surpassing 
interest, the pages of which are constantly 
being more clearly deciphered by skilled in- 
terpreters, and which promises to throw a 
flood of reliable light on just such problems 
as we are dealing with. It is surprising that 
these discoveries have not been made more 
use of by Church historians.' Their effect, I 

' Dr. Scha£F speaks of the importance of these dis- 
coveries, and notes the neglect of them by Church 


take it, must be largely to modify our ideas 
of the numbers of the Christians, and to 
compel the acknowledgment that they formed 
a much larger proportion of the population of 
the Empire than has hitherto been sus- 
pected. It will be convenient to take this 
new evidence first, then to ask how far it is 
corroborated or contradicted by the other 
evidence at our command. 

The Catacombs, as most are now aware 
are immense subterranean burial-places, ex- 
cavated in the soft volcanic tufa, near the 
great roads, within a radius of about three 
miles around Rome. There are certain 
facts regarding them which may now be 
regarded as definitely ascertained.' They 

historians. He himself gives a good account of 
them, but makes little use of their testimony in the 
body of his work. He mentions their witness to 
the numbers of the Christians, but does not well 
know what to make of it. — History of Church (Ante- 
Nic), Preface, and pp. 288, 295. 
' The name, of doubtful derivation, was originally 


are allowed to be Christian, and purely 
Christian cemeteries i; they are of enormous 
extent ; the number of the dead buried in 
them mounts up to millions ; the time 
allowed for this burial is about three cen- 
turies — in reality, little more than two 
centuries and a half, for the excavations 
had hardly begun before the second century 
and the numbers interred after the middle 
of the fourth century were small in propor- 
tion to those in the preceding period. After 
the sack of Rome by the Goths in A.D. 410, 
interment within them ceased. The excava- 

that of a territory adjacent to the cemetery of St. 
Sebastian, and only subsequently was extended 
to all the cemeteries. Over forty catacombs are 
enumerated — twenty -five or twenty-six greater, 
the rest smaller. For particulars see the works (in 
English) of Northcote and Brownlow, Lanciani, 
Withrow, Art. in Dictionary of Christian Antiquities, 
Black's Handbook to Christ and Ecc. Rome, &c. 

' Northcote and Brownlow, I. p. 376;' Diet, of 
Christ. Antiquities, I. p. 296; Northcote' s Epitaphs, 
pp. 22 £F. 


tions consist of galleries and chambers, some- 
times in descending levels of from three to 
five stories, and throughout their entire area 
are literally packed with graves, the dead 
being sometimes buried in the floors, as well 
as in the walls and rooms. What is not so 
certain is the precise figure to be put on their 
extent, or on the number of the dead interred 
in them. On these points estimates widely 
vary. The most careful and reliable calcula- 
tions are those of Michele Stefano de Rossi, 
brother and coadjutor of the famous explorer, 
who, on the basis of the exact measurement 
of six different catacombs, reckons the total 
length of the passages at 587 geographical 
miles.' As respects the numbers entombed, 

' See the details of measurements in Lanciani's 
Pagan and Christian Rome, p. 319. Northcote and 
Brownlow, Roma Sott, I. p. 2, give the apparently 
conflicting estimate of " more than 350 miles." But 
the appended words, "i.e., more than the whole 
length of Italy itself," show that the same calcula- 


we may set aside at once as fabulous the 
earlier computation of Father Marchi, who 
calculating the length of the passages at 
900 miles, reckoned- the dead interred in 
them at 7,000,000 !' A less extravagant 
calculation gives over 3,831,000, or nearly 
4,000,000 graves.2 Michele de Rossi, who 
has had most to do with the surveying and 
mapping out of these subterranean regions, 
adopting a more moderate multiplier, reaches 
a minimum of 1,752,000.3 Now let any one 

tion is intended, and the 350 is either a misprint for 
550, or there is some other confusion about the 

* Dr. Schaff amazingly accepts this computation of 
Father Marchi's in speaking of the number of the 
mart)rrs (I. p. 80), in complete contradiction of his 
calculation of the numbers of the Christians else- 

" See Withrow, p. 21. He mentions that "in the 
single crypt of St. Lucina, 100 ft. by 180 ft., De Rossi 
counted over 700 loculi, and estimated that nearly 
twice as many were destroyed, giving a total of 
2,000 graves -in this area." 

3 Lanciani, ut supra. 

m ■ —Hi 1^"^ 


reflect what computations like these imply. 
Taking the period of interment at its longest, 
we cannot count more than ten generations 
of Christians buried in the Catacombs. As- 
suming that the numbers are anything like 
what the above computations indicate, we 
obtain from them the basis of a simple, but suf- 
ficiently startling calculation. On the larger 
reckoning — that of about 4,000,000 graves — 
we lave a Christian population for one genera- 
tion, in and about Rome, of nearly 400,000 ; 
on the smaller computation, a population of 
about 175,000. But the system of averages 
in such a case is clearly misleading, for the 
number of Christians was undoubtedly small 
in the earlier generations, and reached its 
maximum towards the close of the era of 
persecution. Our number 175,000 would, 
therefore, represent rather a middle point, 
say, about the years A.D. 230 or 250. Com- 
pare this now with the calculation of Gibbon. 


He estimates, as we saw, the population of 
Rome at about 1,000,000, and the number of 
Christians at the beginning of the fourth 
century at about 50,000, or one-twentieth 
part of the whole. In reality, unless the 
testimony of the Catacombs has been 
totally misread, they might have been any- 
thing between one-third and one-half. And 
if Gibbon is right in supposing that the pro- 
portion throughout the Empire was analogous 
to that in Rome, it would be a very moderate 
computation indeed to regard it as one-fifth. 
In any case, it seems to me, there must be a 
heightening of the reckoning. Cut down the 
figures even to 1,000,000 and the proportion 
of Christians to the total population is still 
vastly greater than Gibbon allows. I think, 
therefore, I am justified in speaking of the 
Catacomb discoveries as a " neglected factor " 
in this study of Early Church History — one 
which only recently Church historians have 

-.i- ■ 


taken the trouble to refer to at all, and of 
the bearings of which even yet they show 
generally a most inadequate appreciation. 
I go on to ask whether there is anything 
in the other evidence in our possession 
which helps to corroborate, or which con- 
tradicts, the testimony they have yielded. 

I begin naturally with the Church of Rome, 
to which the Catacomb discoveries refer. Our 
witnesses here are the New Testament, and 
early secular and ecclesiastical history. And 
what do these witnesses tell us ? The facts 
above cited are sufficiently startling, but are 
they really more wonderful than the oldest 
notices we possess of the progress of the 
gospel in the Imperial City would lead us 
to expect? The origin of the Church in 
Rome is hidden in obscurity, but there are 
certain things about its early history which 
we do know very well. We know, for in- 
stance, that as early as a.d. 52, or little more 


that twenty years after the Ascension, accord- 
ing to the now almost universally received 
interpretation of the well-known passage in 
Suetonius (Judaos impulsore Chresto assidue 
iumultuantes Roma expulit)^ the disturbances 
in the Jewish quarters in Rome, arising from 
the disputes of Jews and Christians, were 
such as to lead the EmperOf Claudius to 
issue an 'edict for the banishment of all Jews 
from Rome ' ; we know that six years later 
(a.D. 58), and prior to his own visit to the 
city, the Apostle Paul wrote to the Church 
in this city one of his longest and most im- 
portant Epistles, in which he speaks of its 
faith as "proclaimed throughout the whole 

* Suet. Claud. 25 ; cf . Acts xviii. 2. " Probably the 
measures which the Emperor took against the 
Roman Jews had their origin in the continual dis- 
turbances arising from the strife between Jews and 
Christians" (Neumann, Der Romische Staai, p. 4). 
Thus most writers. Prof. Ramsay puts the date a 
little earlier, a.d. 50 {St. Paul the Traveller, ch. 
xi. 4). 


world " I ; we know that six years later again 
(a.d. 64), according to the testimony of the 
Roman historian Tacitus, the number of the 
Christians involved in Nero's persecution was 
" an immense multitude " (multitudo ingens) * ; 
while, towards the close of the century, the 
Roman Clement, referring to the same per- 
secution, uses an almost identical expression, 
speaking of" a great multitude " (iroXi icKvfioq) 
who had suffered for Christ.3 All this, be it 

' Rom. i. 8 ; of. Lightf oot, Philippians, p. 25. 

' Ann. XV. 44. Schultze remarks that the multitudo 
ingens is of those taken and condemned, and that 
there was certainly as great a number of women, 
children, and other Christians who were not brought 
to judgment (I. 9). Cf. Dr. Lightfoof s remarks on 
the attempts of modern critics to invalidate the force 
of this testimony of Tacitus (Ignatius, I. pp. 9-10). 

3 Epist, to Corinth, 6. The progress of Christianity 
was more than maintained during the rest of the 
century. " During the reigns of Vespasian and Titus, 
and in the early years of Domitian, there is every 
reason to believe that Christianity had made rapid 
advances in the metropolis of the world. In its 


remembered, within the first century, when 
Catacomb excavations had yet hardly begun. 
And what was going on in Rome in this 
century was, we have no reason to doubt, 
going on elsewhere. I cannot go into details, 
but would only ask whether Christianity must 
not have become an exceedingly powerful 
force before, in a city like Ephesus, for ex- 
ample, we could have the adepts in magical 
arts bringing their books, and burning them 
to the value of 50,000 pieces of silver ^ ; or 
have a riot like that instigated by Demetrius 
the silversmith, on the plea that not only the 
trade of shrine-making was brought into dis- 
repute, but the worship of the great goddess 
Artemis was in peril of being subverted " not 

great stronghold — the household of the Caesars — 
more especially its progress would be felt " (Light- 
foot, Clement, I. p. 27). 

' Acts xix. 1 8, 19. About ;£2,ooo of our money. 
" So mightily grew the word of the Lord and pre- 
vailed " (ver. 20). 


alone at Ephesus, but almost throughout all 
Asia " I ; or whether things must not have 
gone far to justify even a hyperbole like Paul's 
that the gospel had been "preached in all 
creation under heaven," and was " also in all 
the world bearing fruit and increasing " ^ j 
or, finally, to furnish a basis for the pictures 
in the Apocalypse of the "great multitude 
which no man could number, out of every 
nation, and of all tribes and peoples and 
tongues," which had " come out of the great 
tribulation " ? 3 

Darkness rests upon the closing decades 
of the first century, but we have the testimony 
from the early part of the next century of 
the newly recovered Apology of Aris tides ^ 
which relates how, after the Lord's Ascension, 
His twelve disciples "went forth into the 
known parts [Gr, provinces] of the world, 

' Vers. 23-27. » Col. i. 6, 23 (R.V.). 

3 Rev. vii. 9. Cf. Lightfoot, Philippians, p. 25. 


and taught concerning His greatness " ' ; and 
through the dim mists of tradition we can 
descry the figures of these Apostles spread 
over the various countries of the world — in 
Parthia and Scythia, on the bleak shores of 
the Euxine, in Mesopotamia, in Arabia, per- 
haps as far even as India — ^and can catch 
glimpses which show that the work of evan- 
gelisation was actively going on.^ When the 
curtain lifts again in the second century, we 
have many indications, direct and indirect, 
of the immense advances that had been made. 
I do not dwell on what Gibbon calls the 

' Apology, 2. The Apology has usually been 
ascribed, after Eusebius, to the reign of Hadrian 
(about A.D. 125-6) ; but a second title in the Syriac 
has led Prof. Rendel Harris and other scholars to 
place it later, under Antoninus. Mr. Armitage 
Robinson, however, prefers the older view {Texts 
and Studies, I. p. 75). 

' The best examination of these traditions is pro- 
bably that of Lipsius in his article on ''Acts of 
Apostles (Apocr3rphal)," in Smith and Wace's Did, of 
Christian Biog., I. pp. 17-32. 


"splendid exaggeration" of Justin Martyr, 
who declares: "For there is not one single 
race of men, whether barbarians, or Greeks, 
or whatever they may be called, nomads, or 
vagrants, or herdsmen dwelling in tents, 
among whom prayers and giving of thanks 
are not offered through the name of the 
Crucified Jesus," ' — though if one reflects that 
Justin does not claim that all the races or 
tribes he speaks of had been converted to the 
gospel, or were even preponderatingly Chris- 
tian, but only that the gospel had reached 
them, and had won from each its tribute of 
believers, the exaggeration need not be so 
great after all — but proceed at once to the 
examination of more particular evidence. 

^ Dial, cum Ttyph., c. 117; of. Gibbon, ch. xv. 
Mosheim remarks, '' There could have been no room 
for this v/sry exaggeration, had not the Christian 
religion at that time been most extensively diffused 
throughout the world" {Commentaries, I. p. 260, 


The case which naturally first arrests our 
attention here is that of Asia Minor, in 
regard to an extensive province of which 
we are exceptionally fortunate in having the 
unimpeachable testimony of Pliny, Pro- 
consul of Bithynia-Pontus, in his famous 
letter to the Emperor Trajan (a.d. 112). 
This letter of Pliny is a remarkable warning 
at the outset of the danger of reliance on 
the mere argument from silence; for, but 
for its existence, it is certain that the actual 
state of the case in this province would not 
have been known to us, and could hardly 
have been conjectured or believed.' Had 

' " Tertullian derived his knowledge of it from the 
correspondence of Pliny and Trajan ; Eusebius from 
^ Tertullian ; later Christian writers from Tertullian 
and Eusebius, one or both. The correspondence of 
a heathen writer is thus the sole ultimate chronicle 
of this important chapter in the sufferings of the 
Early Church " (Lightfoot, Ignatius, I. p. i8 ; of. 
Ramsay, Church in Roman Empire, pp. 146, ii^). 
Ramsay mentions that the text of the letter depends 


any Christian writer, e,g.y chosen to put it on 
record that at this early date in the second 
century the Christian religion was already 
professed by " many of all ages and ranks, 
and of both sexes " {multi ontnis CBtatiSy 
omnis ordinis^ utriusque sexus etiam) through- 
out this extensive region; that the move- 
ment was "not confined to the cities, but 
had spread into the villages and country" 
{neque enim civitates tantunty sed vicos 
etiam atque agros)\ that the temples of 
the heathen were "almost deserted" {prope 
jam desolata templd) ; that the " sacred 
rites" were "interrupted" {solemnia inter- 
missa); that the victims for sacrifice could 
find " very few purchasers " (rarissimus auctor 
inveniebatur) ; and that this had been going 
on for a long time {diu), I fancy we should 
have been disposed to accuse him, as we 

on a single manuscript, found in 1500, used in 1508, 
and never since seen or known, 



do Justin and TertuUian, of rhetorical 
exaggeration.! It is different when these 
statements appear in a cool official docu- 
ment, written expressly to obtain guidance 
in the discharge of proconsular duty. What 
makes the testimony of Pliny the more 
valuable is the wide extent of territory to 
which his witness applies. From recent 
authorities 2 we learn that the region over 
which Pliny's jurisdiction extended stretched 
far to the east ; that it included the district 
known as Pontus, and had in fact for its 
correct designation the Province of Bithynia 
and Pontus; and that Pliny was probably 
in the east or Pontic part of it when he 
wrote to Trajan. His description, there- 
fore, will include Pontus as well as other 
parts of the province. This is corroborated 

« The text of the letter may be seen in Lightfoofs 
Ignatius, and in works on Church History generally. 

' Mommsen, quoted by Ramsay, p. 224 ; Light- 
foof s Ignatius, p. 56. 


by Lucian, the friend of Celsus, who, writing 
after the middle of the century, makes his 
hero, Alexander of Abonotichus, describe 
his native country, Pontus, as "filled with 
Epicureans and Christians."' We have be- 
sides numerous notices of the Churches of 
Pontus in Eusebius.^ Yet Bithynia and 
Pontus, though already possessing Christian 
communities when Peter wrote his First 
Epistle, were not among the parts of Asia 
Minor favoured with Paul's labours. In 
this connection we come on another interest- 
ing fact — important, as again warning us 
against drawing wide inferences from frag- 
mentary data to the prejudice of the 
progress of Christianity. Here in the second 
century we have indubitable evidence that 
Pontus was swarming with Christians. Yet 

* Lucian, Alex,, c. 25. Quoted by Gibbon, Mos-. 
heim, Schultze, &c. 

* Bk. iv. 15, 23 ; V. i6, 23 ; viii. I2, &c. 


in the biography of Gregory Thaumaturgus, 
Origen's pupil, written by his namesake, 
Gregory of Nyssa — a century after his 
death it is true, and with many fabulous 
adornments — it is told that when Gregory 
was made Bishop of Neo-Caesarea, in Pontus, 
in A.D. 240 [not the Pontus of Pliny, but in 
its immediate neighbourhood], he found only 
seventeen Christians.^ It is added, by way 
of preserving the symmetry, that when he 
died he left only seventeen pagans ! The 
conversion of the district was thus accom- 
plished at least within the third century. 
But how unsafe it would be to argue from 
the alleged paucity of Christians in A.D. 240, 
in Pontus — which, if real, must have been 
due to local causes — to the general con- 
dition of Pontus earlier ! 

We are not yet done, however, with Asia 
Minor. There is no reason that I know of, 

' Migne, xlvi. p. 954. 



beyond the fact that we have not a Pliny to 
describe them, for supposing that what was 
true of Bithynia and Pontus, was not true of 
many other provinces as well. There is every 
reason for supposing that some of these pro- 
vinces were even more favourably situated. 
In Phrygia, e,g',, the seat of the outbreak of 
Montanism, Christianity must have been very 
strong. Renan says boldly : " In Hierapolis, 
and many towns of Phrygia, the Christians 
must have formed the majority of the popula- 
tion." I The recent discovery of the singularly 

^ MarC'Aurele, p. 449; so in his Saint Paul (chap, 
xiii.), "Phrygia was thenceforward, and remained 
for 300 years, a Christian country." Cf. Ramsay, 
Church in Roman Empire, pp. 146-7. Renan mentions 
that from the time of Septimus Severus, Apameia, 
of Phrygia, put on its coins a Biblical emblem, 
Noah's Ark. Prof. Ramsay takes this to be Jewish 
{Cities and Bishoprics of Phrygia, II. p. 670). This 
last-named work of Prof. Ramsay's contains much 
new matter and evidence from inscriptions on the 
power of Christianity in certain districts of Phrygia. 
(See specially vol. ii., chaps, xii. and xvii.) " To judge 


interesting epitaph of Abercius, of Hierapolis/ 
identified with the Avircius Marcellus, known 
to us from Eusebius as a prominent opponent 
of Montanism,* sheds light on the presence 
of Christianity in the Phrygian Pentapolis. 
Western Asia, with its historical churches, 
and traditions of St John, would not be far 

from the proportion of epitaphs, the population of 
Eumeneia in the third century was in great part 
Christian. . . . These facts show that Eumeneia was 
to a large extent a Christian city during the third 
century. ... It is clear that, for some reason, Chris- 
tianity spread to quite an extraordinary extent in 
Eumeneia and Apameia" (pp. 502, 511). 

' Cf. Lightfoot, Ignatius, I. pp. 447-86, and Prof. 
Ramsay (to whom belongs the chief merit of this 
discovery) in Cities and Bishoprics of Phrygia, II., 
chap. xvii. " All that is known of the history of 
the Pentapolis," he sayS) "centres round the name 
of Avircius Marcellus. He is presented to us as 
the most prominent Church leader in a district 
already permeated with Christian influence, and the 
chief figure in the resistance to Montanism in the 
latter part of the second century. . . . That the 
Pentapolis was Christianised very early is plain from 
the facts above stated" (pp. 709, 715). 

' Ecc. Hist,, V. i6. 


behind.' I do not touch on the vexed 
question of North or South Galatia — though 
renewed study has failed to reconcile me to 
what I conceive to be the superior impro- 
babilities of the hypothesis to which Prof. 
Ramsay has lent his powerful advocacy^ — 

' Cf . Lightf oot's essay on " The Later School of 
St. John," in Essays on Super, Religion, The Epistles 
of Ignatius, and scenes like the Martyrdom of 
Polycarp, are in evidence here to the existence of 
numerous and well-organised Christian churches. 

' Prof. Ramsay lays, perhaps, too much stress on 
the ignorance of Greek in North Galatia, and on the 
suppression of the Phrygian element in the popula- 
tion (e.g., Church in Roman Empire, pp. 79, 82, 99). 
Mommsen does not take so strong a view. A good 
portion of the inhabitants, he says, "must have 
descended from the older Phrygian inhabitants of 
these regions. Of still more weight is the fact that 
the zealous worship of the gods in Galatia and the 
priesthood there have nothing in common with the 
ritual institutions of the European Celts ; not merely 
was the Great Mother, whose sacred symbol the 
Romans of Hannibal's time asked and received from 
the Tolistobogi, of a Phrygian type, but her priests 
belonged in part at least to the Galatian nobility 
[* Phrygo-Galatic,' surely.] ... As the language of 


but in any case Ancyra, in Galatia, was a 
prominent Church in the middle of the second 
century,' and the gospel was, no doubt, 
diffused through the province then as else- 
where. In Cappadocia there was from early 
times a powerful Christian community. V. 
Schultze finds in the fourth century "the 
power of heathenism longer broken " in that 
province, " than anywhere else in the world," 
and " the majority of the inhabitants " of the 
chief city Christian.^ At the Council of Nice, 

conversation, the Celtic maintained its grounds with 
tenacity also in Asia, yet the Greek gradually gained 
the upper hand," &c. {Provinces of Roman Empire, 

I. p. 341). The topographical difficulties may be 
great, but, in our ignorance of the whole circum- 
stances, can hardly be said to be insuperable, and 
they are not all on one side. Prof. Ramsay's own 
pages show how precarious it is to argue from 
absence or paucity of inscriptions {Cities and Bishoprics, 

II. pp. 484,491, 499-502, &c.). The rival probabilities, 
however, cannot be discussed here. 

' Eus. v. 16. 

' Untergang des Heid,, II. p. 315. See the numerous 
references in Eusebius. 


in 325, the bishops of Asia Minor formed 
about a third of the total number.' 
Compare now these facts with Schultze's 
own estimate that the Christians in Asia 
Minor in Constantine's time may have 
amounted to 1,000,000, out of a total 
population of 19,000,000,^ or about one- 
nineteenth part, and it will probably be felt 
that the estimate is altogether inadequate. 
The proportion may more reasonably be 
assumed to be not dissimilar to that we 
have already met with in Rome. 

It may be held, however, that Asia Minor 
is an exceptionally favourable case for our 
argument, and this may well be so. Our 

' Cf. the list in Schultze, II. pp. 305-7. It may be 
noted that five bishops are named from churches 
in (North) Galatia, and seven from Pamphylia (with 
Lycia) — ^another region of which we are apt to infer 
from silence that it had no church-life of importance. 

' I. p. 18. Of course this is a minimum, and 
Schultze's important qualifications have to be 


knowledge of the state and progress of the 
Church during the greater part of the second 
century is exceedingly fragmentary ; yet 
putting together the pieces of evidence which 
we possess — the numerous references, e,g.^ to 
a chronic state of persecution in Justin and 
other Apologists ' ; the rise of Apology itself, 
as bearing witness to the growing importance 
of the Church, and its entrance into literary 
circles « ; the edicts and rescripts of emperors 
on the treatment of the Christians, some of 
them embracing a wide areas; the extra- 
ordinarily rapid development and propaga- 
tion of Gnostic errors — multiplying, as 

' See references in Lightfoot, Ignatius, I. pp.518 ff. 
* On this and the rise of Gnosticism, see Lecture IIL 
3 Those of Hadrian, Antoninus, and M. AureHus ; 
cf. Lightfoot, Ignatius, I. pp. 461 ff. The passage 
from Melito (Eus. iv. 26) on letters of Antoninus to 
the cities forbidding them to take any new measures 
against the Christians, "among the rest to the 
Larissasans, to the Thessalonians, to the Athenians, 
and to all the Greeks," deserves special notice. 


Irenaeus says, like mushrooms out of the 
ground ' ; the vivid light-flashes which 
illuminate such martyr-scenes as those of 
Ignatius of Antioch, of the venerable Poly- 
carp of Smyrna, of Justin and his com- 
panions at Rome, of the Martyrs of Vienne 
and Lyons,2 and serve to reveal how much 
there is to see if only we had light enough 
to see it — ^piecing these notices together, we 
may find good reason for believing that, 
with the exception of outlying regions like 
those of Gaul and Germany, the state of 
matters in the other parts of the Roman 
world was not essentially different from 
what we have already found. Celsus dreads 
the growing numbers of the Christians, and 
writes an elaborate book to ridicule and 
refute them 3 ; and Eusebius, looking back 

* Ad, Hccr., bk. i. 29. 

' Cf . Eus. iii. 36 ; iv. 15, 16 ; v. i ; and of. Lightfoot 
on incidents. 

3 Cf. IJhlhorn!^ Conflict of Christianity (E.T.), p. 279 ; 
and see Lecture III. 


upon the period, speaks of Christianity as 
spreading at this time so as to embrace 
the whole human race.' 

I do not wait on these general con- 
siderations, but pass to the close of the 
second century, when light fully returns to 
us, and we are able to see how immense a 
progress has been made. We are now in 
the age of the Old Catholic Fathers, and find 
ourselves confronted with a great and firmly- 
organised Church, claiming to be "Catholic 
and Apostolic," spread throughout the chief 
provinces of the Empire, and waging a 
victorious conflict with Gnosticism and 
Montanism on the one side, and Paganism 
on the other. Now for the first time the 
important Churches of Carthage and Alex- 
andria come fully into view. Only now, as 
formerly observed, do we gain a hint of the 
existence of a Church in the former of these 

' Eus. iv. 7. 


cities. It is as when a traveller, at a sharp 
turning of his road, suddenly finds himself 
in presence of a busy and populous city, the 
very name of which had previously been 
unknown to him. Yet both Churches are 
discovered to be already of long standing; 
both are wealthy and numerically prosperous ; 
both have to bear the brunt of the greater 
persecutions ; in both the crowds of the 
lapsed in the hour of trial reveal at once 
the high social position of many of the 
converts, and the nominal character of much 
of their profession. 

The situation in Carthage is vividly 
depicted for us in the pages of the fiery and 
eloquent Tertullian. There is no longer any 
mistaking the fact that, notwithstanding the 
numerical inferiority of the Christians, and 
the severe persecutions by which it was 
sought to restrain them, the flowing tide was 
with the new faith, and the heathen them- 


selves were profoundly alarmed at its progress. 
" Men cry out," says the Apologist, " that the 
State is besieged; the Christians are in the 
fields, in the forts, in the islands ; they mourn, 
as for a loss, that every sex, age, condition, 
and even rank, is going over to this sect " ' ; 
and he tells us, in language resembling 
Pliny's, " the temple revenues are every day 
falling off; how few now throw in a con- 
tribution." 2 This is rhetoric, no doubt, but 
we do not feel it necessary to set it aside 
wholly as the language of exaggeration. It 
is only what the facts before us would lead 
us to expect. In his address to the Proconsul 
Scapula in deprecation of the persecutions 
of the Christians — to a man, therefore, who 
must have been able to detect any extrava- 
gant exaggeration — ^Tertullian asserts boldly, 
"Though our numbers are so great — am- 
stituting all but^ majority in every city (J>ars 
' Apol I, » C. 42. 


pene major civitatis cujusque) — we conduct 
ourselves in quietness and modesty " ^ ; and 
says again that if the Christians in Carthage 
were to present themselves in a body before 
his tribunal, he would have to decimate the 
city to make an example of them.^ Ter- 
tullian, we allow again, was a rhetorician, 
but there was at least some method in his 
rhetoric, and it would plainly have defeated 
the very end he had in view had he addressed 
an appeal to a proconsul, intended to influence 
his action, which was on the face of it 
monstrously and even ludicrously at variance 
with the facts. It is in the light of state- 

' Ad, Scap. 2. 

» C. 5. If one wishes to press the "decimate" 
of the rhetorician, as implying that the numbers 
whom he had just described as "all but a majority 
in every city," constituted literally and exactly one- 
tenth, it may be observed that it is only the adult 
part of the community that would figure in such 
a scene as this, and all need not be conceived of as 


ments like these, where a certain caution — 
not to say verisimilitude — must have been 
employed, that we must judge of his other 
celebrated outburst in the Apology, "We 
are but of yesterday, and yet we have filled 
every place belonging to you — cities, islands, 
castles, towns, assemblies, your very camps, 
your tribes, companies, palace, senate, forum 
— we leave you your temples only."^ Neu- 
mann justly remarks that, to prove effective, 
such statements must have been able to 
attach themselves to known facts.^ It is a 

' C. 37. 

' Der Romische Staat, p. 121. So Hasenclever, in 
his able articles on the social rank of the early Chris- 
tians in Jahr.f, ProL Theol.,wm.f speaks of Tertullian's 
rhetorical exaggeration as "essentially agreeable to 
the fact " (p. 36). There is more reason for supposing 
that Tertullian gives wings to his imagination in his 
enumeration of the various nations that had received 
the gospel {Ad. Jud, 7) ; but even this passage is 
only an expansion of similar statements in Justin and 
Irenaeus. The latter Father speaks of the Church in 
his day as "dispersed throughout the whole world, 
even to the ends of the earth," and enumerates the 


very moderate interpretation, I think, to put 
upon them to say that the Christians by the 
end of the second century must have formed 
one-fifth or one-sixth of the population of 
the North African cities, and perhaps one- 
tenth of the whole province. Schultze, on 
the other hand, computes as a minimum only 
50,000 out of a population of half a million 
in the city, or one-tenth, $ind a total of 
100,000 in the province, or about i per 
cent! I A totally inadequate estimate, as 
it seems to me. But if this region had one- 
tenth of a Christian population at the end of 
the second century, the proportion must have 
been vastly greater by the end of the third 
century, after the long period of peace, during 
which, as we know from Eusebius,^ the Church 
was progressing by "leaps and bounds." 

Churches in Germany, Spain, Gaul, in the East, in 
Eg3rpt, in Libya, &c., as agreeing in the symbol of 
the faith (Adv. Hcer, i. 7). 
* Untergang, p. 4. « VIII. i. 



This IS borne out by the glimpses we get 
of the affluence of the Church in the times 
of Cyprian and Constantine ; by the very 
magnitude and importance of its schisms — 
notably the Donatist; and by the extra- 
ordinary number of its bishops.^ 

The situation was not widely different in 
Alexandria. From the appearance of the 
developed Gnosticism of Basilides and 
Valentinus in the early part of the century, 
we know that Christianity must early have 
taken hold upon this famous city — as motley 
in its habits of thought as in its population ; 
and we have an incidental corroboration of 
this in a curious letter of the versatile 
Emperor Hadrian, whose satiric strain throws 
an interesting light on the strangely mixed 
state of affairs which prevailed. " I have 

' In the year 330 the Donatists alone could bring 
together a Synod of 270 bishops. See further on the 
Church of Carthage in Lect. II. 


found the people," he says, " vain, fickle, and 
shifting, with every breath of opinion. Those 
who worship Serapis are in fact Christians, 
and they who call themselves Christian 
bishops are actually worshippers of Serapis. 
. . . The patriarch himself, when he comes 
to Egypt, is compelled by one party to 
worship Serapis, by the other Christ." ^ 
Here is already a very marked diffusion of 
Christianity in Alexandria in the first quarter 
of the second century, and a corresponding 
interest of the emperors in it. While naming 
Hadrian, I may simply quote for what it is 
worth the remarkable statement made regard- 
ing him by the heathen writer Lampridius, viz., 
that he at one time contemplated dedicating 
to Christ statues without temples which he 
had caused to be erected in every city, and 
was only deterred by the consideration " that 

' Letter to Servianus. See text in Lightfoot, 
Ignatius, I. p. 464. 


the temples of the old gods would become 
deserted, and all the people would become 
Christian " (pmnes Christianos futuros)^ Even 
if erroneous, the statement is a striking 
testimony to the impression which Chris- 
tianity must have produced on the Hadrianic 

The position of the Church in Alexandria 
was greatly strengthened by the rise of its 
famous Catechetical School. It is right, 
however, that I should say a word at this 
point on the testimony of Origen, who, from 
Gibbon downwards, has been cited as a 
counter-witness that the number of Christians 
in his days were "very few." It would be 
strange indeed if Origen, in the middle of 
the third century, should make any such 
assertion, and we shall find, I think, that it 
is forcing his language to put this interpreta- 

* Lamprid. in Sev, Alex., 43 ; cf. Lightfoot, Igna- 
tius, I. p. 441 ; Lanciani, Pa^. and Christ. Rome, p. 11. 


tion upon it. The passage in question is in 
the eighth book against Celsus, where Origen 
is urging the blessing that would accrue to 
the Roman Empire if men universally acted 
on the spirit of Christ's precept, "If two of 
you shall agree on earth as touching any- 
thing that they shall ask," &c. "What 
might we expect," he says, "if not only a 
very few (iravv oXtyoi) agree, as at present, 
but the whole of the Empire of Rome ? " ' 
It is plainly unfair, especially in view of 
Origen's own declarations elsewhere,^ to 

* Origen, Contra Celsutn, viii. 69. 

= In bk. viii. 14, eg., Origen speaks of the " multi- 
tude" of believers in the Church. He cites Celsus 
(iii. 10) to the effect that when the Christians were 
few in number they held the same opinion, but when 
they became " a great multitude " they were divided 
and separated ; and he says " That Christians at first 
were few in number is undoubted, in comparison 
with the multitudes who subsequently became Chris- 
tians ; and yet, all things considered, they were not 
(even then) so very few." (Cf. iii. 9, 29, and see 
below.) Of course the numerical inferiority of the 
Christians to the Pagans was still striking, even if 


Strain a passage like this, which relates to 
agreement in prayer, into a statistical decla- 
ration of the number of Christians in the 
Empire. Remembering the distinction be- 
tween real and nominal Christianity, we 
might say with some justice that even in 
Britain and America at the present day 
" very few " agree. The truth is that Origen 
is one of the most explicit witnesses we 
possess to the victorious progress which 
Christianity was making in the Empire. In 
that very eighth book to which reference is 
made, and in the same context, he expresses 
his triumphant confidence, despite of oppres- 
sion and persecution, that Christianity is the 
power destined to overcome every other. 
" Every form of worship," he says, " will be 
destroyed except the religion of Christ, which 
will alone prevail. And indeed it will one 

the former composed one-fourth or one-fifth of the 
population of the cities. 


day triumph, as its principles take possession 
of the minds of men more and more every 
day." I And repeatedly in the course of his 
work he bears witness to the all-conquering 
power of the truth. " It proved victorious," 
he says, " as being the word of God, the very 
nature of which is such that it cannot be 
hindered ; and becoming more powerful than 
all such adversaries, it made itself master of 
the whole of Greece and a considerable 
portion of barbarian lands, and converted 
countless numbers (myriads) to his religion." ^ 
This is a considerably different picture from 

' Bk. viii. 68. This, it has been remarked, is a 
new idea, remarkably opposed to the tone of the 
earlier writers, who always look on the Roman power 
as hostile and persecuting, an oppression from which 
there could be no deliverance except through the 
coming of the end. 

^ Bk. i. 27 ; cf. ii. 13 : " For the Word, spoken 
with power, has gained the mastery over men of 
all sorts of nature, and it is impossible to see any 
race of men which has escaped accepting the teach- 
ing of Jesus"; and iii. 24: "He (Celsus) cannot 


the " very few," to which we are wont to be 
referred as Origen's sole testimony! The 
most convincing evidence, perhaps, of the 
enormous progress the gospel must have 
made in Alexandria and similar great cities 
by the time of Constantine is the way in 
which, after the victory of Christianity, the 
conflicts of Christians and Pagans seem to 
sink into the background, while the stage is 
filled with new disputants. Catholic and 
Donatist, Orthodox and Arian, in whose 
disputes the very heathen take their share, 
ridiculing them in the theatres, and discuss- 
ing them in the baths, shops, and streets. ^ 

demonstrate that an unspeakable number, as he 
asserts, of Greeks and Barbarians acknowledge the 
existence of iEsculapius ; while we, if we deem this 
a matter of importance, can clearly show a countless 
multitude of Greeks and Barbarians who acknow- 
ledge the existence of Jesus." (Also vii. 26.) 

* Cf. Histories of Socrates (i. 6, 8 ; ii. 2) ; Theod., 
(i. 6), &c. See passages collected in Newman's Arians, 
Note V. 


Yet in this great and populous city — the 
second in the Empire — Schultze will only 
grant a minimum of some 50,000 Chris- 
tians, or one-twelfth of its 600,000 inhabi- 
tants.i Surely, again, an immense under- 
statement ! 

From Carthage and Alexandria our eyes 
turn to a third great centre — Antioch, the 
gay and voluptuous capital of Syria, which, 
in point of population and influence, stood 
only behind Alexandria. Apart from the 
list of its bishops and other slight notices in 
Eusebius,^ the Church of this city, renowned 
in the earliest age as the Mother Church of 
Gentile Christianity, is another of those that 
only come late into view. When it does 

' Pp. 2c>-i. He allows 150,000 for Eg3rpt and 
Lib}^. On the opulence of the Alexandrian Church, 
see next lecture. 

" These, however, suffice to show an important 
Church. The Epistles of Ignatius alone are evidence 
of this. 


become distinctly visible in the middle of 
the third century, it is as a seat of eccle- 
siastical influence of the first rank. The 
extraordinary splendour of its episcopate, 
and elaboration of its Church service, under 
the notorious Paul of Samosata ^ ; its influen- 
tial councils and important theological 
school ; the magnificent Golden Church 
reared later by the liberality of Constan- 
tine 2 ; its prominence in the Arian contro- 
versies ; the utter failure of Julian's attempt 
to restore Paganism in it — readers of Church 
History will remember his chagrin when, 
having gone to celebrate with all pomp the 
festival of Apollo at the Temple of Daphne, 
he found only a single old priest, sacrificing a 
goose at his own expense 3 ; the flourishing 

' Cf. Eus. vii. 30. 

» Eus., Life of Constantine, iii. 50 ; Orat. 9. " A 
church of unparalleled size and beauty." Its dedica- 
tion in A.D. 341 was the occasion of a Council famous 
in the Arian strife. 

3 Julian himself relates the incident in his Misopogon, 


State of the Church, numerically at least, 
under Chrysostom — all this shows that, even 
before the change of the political relations, 
Christianity must have been practically in 
the ascendant in the city. " Probably al- 
ready (in first half of fourth century)," says 
V. Schultze, "at all events soon after, the 
majority and power were in the possession of 
the Christians." i We have the express testi- 
mony of Chrysostom that in his day the 
Christians were a majority in the city (to 
irXiov Trig ttoXcwc Xpi(maviK6v) ^ ; and this is 
borne out by the separate figures he gives, 
showing the population to have been 2oo,ooo,3 
and the number of the Christian community 
about 100,000.4 But even these figures pro- 

' Untergang, II. p. 261. ' Adv, Jud. horn,, i. 4. 

3 Horn, in St, Ignat., 4. 

* Horn, in Matt, Ixxxv. (Ixxxvi.) 4. On the deba- 
table elements in these figures, see Gibbon, eh. xv., 
and V. Schultze, II. p. 263. These writers suppose 
that Chrysostom reckons only citizens of Antioch — i.e., 


bably do not do justice to the strength of 
Christianity in the Syrian capital. In a vast 
city like Antioch, as in other great centres, 
there must have been a large floating mass 
favourably disposed to Christianity, who yet 
never definitely connected themselves with 
the Church. What it is still more important 
to remember, Antioch at this time was a 
city deeply rent with schisms. There were, 
in fact, in Chrysostom's day, no fewer than 
three rival parties in it, two at least with 
separate organisations, all holding eccle- 
siastically aloof from each other — the Catholic, 
the Arian, and the Meletian.''^ It is an 
interesting question whether in his 100,000 

not children or slaves — and Gibbon would raise the 
number of inhabitants to half a million. Be that as it 
may, Chrysostom's statements, which V. Schultze 
accepts, as to the relative proportions, are not to be 
set aside. 

' "The unhappy Church" was thus, as Mr. 
Stephens says, "torn to tatters" {Saint Chrysostom, 
pp. 21, 140). 


Christians he reckoned, as his language 
would seem to imply,^ only the frequenters 
of his own cathedral, or the schismatics as 
well. In any case, we have his explicit 
statement that the Christian community in 
Antioch outnumbered the Jews and Pagans 
combined ; and this reflects light back on the 
prosperous condition of the Church in the 
beginning of the century. It is a natural 
but hasty assumption that the mere change 
in imperial favour constituted a reason why, 
in such a city, those who had been previously 
Pagan should at once rush into the arms of 
the Church. Imperial favour would no doubt 
have its effect, but the city did not change 
its creed at the bidding of Julian, and there 
is no reason to suppose it would have become 

' He speaks of " those assembling there " (roi^g 
kvravQa avvayofuvovo), ue,, at the church. This is to be 
borne in mind, if it should be the case (see above), 
that the 200,000 in Antioch only represent the citizen 
element in the population. 


Christian at the bidding of Constantine had 
the public mind not otherwise been prepared 
to receive the new faith. 

The case of Antioch just cited has a bear- 
ing on our subject in another way; has, in 
fact, been used by Gibbon for the directly 
opposite purpose of establishing his low 
estimate of the numbers of Christians in 
Rome and in the Empire. Unwarrantably, 
as I think, in face of Chrysostom's statements, 
reckoning the Christians in Antioch at one- 
fifth, instead of one-half, of the population, he 
makes this the basis of a calculation to show 
that the proportion elsewhere is, as he repre- 
sents it, one-twentieth of the whole.' Two 
sets of data are combined in his argument — 
one, a notice in Chrysostom that in Antioch 
3,000 widows and virgins were supported by 
the bounty of the Church » ; and the other, a 
statement in Eusebius that in the year 250 

* Ch. XV. » Horn, in Matt. Ixvi. (Ixvii.) 3. 


A.D., the clergy of Rome consisted of a bishop, 
46 presbyters, 7 deacons, as many sub- 
deacons, with 42 acolytes, and 52 readers, 
exorcists, and porters, while the number of 
widows and poor supported by the Church 
was over 1,500.^ This small number of the 
clergy, and the fact that the numbers sup- 
ported by charity are half those in Antioch, 
with its Church community of 100,000, are 
held to justify the inference that the Church 
in Rome must have had about 50,000 mem- 
bers. But it must surely be felt that we are 
on very precarious ground indeed in reason- 
ing from a bare enumeration of the clergy 
and poor to the strength of the Christian 
community in Rome, especially when Cor- 
nelius, the author of the above enumeration, 
in the same breath speaks of " the very great, 
even innumerable people" (juyt<rrov icai avapid- 
fxirrov Xaov) whom these clergy served ! We 

« Ecc, Hist vL 43. 


know far too little of the ecclesiastical ar- 
rangements of the Roman Church — of the 
number of its parishes, and of the manner in 
which the work of these parishes was over- 
taken by the clergy,' to be able to hazard 
even a guess at the ratio of the clergy to the 
body of the people. In Antioch itself, where 
the Christians were at least 100,000 strong, 
we have no hint of a division into districts 
with separate churches at all ; the preaching, 
so far as would appear, was done, chiefly by 
Chrysostom, at the one great cathedral* The 
great church at Constantinople, again, in the 
time of Justinian, when practically all were 

* We have the testimony of Optatus of Mileve that 
there were forty churches in Rome before the last 
persecution. Cf. Bingham, III. p. 133. But we have 
only to reflect on the number of parochial churches 
in some of our large continental cities still — a dozen 
or little more to an immense population — to see that 
this is no index to the numbers of the Christian 

' Mr. Stephen points this out in his Saint Chrysostom, 
p. 108. 


nominally Christian, was served, as we learn 
from Gibbon himself,' by but sixty presbyters. 
It is at least equally precarious to reason from 
the numbers of widows and poor in one com- 
munity to those in another, without knowing 
the precise circumstances of each and their 
respective modes of administering help. But 
apart from all other considerations, the Cata- 
comb discoveries already adverted to seem 
to demonstrate the baselessness of Gibbon's 

Our materials are scantier, when from these 
large and flourishing churches of the East, 
we turn to the West, and inquire concern- 
ing the progress of the gospel in Gaul and 
Spain. The origin of the Church in Gaul is 
wrapped in obscurity, and we need not seek 
to penetrate the mists of legend which en- 
shroud it The first real glimpse we get of it 
is in the beautiful and pathetic narrative of 

' Ch. XX., footnote to No. 2, on Clergy. 



the martyrdoms at Vienne and Lyons, under 
Marcus Aurelius, in the year 177 A.D.' We 
have only to think, however, of what that 
single glimpse reveals, to be satisfied that 
the Kingdom of Heaven, which is like 
leaven, was silently and secretly diffusing 
itself in Gaul, as everywhere else. Dean 
Milman speaks too sang^inely, perhaps, 
when he says of this reign of Marcus, 
"The western provinces, Gaul and Africa, 
rivalled the East in the number, if not in 
the opulence of their Christian congrega- 
tions. In almost every city had gradually 
arisen a separate community," « but we may 
safely conclude that our temptation is to 
minimise, rather than to magnify, what had 
actually been accomplished. The Church at 
Lyons, at the head of which Irenaeus now 
takes his place as bishop, shows how Chris- 
tianity had established itself at a point of 

' Eus. V. I., 2. » Hist, of Christ, bk. ii. ch. 7. 


vantage, the value of which, for purposes of 
aggression, can hardly be over-estimated. 
Only when, with the aid of a description like 
Renan's,^ we picture to ourselves the super- 
lative importance of Lyons as a political 
religious, and commercial centre, and realise 
its geographical advantage as situated at the 
junction of the Rhone and the Sa6ne, do we 
appreciate how much was secured by the 
gospel having attained, even at the cost of 
fearful sufferings, a firm footing in its midst. 
Neither does it follow that, because Chris- 
tianity was late in taking root in Gaul, its 
progress, once it had established itself, was 
not as rapid there as elsewhere.^ Unfor- 
tunately, our means of information are so 
slight that we are unable to trace its 
advances in detail. We have a rhetorical 

* Marc-Aurhle, ch. xix. Cf. Schultze, Untergang, II* 
pp. iia-12 : "The capital of the Three Gauls." 

' The culture conditions were exceptionally favour- 
able. Cf. Schultze, II. p. loi. 


allusion in Tertullian ^ ; we glean scattered 
hints of congregations in the third century ; 
we find Constantius and Constantine protect- 
ing the Christians in Gaul in the Diocletian 


persecution 2; we have the important Synod in 
Aries in 314 A.D., another Church coming 
suddenly into view 3 ; we have notices thus 
early of numerous bishoprics 4 ; while by the 
middle of the fourth century we find an epis- 
copal organisation in all the provinces of the 
country.^ But the best proof of all of the 
rapid march of the gospel to victory is the 
fact which Schultze emphasises, of the com- 
plete overthrow of the classical Paganism in 
the cities throughout all Gaul in the course of 

' Adv, Jud, 7. 

" Eus., Life of Constantine, i. 8, 15, 16, &c. 

3 Cf. Hefele, II. pp. 180 ff. (E. T.) "The Synod of 
Aries/' says Schultze, " shows that Christianity had 
gone deep into the community" (II. p. 13). 

4 Schultze, Untergang, I. p. 12. 

5 Ibid. II. p. 103. 


the fourth century.^ How this victory was 
brought about we can only imperfectly discern 
in the labours of such outstanding individuals 
as Hilary of Poitiers and Martin of Tours, 


but the decisiveness and completeness of the 

transition is as remarkable as anything in 


If, in this paucity of evidence, it be thought 

that we assume too much in supposing a 

silent but steady progress of the gospel 

through Gaul in the third century, an apt 

instance to rebuke our scepticism comes 

from the neighbouring Church of Spain. 

The early history of the Church in Spain 

is even more completely hidden from us 

than that of the Church in Gaul. The sum- 

* Untergang, II. pp. 103, 1 13-14, &c. The Celtic 
heathenism lingered longer, though also in the 
main overcome. Mommsen (quoted by Schultze, 
II. p. 116) says, " More rapidly still than the (Celtic) 
native speech, the native religion lost ground, and 
Christianity, pressing in, found in it scarcely any 


total of our knowledge of it during the first 
three centuries is comprised in casual allu- 
sions in Irenaeus and TertuUian,' and in an 
Epistle of Cyprian which relates to the 
deposition of two Spanish bishops.^ The 
curtain lifts, as usual suddenly, at the Council 
of Elvira, A.D. 305 or 306 — a, Council whose 
81 Canons still remain to us.3 And what do 
we find then ? A Church long rooted in the 
land, with splendid basilicas, reckoning in its 
membership great landowners, and numerous 
magistrates, and high civic dignitaries, while 
an eagerness is shown on the part of all 
sections of the community — including those 
whose occupations are most questionable — 

* Iren. i. 10 ; Tert. Adv. Jud. 7. 

* Epistle 67. The Epistle is, however, of import- 
ance as showing the sufferings of the Spanish Church 
in the Decian Persecution. 

3 See in Hefele, 1. pp. 180 ff. The Council of 
Elvira, with its 19 bishops and 24 presbyters, is only 
to be regarded as representing the South of Spain. 
Cf. Schultze, I. p. 5. 


to be received into its communion. Hosius, 
the Bishop of Cordova, was the trusted adviser 
of Constantine, and a leading figure in the 
Nicene Council. 

This brings us to the final struggle — the 
persecution inaugurated by Diocletian — from 
the midst of which come some remarkable 
testimonies, with the mention of which I may 
fitly conclude this portion of the argument 
The Church had been at peace for forty 
years, and its progress during that period 
had been extraordinarily rapid. " Who could 
describe," says Eusebius, "those vast col- 
lections of men that flocked to the religion 
of Christ, and those multitudes crowding in 
from every city, and the illustrious concourse 
in the houses of worship ?" ' Then the perse- 
cution burst, sifting the ranks of the Church, 
and scattering nominal professors before it, as 
a storm breaking through a forest makes the 

' Ecc, Histt viii. i ; Cf . Lect. III. 


leaves fly in every direction. The incidents 
of the persecution — which we do not stay 
to describe — and other notices of the time, 
show how widely Christianity must have 
been spread. We read, for example, of a 
town in Phrygia burned with all its popula- 
tion, including women and children, because 
the inhabitants, those in high rank as well 
as persons of humbler station, confessed 
themselves Christians, and would not recant!' 
We should notice also the case of Armenia, 
which at this time received the gospel from 
Gregory the Illuminator, and that so de- 
cidedly that probably two -thirds of the 
people may be reckoned as professing Chris- 
tianity." Incidentally, we get a glimpse of 
the predominance of Christianity in Rome 
in the fact, narrated by Eusebius, that 
Maxentius, the usurper of imperial power 
in Italy during the persecution, sought 
" Ecc, Hist, viii. ii. ' Cf. Schultze, I. p. 17. 


to ingratiate himself with the Romans by 
pretending that he was of the Christian faith. 
"He pretended," says the historian, "by a 
species of accommodation and flattery to- 
wards the Romans, that he was of our 
faith." ^ But there are yet more striking 
testimonies. The complete failure of the 
persecution may be said to be itself such 
a testimony, as, indeed, the persecutors them- 
selves, in their edicts and rescripts of tolera- 
tion, acknowledge.^ Maximin, Emperor of 
the East, was the most obstinate and cruel, 
perhaps, of all these persecutors. And what 
does he say ? In an Epistle to his governors 
ordering the persecution to cease, he gives 
as the reason why it had been under- 
taken actually this — that the emperors ^* had 
seen that almost all men (ax^Sov aTravrac 
aydpcuTTovc) ^^^^ abandoning the worship of 
the godSy and attaching themselves to the party 
' Eus. viii. 14. ' Ibid. viii. 16, 17 ; ix. 9. 


(Wvh) of the Christians^ ^ Rhetoric may be 
charged against Tertullian, but a procla- 
mation of this kind is hardly one in which 
we should look for rhetorical exaggeration ! 
The only other testimony I shall adduce is 
one from the Christian side. It is that of 
the famous Lucian of Antioch, teacher of 
Arius, and founder of the school usually 
known as the Antiochian, who perished in the 
persecution. And the words of Lucian are 
that, prior to the last persecution, ^^ almost 
the greater part of the worlds including 
whole cities^ had yielded obedience to the 
truth " (" pars paene mundi jam major 
huic veritati adstipulatur ; urbes integrae ").* 

These are utterances from the midst of 

' Eus. ix. 9. 

° The passage is only in the Latin translation of 
Eusebius by Rufinus, but Dr. Milman thinks it 
authentic; cf. Hist, of Christ, Bk. iii., ch. i. He 
quotes also a note from Routh, who gives on the 
authority of Porson a statement from Porphyry, 
that the Christians were roifs nXiiovas — a majority. 


the conflict, not easily to be explained 
away by those who would persuade us 
that the Christians constituted only an 
insignificant one-twentieth, or even one-tenth 
of the population, at the time of the victory ; 
and they surely warrant us in holding that 
there has been an undue timidity in recog- 
nising the powerful hold which Christianity 
had taken, numerically, on society by the end 
of the third century. If we allow the simple 
facts of the case to produce their natural 
impression on our minds our verdict, I think, 
must be — there are factors here which have 
been neglected. 





Influence of Christianity on the higher ranks of 
society under-estimated— New Testament evi- 
dence — ^Witness of the Catacombs — Pomponia 
Grsecina — Flavins Clemens and Domitilla — Aci- 
lius Glabrio — Notices in Second Century — ^The 
wealth of the Church of Rome — ^The witness of 
the persecutions — ^TertuUian and Clement on 
luxury of Christians— Relations of Christianity 
with the Imperial Court in the Third Century — 
The Decian persecution and its effects — The 
Church before and under Diocletian — Social 
status of Church teachers— -Result : membership 
of the Early Church not drawn mainly from the 
lowest, but from the intermediate classes, and 
embraced many of the wealthier and higher 




IN the previous lecture I defended the 
position that Christianity in the early 
centuries had manifested an energy of propa- 
gation, and diffused itself with a rapidity 
much greater than the majority of Church 
historians seem prepared to allow ; I am now 
to seek to strengthen this position by taking 
society, as it were, in vertical section, and 
inquiring into the degree in which Chris- 
tianity can be shown to have affected the 



wealthier and better-educated classes in the 
Empire, as well as those of inferior social 
station. Here also, I think, the influence of 
the gospel has generally been under-esti- 
mated. It may be going too far to say, 
with Prof Ramsay, that Christianity " spread 
at first among the educated more rapidly 
than among the uneducated " ' ; but I am 
persuaded that even this is nearer the truth 
than the opinion often expressed that Chris- 
tianity drew the great bulk of its adherents 
in the earliest times from persons of the 
lowest and most servile positions — that, in 
Gibbon's well-known words, the new sect 
was " almost entirely composed of the dregs 
of the populace — of peasants and mechanics, 
of boys and women, of beggars and slaves." « 
To say that Christianity began with the 

' Church in Roman Empire, p. 57. 
' Gibbon gives this as ** the charge of malice and 
infidelity," which he proceeds in part to qualify. 


lowest classes, and gradually worked up to 
the higher, is at best a half-truth. It is not 
less true that the gospel often laid hold 
first of persons in better social position, and 
from them worked around and down. Its 
Divine power drew to it men of all classes 
of society from the beginning, and often 
the persons in higher station were the first 
to come, and, through their example, brought 
others. The evidence on this, as on the 
other branches of our subject, has been 
gradually accumulating, and in recent years 
has come to be much better appreciated. 
Still, as respects the ordinary treatment of 
Church History, it may justly be said that 
not a little of it is " a neglected factor." 

In supporting this thesis, which will seem 
to many paradoxical, I do not wish to be 
misunderstood. It is not disputed that in 
the days of the Apostles, and so long as 
Christianity was a proscribed religion, the 




numbers of the wealthy, and learned, and 
powerful, belonging to it were still compara- 
tively few, and that the body of the mem- 
bership of the Church consisted of persons 
of the humbler and middle ranks of society.' 
The wealthy and noble must always be few 
in comparison with others in the Church, for 
this, if for no other reason, that there are 
fewer of them. This is Origen*s reply to 
Celsus as respects the intelligence of the 
Christians, that " among the multitude of 
converts to Christianity, the simple and igno- 
rant necessarily outnumbered the more in- 
telligent, as the former class always does the 
latter^^ 2 Even yet the greater part of our 

' The rude, misspelt scrawls and execrable Latinity 
of many of the Catacomb inscriptions are sufficient 
evidence of this. The contrast has often been drawn 
between the finely executed Pagan epitaphs on one 
side of the Lapidarian Gallery in the Vatican, and 
the hasty, illiterate scribbles of the Catacomb series 
opposite. (Cf. Hasenclever in Jahr. /. Proi. TheoL, 
VIII. pp. 34-5.) • Contra Celsum, i. 27. 


Christian congregations does not consist of 

nobles and millionaires, but of persons 

drawn from the intermediate and humbler 

classes of society — tradespeople, artisans, 

peasants, and the best part of these — and 

still more must this have been the case when 

there was far less of a middle class than there 

is now,' and trade and industry were left 

chiefly in the hands of freedmen, foreigners, 

and slaves. But this inferior social rank of 

the earlier converts to Christianity has been 

greatly exaggerated. The sneer of Celsus^ 

which Origen refutes has been repeated as 

if it were a true description of Christian 

society, instead of a caricature. We shall 

see that if, as Paul says, " not many wise 

after the flesh, not many mighty, not many 

noble" were called,3 there were still, all 

' Some would say no middle class at all, but this 
is an exaggeration. Cf . the sentence from Schultze, 
on p. 112. 

• Contra Celsum, iii. 55. ' i Cor. i. 26- 


things considered, a surprising number from 
these very classes and from the intermediate 
ranks — and as time went on still more — who 
adorned by their faith the doctrine of God 
their Saviour. I am far, indeed, from sug- 
gesting that Christianity derives a lustre 
from the mere social rank of its converts, 
which would not be lent to it by the virtues 
of the humblest.^ The flow of rank and 
wealth into the Church, far from proving 
a source of blessing to it, has proved often 
a cause of backsliding and corruption. But 
it may fairly be contended that just in pro- 
portion to the obstacles which lay in the 
way of persons of rank and wealth becoming 
members of an obscure and uninfluential 
sect, the more signally was the power of the 
gospel magnified in overcoming these ob- 
stacles, and bringing them to the feet of 
the Crucified. Neither must we under-esti- 

' Cf. James ii. 5. 


mate the effects on the progress of Chris- 
tianity of the influence and example of 
persons of this class. That influence was 
great, and in the providential order had 
much to do with the commending of 
the gospel in the circles in which it 

An instructive fore-glimpse of what is 
afterwards to be illustrated in the history 
of the Church is already furnished in the 
personal ministry of its Founder. The 
wealthy and official classes, we know, as a 
body rejected Christ, while "the common 
people heard Him gladly."' The question 
could be asked, " Have any of the rulers 
believed on Him, or of the Pharisees ? " 2 
Yet if we look a little more carefully into 
the list of Christ's personal disciples and 
followers, we shall find, I think, that they 
are drawn neithier from the highest, nor 
» Mark xii. 37. ' John vii. 48. 


preponderatingly from the lowest, ranks of 
society, but from what we should now call 
the middle classes ; while instances are not 
wanting to show the power of the gospel 
on persons of higher social position. Thus, 
among the friends and followers of Jesus 
we have mention made of certain women 
who had been healed by Him and attended 
Him, including — with Mary Magdalene — 
" Joanna, the wife of Chuza, Herod's steward, 
and Susanna, and many others, who minis- 
tered to Him of their substance " ' ; we have 
the family of Bethany — Lazarus and his 
sisters — evidently of good social position ; 
we have, in the band of the Apostles, the pairs 
of brothers, Simon and Andrew, and James 
and John, who, though fishermen, were at 
least in comfortable circumstances — Zebedee 
with his sons owning boats and hired ser- 
vants, and carrying on a fishery business in 

' Luke viii. 2. 


partnership with Simon ' ; we have the pub- 
licans, Matthew and Zacchaeus, the one able 
to make "a great feast in his house" on 
occasion of his Call,^ the other " a chief pub- 
lican," and " rich " 3 ; we have the Roman 
centurion, who had built the Jews a synagogue 
— no mean personage therefore 4 — and Jairus, 
one of the rulers of the synagogue 5 ; we have 
Nicodemus, a ruler of the Jews, and Joseph 
of Arimathea, "a councillor of honourable 
estate," and " rich " ^ ; we have the testimony 
in John, " Nevertheless even of the rulers 
many believed on Him, but because of the 
Pharisees, they did not confess it"7; we 
have such instances as that of the rich young 

' Mark i. 20; Luke v. 10. Nathanael (= Bar- 
tholomew ?) also seems to have been a man of gdod 
social standing (Cf. John i. 45-51). It is noticeable 
that members of this group are found at a distance 
from their homes in Judea, waiting as disciples on 
the Baptist (John i.). ^ Luke v. 29. 

3 Ibid. xix. 2. * Ibid. vii. 5. s Mark v. 22. 

^ Mark xv. 43 ; Matt, xxvii. 57. ^ John xii. 42. 


ruler attracted to Christ,* of the candid scribe 
who was "not far from the kingdom of God," « 
of the other scribe who impulsively offered 
his service.3 All this, when, to use the 
words of the Evangelist, "the Spirit was 
not yet given, because Jesus was not yet 
glorified." 4 

Passing from the Gospels to the Church 
in the Apostolic age, we have a new 
series of examples which look in the same 
direction. The mother church at Jerusalem 
had among its members possessors of lands 
and houses, apparently not a few, who sold 
them and laid the proceeds, in whole or 
part, at the Apostles' feets We have specific 
instances in Barnabas of Cyprus, who, having 
land, sold it ^ ; in Ananias and Sapphira, 
who sold a possession and deceitfully kept 

back part of the price 7 ; in the mother of 

* Luke xviii. i8, 23. ' Mark xii. 34. 

3 Matt. viii. 19. * John vii. 39. s Acts iv. 34, 35. 

* Ibid. iv. 37. 7 Ibid. v. i, 2. 


John Mark, who had a house of her own in 
Jerusalem.' Many of the converts at Pente- 
cost were persons who had come long and 
expensive journeys to the feast ^ ; and at an 
early stage in the history it is testified, "A 
great company of the priests were obedient 
to the faith." 3 The eighth chapter of the 
Acts records how the eunuch of Ethiopia, 
described as "of great authority under 
Candace, queen of the Ethiopians, who 
was over all her treasure," 4 was brought 
to faith by the Evangelist Philip ; the ninth 
chapter relates the conversion of the great 
Apostle of the Gentiles, Saul of Tarsus, 
pupil of Gamaliel, who, as a Roman citizen, 
may be assumed to have been of a good 
family 5 ; the eleventh chapter tells of 
Peter's successful mission to the devout 

centurion Cornelius — a man noted for his 

* Acts xii. 12. » Ibid. ii. 5. 

5 Ibid. vi. 7. * Ibid. viii. 27. 

5 Cf. Ramsay's St. Paul the Travellery ch. ii. 


alms. I A new beginning in the Christian 
propaganda is made in the Gentile Church 
at Antioch, and here, among the prophets 
and teachers who designate Saul and 
Barnabas to their work, we find Manaen, 
the foster-brother of Herod the Tetrarch.* 
Following the Apostle in his missionary 
joumeyings, we have continual examples of 
how the word took root in the hearts of 
persons of the higher ranks, even more 
readily, often, than in the minds of the 
multitude. Thus the visit to Cyprus issued 
in the confusion of Elymas the Magian and 
the conversion of the Proconsul Sergius 
Paulus.3 The first convert in Philippi 
was Lydia, the well-to-do seller of purple.4 
In Thessalonica a great multitude of devout 
Greeks (proselytes) believed, and it is ex- 
pressly recorded, "of the chief women not 

* Acts X. 2, 31. « Ibid. xiii. i. 

3 Ibid. xiii. 12. * Ibid. xvi. 14. 


a few." ' Jason, who received the preachers 
into his house and shielded them from 
violence, was evidently a man of substance.^ 
In the neighbouring city of Beroea it is 
attested that many believed, "also of the 
Greek women of honourable estate, and of 
men, not a few." 3 Athens gave but few 
converts, but one of them was Dionysius 
the Areopagite.4 There is nothing in all 
this of the gospel working its way gradually 
up from below. It goes straight to the 
hearts of these people jof honourable estate 

from the lips of the preacher, or from the 
Scriptures " searched daily." 5 At Corinth, 
besides the tent-makers Aquila and Priscilla, 
who cannot be described as poor, we have 
Crispus, the chief ruler of the synagogue, 
" believing in the Lord with all his house " ^ ; 

' Acts xvii. 4. ' Ibid. xvii. 9. 

3 Ibid. xvii. 12. ^ Ibid. xvii. 34. 

Ibid. xvii. 11. ^ Ibid, xviii. 8. 


at Ephesus we have the conversion of the 
dealers in magic arts and the burning of 
the great pile of their books of sorcery ^ ; at 
the island of Malta we have the cure — ^if not 
the conversion — of the governor Publius 2 with 
many other indications of a similar kind. 
Moving through the history of the " Acts," 
in fact, and gleaning the impressions which ' 
its pictures of the life and work and trials 
of these early Christian brotherhoods make 
upon us, we never feel ourselves in contact 
with Gibbon's " dregs of the populace," but 
are consciously at every point in touch with 
intelligent, well-ordered, and socially reput- 
able communities. These notices in the 
Book of Acts receive confirmation and 
amplification from the Epistles, if there 
also, in many of the Churches, darker 
shades appear. The Church at Corinth, 
to which Paul wrote that "not many 
Acts xix. 19. ' Ibid, xxviii. 8. 


wise, not many mighty, not many noble" 
are called, embraced in its membership, 
besides Crispus, the chief ruler of the 
synagogue, Erastus, the chamberlain of the 
city I ; while the disorders at the Agape and 
many other indications — the taste of the 
Church, for instance, for rhetoric and 
Alexandrian wisdom, its conceit of know- 
ledge, its lawsuits of the brethren one 
with another,* its heresies on the resurrec- 
tion — show that it was not a church 
composed exclusively, or even predominat- 
ingly, of the poorer classes, but a church, 
rather, intellectually disposed, and containing 
in it a good many people of better social 
position. To this church belonged the 
much-praised " household of Stephanas." 3 
I need only allude — for I cannot delay long 
on this part of the subject — to such other 
characters in the Epistles as Philemon of 
* Rom. xvi. 23. » I Cor. vi. 6. ^ jbid. xvi. 15. 


Colosse, the master of Onesimus,' the 
hospitable Onesiphorus,* the well-beloved 
Gaius,3 and the most excellent Theophilus, 
to whom the Evangelist Luke,4 himself a 
physician,S writes his " treatises." The Epistles 
bear witness in our favour in other and less 
direct ways. If the Apostolic Churches had 
slaves in their membership, they had also 

masters, to whom exhortations are ad- 

dressed.^ Specially instructive in this con- 
nection are the passages directed against 
the dangers and abuses of wealth, as, e^,^ 
where Paul exhorts, "Charge them that are 
rich in this present world, that they be not 
high-minded, nor have their hope set on the 
uncertainty of riches, but on God "7; or 
where James cautions against partiality to 
the man with the gold ring and fine 

\ Ep. to Phil. » 2 Tim. i. i6. 

3 3 John. * Luke i. 3 ; Acts i. i. 

s Col. iv. 14. * Eph. vi. 9; Col. iv. i. 

' I Tim. vi. 17 ; of. vers. 9, 10. 


clothing, and denounces the rich men who 
rob the labourers of their hire.^ Finally, 
we have the picture of the Church of 
Laodicea in the Book of Revelation, which 
boasts of being rich and increased with 
goods — if this is to be taken literally — and 
is all the while in dedp spiritual poverty.* 
We have, on the other hand, the testimony 
of the Apostle to " the deep poverty " of the 
Churches of Macedonia — connected, however, 
with a special season of tribulation — but also 
his witness to their abundant liberality in the 
collection made for the poor saints in Judea.3 
I do not think it is an unreasonable con- 
clusion to draw from these data that, while 
there were doubtless poor churches, and 
many poor people in all the^ churches, the 
general membership of the congregations 

' James ii. 2, 5 ; v. 4. The latter passage need not 
refer to Christians. 

• Rev. iii. 17. » 2 Cor. viii. 2 ; ix. 


was, contrary to the usual view, composed 
of fairly well-to-do and intelligent people 
and commonly had among them also 
persons of highly respectable, and some- 
times quite conspicuous positions. I am glad 
in this view to find myself supported by the 
writer already frequently quoted — V. Schultze. 
"It was not the base elements," he says, 
" which came into the Church ; but, on the 
contrary, the better strata of the Roman 
population, the artificers, the shopkeepers, 
and the small landed proprietors, therefore 
preponderatingly the under and middle 
portion of the citizen class, who, in the 
general moral and religious dissolution of 
heathenism, still proved themselves the 
soundest classes in the community."' I 

* Untergang, I. p. 25. I may quote here also Dean 
Meri vale's judgment. ''I have shown in another 
place," he says, " that the gospel was not embraced, 
on its first promulgation in Judea, by the despair of 
thQ most wretched outcasts of humanity, but rather 


propose, in the remainder of the lecture, 
to adduce some of the evidence furnished 
by early ecclesiastical history, which, I 
think, makes clear the justice of this 

Here, again, I can have no hesitation in 
placing in the forefront of my argument the 
comparatively recent and singularly impres- 
sive testimony of the Catacombs. The evi- 
dence which has come to us from this quarter 
is partly elucidatory and corroborative of 
what had formerly been conjectured ; but 

by the hopeful enthusiasm which urges those en- 
jo3ang a portion of the goods of life to improve 
and fortify their position. And so again at Rome 
we have no reason to suppose that Christianity 
was only the refuge of the afflicted and miserable ; 
rather, if we may lay any stress on the monuments 
above referred to, it was first embraced by persons 
in a certain grade of comfort and respectability ; by 
persons approaching to what we should call the 
middle classes in their condition, their education, 
and their moral views." — The Romans under the 
Empire, ch. liv. See further Dean Milman's judg- 
ment cited below, p. 142. 



much of it, also, is entirely new, and to it 
chiefly, perhaps, is due the revived interest 
which of late years has been shown in this 
subject of the social rank of the early 
Christians. The very existence of these 
Catacombs, it may be remarked at the outset, 
taken in connection with the circumstances 
of their origin, is a proof that the Church 
of Rome must from the earliest period have 
had among its members persons of wealth 
and distinction. The oldest of the Catacombs 
go back to the first century — one or two 
perhaps to Apostolic days. In nearly all 
cases they seem to have been begun as 
private burial-places in the gardens or vine- 
yards of persons of the wealthier class,^ while 
the elegance and refinement of their con- 
struction, and the elaboration of their decora- 
tions, point to lavish outlay by their owners.^ 

* Of. Northcote and Brownlow, I. pp. loi, 114 ff. 
' This artistic elegance and finish is characteristic 


In some cases spots of ground were directly 
gifted to the Church for the burial of the 
brethren.^ But it is chiefly in the inscrip- 
tions, enabling us positively to identify par- 
ticular crypts with individuals and families, 
that the interest of this class of discoveries 
culminates. The amount of light thrown in 
this way on the extent to which Christianity 
had penetrated into th$ higher Roman circles 
is really very surprising. I shall notice a 
few of the best known cases, combining with 
the light furnished by the Catacombs such 
knowledge of the facts as comes to us from 
other sources. 

An early case of great interest is that of 
Pomponia Graecina in the reign of Nero. The 
New Testament acquaints us with the fact 

of all the cemeteries which on other grounds are 
shown to go back to first century. Cf. Northcote 
and Brownlow, as above ; Diet, of Christ, Antiq,, I. 

p. 303- 
* Cf. Lanciani, Pagan and Christ, Rome, p. 336. 


that Christianity had early obtained a foot- 
ing in that immense establishment known 
as "Caesar's Household." ^ Prof. Ramsay 
quotes from Mommsen the observation 
that nowhere had Christianity a stronger 
hold than in the household and at the 
court of the Emperors.^ But that, beyond 
this household, Christianity had found its 
way into the highest circles, had long been 

' Phil. iv. 22. See the description of this gigantic 
ostablishment in Lightfoof s PhilippianSf pp. 171 ff ; 
also Friedlander, Sitiengeschichte Roms, I. pp. 71-210. 
Withrow states : l" In remarkable confirmation of 
this fact is the discovery in the recent explorations 
of the ruins of the Imperial Palace [Nero's " Golden 
House"], of several Christian memorials, including 
one of those lamps adorned with evangelical symbols 
so common in the Catacombs " (p. 56). 

■ Church in Roman Empire, p. 57. Harnack sa3rs : 
" We are able to»day, on the basis of fully authenti- 
cated records, to declare, with satisfactory certainty, 
that even in the time of the Aposties the palace of 
the Emperor was one of the chief seats of the 
growing Christian Church in Rome." Art. on "Chris- 
tianity and Christians at the Court of the Emperors," 
in Princeton Review, July, 1878, p. 257. 


surmised from an obscure notice in Tacitus, 
which relates how in A.D. 57, a lady of 
illustrious birth, Pomponia Graecina, wife of 
Aulus Plautius, the conqueror of Britain, was 
accused before the Senate, and was tried and 
acquitted before a domestic tribunal on a 
charge of " foreign superstition " {superstir 
tionts exterfUB\ and how her life was there- 
after spent in deep gloom.^ The peculiarity 
of the diarge in this case led to the con^ 
jecture that the "foreign superstition" in 
question was none other than Christianity. 
So long as it depended solely on this passage, 
the inference was felt to be precarious, and 
we cannot feel surprised that while the 
majority of scholars acquiesced in it, others, 
equally learned, took an opposite view. Now, 
however, the conjecture h£is practically been 
converted into certainty by the discovery by 
De Rossi in the crypt of Lucina — one of 

' Tac. Annmls, xiii. 32. 


the very oldest parts of the Catacombs — of 
several inscriptions unmistakably showing 
a connection of the vault with members 
of the Pomponian gens — one descendant 
bearing this very family name : Pomponius 
Grsecinus.1 It is an ingenious conjecture 
of De Rossi that probably the Lucina who 
gives her name to the crypt is Pomponia 
Grxcina herself. Lucina would then be the 
name assumed by this lady at baptism.' 
This same distinguished person, it should 

' Cf. Northcote and Brownlow for details, I. pp. 
277-281. Lightfoot says, "It is clear therefore that 
this burial-place was constructed by some Christian 
lady of rank, probably before the close of the first 
century, for her fellow-religionists, aod that within 
a generation or two a descendant or near kinsman 
of Pomponia Grscina was buried." {Clemenl, I. p. 
31. Cf. Harnack, Princeton Raiiew, July, 1878, p. 263.) 

" Hasenclever.iii his interesting articles on"Christian 
Proselytes of the Higher Rank in the First Century," 
in fahr. f. Prot. Tktol. viii., seeks to minimise the 
evidence in the above aod later cases, but in view of 
the Catacomb testimony, his arguments need scarcely 
be discussed. Cf. Lightfoot, Clement, pp. 30, 32, &c. 


be said, is connected by many scholars with 
the gospel in another way, though much 
weight, I fear, cannot be allowed to their 
speculations. Specious grounds have been 
alleged for the identification of the Pudens 
and Claudia named in 2 Tim. i v. 21 as 
prominent members of the Church of Rome 
with a Pudens and Claudia repeatedly men- 
tioned in the epigratns of Martial,^ the former 
a Roman centurion of distinction, the latter 
a British princess whom Pudens wedded. 
Last century (1722) there was discovered at 
Chichester an inscription which tells how a 
site was presented by one Pudens to the 
British king, Claudius Cogidubnus — the same 
with whom, as we learn from Tacitus, Aulus 
Plautius had friendly relations in his cam- 
paigns. The presumption is strong that the 

* See the passages quoted in Alford's Excursus on 
Pudens and Claudia, in Proleg. to 2 Timothy, Greek 
Test., iii. p. 104. 


Pudens of Bfartial is an officser who sa:ve9ci 
under Plautius in Britain, and that the princess 
he married was the daughter of this King 
Claudius Cogidubnus. If so, we have a link 
connecting her with Pomponia Graedna, under 
whose protection it may be presumed that 
she joume3red to Rome, and whose connection 
with the family of the Rufi furnishes a reason 
for the assumption by her of the second 
name she bears — Claudia Rufina, The same 
link connects Claudia with Christianity, and 
gives plausibility to the suggestion that 
through Pomponia she may have been intro- 
duced into Christian circles, and with her, 
Pudens.^ The weak point in this train of 
reasoning, otherwise so seductive, is the 
absence of any evidence that Claudia Rufina 
was brought under Christian influences,, for 

' The Pudens and Claudia of Martial were not 
married at the date of the epistle ; neither apparently 
were the pair in the text, since the name of Linus 


the mere occurrence of two names so common 
a& Pudens and Claudia in 2 Tim. iv. 21 does 
not prove it. The identification with the 
members of the Roman Church is favoured, 
however, by writers like Alford, Conybeare 
and Howson, Lewin, and Plumptre ; while 
Lightfoot and others, on chronological and 
moral grounds, decidedly — possibly too de- 
cidedly — ^reject it* The utmost that can be 
said for it at present is that the coincidences 
are unquestionably striking. 

Pomponia Graecina lived on into the reign 
of Domitian, and her influence, as Lightfoot 
su^ests,^ may not have been without its 
share in bringing about the next outstand- 
ing cases of conversion we have to record 

* See in favour of the identification, Alford ut supra 
and against, Lightfoot, Clement, I. pp. 76-79. Light- 
foot gives the references to the others. Farrar, who 
scouts the identification in his St. Paul (ch. 56), uses 
it to garnish his picture in his Darkness and Dawn, 

' Pp. 32-3. 


— those of Flavius Clemens, the consul, and 
Domitilla, his wife — the former the cousin, 
the latter the niece, of the Emperor Domi- 
tian. The basis here again is the statement 
of a heathen writer. Dion Cassius (or his 
epitomiser Xiphilinus) informs us that these 
two persons were accused of " atheism," and 
"going astray after the customs of the Jews" 
{aOBOTtiTog . . . ig TCL lovSaitov Wti s^okIX- 
XovTcc), for which offence Clement was put 
to death, and Domitilla w£is banished to the 
island of Pandatereia in the iEgean.' The 
peculiar wording of the charge long ago 
suggested that, as in the previous case, it 
was really the offence of Christianity for 
which Clement and his wife suffered «; and 

' Dion Cassius, Ixvii. 44 : Suet. Dom. 15. See 
on these passages Lightfoot, Philippians, pp. 21-23 • 
Clement, I. pp. 33-35 : IgnaL I. pp. 12, 13. 

' Thus already Gibbon (ch. xvi.) : " A singular 
association of ideas, which cannot with any propriety 
be applied except to the Christians." Most modern 
scholars agree. 


this conjecture was strengthened by a notice 
in Eusebius, derived from the Roman his- 
torian Bruttius, that Flavia Domitilla, whom 
by a confusion he calls the niece (not the 
wife) of Flavius Clemens, was banished for 
confessing Christ.' It has been reserved for 
Catacomb exploration to clear up the am- 
biguity attaching to this case also, and to 
establish beyond doubt the Christianity of 
the illustrious pair. The cemetery of Domi- 
tilla has been discovered by the labours of 
De Rossi, with inscriptions abundantly at- 
testing her ownership of the ground, and 
its use for Christian burial ^ ; while the 

' Eus. Ecc. Hist, ill. 18 : of. Chronicle under a.d. 
95. On the discrepancies with Dion Cassius, &c., 
see Lightfoot, Phil., pp. 22, 23, and Clement, I. pp. 
44-51 ; and Harnack, Princeton Review, July, 1878, 
pp. 266-69. Harnack favours the theory of two 

» Cf. for details, Northcote and Brownlow, Rom. 
Sott, I. pp. 120-6 : Lanciani, Pag, and Christ, Rome, 
pp. 316, 335-340 ' Lightfoot, Clement, I. pp. 35-37 ; 



furdier discoveiy of an el^jantly-constructed 
crypt of the Flavians shows that, in the 
words of Hamack, "an entire branch of 
the Flavian family embraced the Christian 
faith." I It will not be denied that these 
facts furnish startling illustration of the 
extent to which, by the close of the first 
century, Christianity had pushed its con- 
quests. Next to the Emperor himself, these 
two personages held the highest rank in 
the Empire; Aey stood nearest to the throne; 
their two sons had even been designated by 
Domitian as his heirs to the purple.^ It 

Harnack, Princeton Review, July, 1878, pp. 268-9. 
The cemetery is that of Domitilla, who alone is 
mentioned by Eusebius, but the charge was the 
same against both husband and wife. 

» Harnack (ut supra) says, "What a change I Be- 
tween fifty and sixty years after Christianity reached 
Rome, a daughter of the Emperor (Vespasian) em- 
braces the faith, and thirty years after the fearful 
persecutions of Nero, the presumptive heirs to the 
throne were brought up in a Christian house " 
(p. 269). ' Suet. Dom, 15. 


seemed almost as if, ere the last Apostle had 
quitted the scene of his labours, Christianity 
were about to mount the seat of empire ! 

There is, however, yet another case, belong- 
ing to this period, quite as striking in its 
elements of surprise as that of Clemens and 
Flavia Domitilla. Dion informs us in the 
passage already cited that besides these two, 
" many others " (aXXoi iroXXoO were arraigned 
on the same charges — among them Glabrio 
who had been consul with Trajan, who also 
was condemned, and put to death. The full 
name of this victim of Domitian's persecuting 
zeal was Manius Acilius Glabrio, and his 
family was conspicuous as one of the very 
wealthiest and most illustrious in the State. 
" Towards the end of the republic," says 
Lanciani, "we find them (the Acilii) estab- 
lished on the Pincian Hill, where they had 
built a palace, and laid out gardens which 
extended at least from the convent of the 


Trinita dei Monti to the Villa Borghese. 
The family had grown so rapidly to honour, 
splendour, and wealth, that Pertinax in the 
Senate in which he was elected emperor, 
proclaimed them the noblest race in the 
world."* Doubt was still entertained by 
many, however, whether the terms of the 
passage in Dion necessarily included Chris- 
tianity among the charges on which Glabrio 
was condemned, and Lightfoot, in reviewing 
the evidence, declared that the case seemed 
to him to break down altogether. » It is 
permissible to think that were this eminent 
scholar writing now, his opinion would be 
somewhat modified. For here, again. Cata- 
comb discovery has come to our help. In 
the year 1888, a crypt was laid bare by the 
indefatigable De Rossi, which proved to be 
that of the Acilii Glabriones. A fragment of 

' Pagan and Christ Rome, p. 5. 
» Clement, I. p. 82. 


a marble coffin was found, inscribed with the 
words Acilio Glabrioni Filio^ and additional 
inscriptions have since confirmed the identifi- 
cation.2 As this crypt forms the centre of a 
large group of galleries, its Christian character 
can hardly be doubted. Thus again we see 
Christianity penetrating into one of the 
wealthiest and most renowned families of 
the Flavian age.3 

The individual instances I have cited are 

' Probably son of Manius Acilitts Glabrio, Consul, 
124 A.D. On the Acilian inscriptions, see Frontis- 
piece and Note in Appendix. 

» Lanciani, pp. 4-8. " His end helped, no doubt," 
this writer says, "the propagation of the gospel 
among his relatives and descendants, as well as 
among the servants and freedmen of the house, as 
shown by the noble sarcophagi and the humble 
loculi found in such numbers in the crypt of the 
Catacombs of Priscilla" (p. 7. Cf. Ramsay, Church 
in Roman Empire, pp. 262-3). 

3 A Catacomb inscription furnishes good reason 
also for believing that Bruttius, the historian on 
whom Eusebius depends for his information about 
Domitilla, was, or became, a Christian. Thus Light- 
foot and Lanciani. 


far from exhausting the evidence supplied 
by the Catacombs to the acceptance of the 
gospel by persons of the upper ranks in 
society in the first century, but they may 
sufiice. As an interesting indication from 
the literary side, I may refer to the apocry- 
phal Acts of Paul and Thecla^ which most 
scholars now believe to have at least a basis 
of historical truth. Thecla was the daughter 
of a noble and wealthy family in Iconium, 
and Queen Tryphaena, of Pontus, who is 
shown by recent discovery to be a real 
historical personage, is related to have been 
converted by her. Prof. Ramsay accepts 
these facts as probably historical <; Hamack 
also regards the book as "without doubt 
resting upon historical accounts."^ It there- 
fore adds its grain of testimony to our 
general contention. 

' Church in Raman Empire, p. 414. 
« Princeton Review, July, 1878, p. 261^. 


When we pass to the second century we 

are not so entirely dependent upon Catacomb 

witness as in the first, though here also, as 

we shall see, the Catacombs have important 

aid to oflFer us. The river of Church History 

still flows, indeed, so much underground as 

to be for long periods almost entirely out of 

sight. Yet numerous illustrations are not 

wanting to show us that the gospel was 

drawing its converts on every side from the 

higher as well as the lower orders of society. 

Pliny, it will be remembered, bears emphatic 

testimony to this in Bithynia and Pontus. 

Persons of all ages, of ail ranks, and of both 

sexes, he reports to Trajan, had accepted 

Christianity, and the number was daily in- 

creasing.i The Epistle of Ignatius to the 

Romans, about the same time, presupposes, 

as Dr. Lightfoot points out, that there were 

persons in high quarters in Ronie so in- 

' Ep., 96. 


fluential that the writer fears their inter- 
cession may deprive him of the crown of 
Martyrdom.' Hermas, in his Shepherd — 
that Pilgrim's Progress of the Early Church 
— has numerous references to the wealthy in 
the Church of Rome — possessors of lands 
and houses — whom he rebukes for worldli- 
ness and luxury.^ The wealth of the Church 
is witnessed to us in a more pleasing way 
by its reputation for an abundant liberality. 
Dionysius, the Bishop of Corinth, about 
170 A.D., extols the Church of Rome for 
this grace. "For this," he says, "is your 
practice from the beginning, to do good to 
all the brethren in various ways, and to send 

' Ep, to Ront.f I, 2 ; of. Lightfoot, IgnaL I. p. 356. 
To the same effect Harnack : '' Before what other 
person than the Emperor could this intercession be 
made. . . . We must conclude that there were per- 
sons at that time among the Roman Christians who 
possessed great influence at the Court" (Princeton 
Review, July, 1878, p. 278). 

' Hermas, Sim. i. ; ii. ; viii. 9 ; ix. 20, &c. His 
own mistress was a rich lady. 


contributions to many churches in every 
city, thus refreshing the poverty of those in 
need, and furnishing supplies to the brethren 
in the mines. By these gifts, which ye send 
from the beginning, as Romans, ye maintain 
the ancestral custom of the Romans, which 
your blessed Bishop Soter has not only 
observed, but also increased, providing great 
abundance for distribution to the saints, 
and with blessed words encouraging the 
brethren from abroad, as a loving father 
his children." i When we reflect that the 
bulk of the Roman mob was practically idle 
— clamouring for bread and games, or 
dangling as clients in attendance on the 
rich — and that slaves had little, we see that 
a considerable portion of the membership 
of the Church must have been composed 
of persons in higher social station, or at 
least of the sections which possessed wealth. 
' Euseb., Ecc. Hist, iv. 23. 


Justin's picture of the Christian worship 
bears out this idea. "The wealthy among 
us," he says, "help the needy. . . . They 
who are well to do, and willing, give what 
each thinks fit; and what is collected is 
deposited with the president, who succours 
the orphans and widows, and those who, 
through sickness or any other cause, are in 
want, and those who are in bonds, and the 
strangers sojourning among us." ^ 

This brings us again to the corroborative 
testimony of the Catacombs, and to the 
interesting additional information which they 
supply. I can only draw attention to the 
costly crypts and tombs of the cemetery of 
Praetextatus — a Catacomb of the second 
century — ^which are constructed in the finest 
style of arts in a tomb cased with marble, 

' ist Apol., 67. 

' See the remarkable descriptions of the archi- 
tecture, paintings, and rich tombs in Northcote and 
Brownlow, I. pp. 133-44. 


in one of the chambers of this cemetery, 
lie two bodies, one wrapped in cloth of 
gold, the other in purple, while on a 
grave in the wall is an inscription marking 
the resting-place of " Urania, daughter of 
Herod." ^ It is hardly possible to avoid 
connecting this Urania with the daughter 
of the same name of the famous Herod 
Atticus,2 whose villa and mausoleum are in 
the immediate neighbourhood. If so, the 
identification is one of the most remarkable 
we have yet met with. Herod Atticus is 
known to history as a celebrated rhetorician, 
and the tutor of Marcus Aurelius, but also, 

' Northcote and Brownlow, I. p. 134. 

' " Daughter of Herod Atticus by his second wife, 
VibuUia Alcia'* (Lanciani, Pagan and Christ, Rome, 
p. 9). For a full account of Herod Atticus and his 
extraordinary wealth, see the same work, pp. 287 &. 
Herod's father, through the discovery of a treasure, 
''suddenly became the richest man in Greece, and 
probably in the world." Cf. also Merivale, Romans 
under the Empire, ch. Ixvi. 


through the inheritance of an immense 
treasure, as probably the wealthiest man of 
his time. And here we have apparent evi- 
dence that his daughter had embraced the 
Christian faith. We have, besides, inscrip- 
tions attesting the Christianity of members 
of consular families, and many of equestrian 
rank.' A special interest attaches to the 
discovery by De Rossi, in the cemetery of 
Callistus, of the crypt of Caecilia, the virgin- 
martyr round whom so many legends of 
the Roman Church subsequently gathered. 
Some obscurity rests on the date of Caecilia's 
martyrdom, but it was probably in the reign 
of Marcus Aurelius.* De Rossi's account of 
this lady, which Lightfoot in the main 
accepts, is briefly as follows : That Caecilia 
was a lady of noble birth ; that the land 

' Pagan and Christian Rome, p. lo. 
' See the questions fully discussed in Lightfoot, 
Ignatius, I. pp. 503-4. 


in this place belonged to her gens ; that 
some members of the family were converted 
to Christianity in the second century, so that 
Caecilia was a Christian from her cradle; 
that these Christian Caecilii made over the 
subterranean vaults for the purposes of 
Christian burial, and subsequently were them- 
selves laid here ; and that this was the origin 
of the cemetery of Callistus, or of parts of 
it.^ There is no question in view of the 
inscriptions found that the crypt discovered 
by De Rossi is that in which the body of 
the martyr was originally laid, and from 
which it is related to have been removed 
with honour by Pope Paschal in the ninth 
century .2 The spread of Christianity in the 

' Lightf oot, ut supra, 

^ The body was placed in the Church of St. Caecilia 
in Trastevere. In 1599, in the course of excavations, 
the marble sarcophagus, with the body enclosed, 
clothed in blood-stained robes of golden tissue, was 
brought to light. Cf. Northcote and Brownlow, I. 
pp. 320-1. 


gens is abundantly attested by the numbers 
of epitaphs of these Christian Caecilii and 
other noble families connected with them 
by blood and marriage in adjoining parts 
of the Catacomb, and these not mere de- 
pendents, but, as their titles, ClarissimuSy 
Clarissima^ and the like, show, illustrious 
members of their houses.^ 

All this speaks with great distinctness to 
the highly influential position of the Church 
at Rome, and if we cannot pronounce with 
the same definiteness of other places, it is 
only because, till near the end of the 
century, light almost wholly fails us. When 
we do get a glimpse, as in the beauti- 
ful Epistle of the Churches of Vienne and 
Lyons, giving an account of the martyr- 

' Northcote and Brownlow, I. pp. 278, 327. Twelve 
or thirteen of these epitaphs, all of Caecilii or 
Caeciliani, are found in the crypt of Lucina. There 
seems further to have been some connection between 
this family and that of Pomponia Graecina. 


doms in these places in 177 A.D., the same 
mixture of classes is forced on our atten- 
tion. If we have Blandina, the slave-girl, 
a " noble athlete " in confessing Christ, we 
have also among the sufferers the mistress of 
Blandina; we have one prominent confessor, 
noted as " a man of distinction " {liritrqiioq) ; 
we have a number of Roman citizens ; we 
have heads of households, whose domestics 
are seized to give evidence against them ; we 
have a well-known physician ; and generally 
the martyrs seem to be of the middle or 
better class.' Another remarkable instance — 
again from Rome — is that of the senator 
Apollonius, a man "renowned for learning 
and philosophy," who, on being denounced 
by an informer, made an eloquent defence of 
his religion before the Senate, and was sen- 
tenced to decapitation.2 When, however, we 

' See the Epistle in Eusebius, v. i. 

" Eus. V. 21. The Acts of Apollonius have re- 


approach the close of the century, full light 
returns to us; and we see in the Churches 
of Carthage and Alexandria, and elsewhere, 
how completely Christianity had succeeded 
in penetrating the wealthiest classes in the 
chief centres of population. 

The fatal edict (or, as Neumann will have 
it, rescript^) of Septimus Severus, in 202 A.D., 
which initiated the persecution connected 
with his name, came as a great revealing 
blow to the Churches affected by it It made 
manifest, not only how many of the wealthier 
and dignified classes had, nominally at least, 
embraced Christianity, but also how unfit 
much of their profession was to endure the 
fire of trial. Here it may be noted as 
singular that the brunt of the persecution 

cently been recovered. Of. Conybeare, The Armenian 
Apology and Ads of Apollonius (1896) ; then, after the 
discovery of the Greek Acts, Klette, Der Process und 
Die Acta S, Apollonius (1897). 
' Der Rom, Staat, p. 161. 


was borne, not by the Church of Rome, but 
by the comparatively remote Churches of 
North Africa and Egypt The same thing 
may be observed in other persecutions. 
Why was this? Was it that the Roman 
Christian community was socially obscure 
and insignificant ? Or was it for the opposite 
reason, which Tertullian suggests, that Chris- 
tianity had struck its roots so deeply into the 
State, and had drawn to itself in Rome so 
many illustrious men and women — people in 
the highest positions ' — that even an emperor 
might shrink from the upturning of society 
which a general proscription would involve? 
Septimus Severus himself, as we know, for 
a time looked favourably on Christianity, 
having been healed, it is said, of some dis- 
order by a Christian slave.^ Whatever the 
explanation, the blow did fall pre-eminently, 
not on the capital, but on Carthage and 

' Ad Scafulam, 4. ^ Ibid. 


Alexandria, and its effect in both places was 
to discover at once the hold which the new 
religion had on the people of rank and wealth. 
Tertullian is an unexceptionable witness for 
Carthage. In his address to the proconsul 
Scapula, pleading his cause with that digni- 
tary, he pictures the Christians presenting 
themselves in a body before his tribunal, 
and asks, " What will you make of so many 
thousands, of such a multitude of men and 
women, persons of every sex, and every age, 
and every rank, when they present themselves 
before you? How many fires, how many 
swords will be required? What will be the 
anguish of Carthage itself, which you will have 
to decimate, as each one recognises there his 
relatives and companions, as he sees there, it 
may be, men of your own order, and noble 
ladies, and all the leading persons of the city, 
and either kinsmen or friends of those of your 
own circle? Spare thyself, if not us poor 


Christians ! Spare Carthage, if not thyself ! " ' 
When the storm burst, it was naturally those 
classes which had to make the greatest 
worldly sacrifices which showed the largest 
number of defections. If they did not deny 
Christ, they sought by expedients of bribery 
to secure exemption from trouble. "Whole 
churches," says Tertullian, in this way "im- 
posed tribute en masse on themselves." 2 

Clearest of all among the proofs, how- 
ever, of the extent to which the wealth 
and fashion of these luxurious cities had 
found their way into the Churches, are 
the satirical descriptions and denunciations 
of Tertullian and Clement of Alexandria 
in picturing a state of Christian society 
deeply infected with the vices and follies of 
the age. The rules of living in Clement's 
Padagoguey with their "caustic sketches," 
to use Farrar's words, "of the glutton, and 
* Ad Scap. 5. * De Fuga, 13. 


the dandy, and the painted, perfumed, be- 
wigged, and bejewelled lady of fashion," ^ 
would have no application at all to a Church 
composed wholly or mainly of " dregs of the 
populace " ; and the same may be said of 
Tertullian's denunciation of the luxury and 
extravagance of the women of his time in his 
tract on The Attire of Women? Dean Milman 
takes what seems the only just view of the 
matter. "It appears unquestionable," he 
says, " that the strength of Christianity lay in 
the middle, perhaps the mercantile, classes. 
The last two books of the Pedagogue of 
Clement of Alexandria, the most copious 
authority for Christian manners at that time, 

' The Fathers, I. p. 375. 

' Cf ., e.g,, the picture of extravagance in the close of 
bk. i., and the denunciations of cosmetics, dyeing the 
hair, elaborate hair-attire, splendid and excessive 
dress, in bk. ii., with the concession to those ** whom 
the exigencies of riches, or birth, or past dignities, 
compel to appear in public so gorgeously arrayed as 
not to appear to have attained wisdom " (ii. 9). 


inveigh against the vices of an opulent 
and luxurious community ; splendid dresses, 
jewels, gold and silver vessels, rich banquets, 
gilded litters and chariots, and private baths. 
The ladies kept Indian birds, Median pea- 
cocks, monkeys, and Maltese dogs, instead of 
maintaining widows and orphans, the men 
had multitudes of slaves. The sixth chapter 
of the third book (that the Christian alone 
is rich) would have been unmeaning if 
addressed to a poor community."^ 

But if many were vain and foolish, and 
fell in the stress of the persecutions, there 
were honourable exceptions. The gem of 
the martyrology of this period is the un- 
doubtedly genuine narrative^ of the martyr- 
dom of Perpetua and her companions. 
Perpetua, a young married lady, of noble 
birth, was, with her brother, a catechumen 

* HisL of Christ, ii., ch. ix. (note). 

* Cf. Acts, written partly by Perpetua herself. 


of the Church at Carthage.^ Thrown into 
prison, and tried in the sorest way a woman 
can be, through the entreaties of her aged 
father, and the tenderest appeals to her 
motherhood, she yet, through all, remained 
constant. With her perished four others, 
one of them, Felicitas, a slave. Here, again, 
high-born and humble receive together the 
baptism of blood. In the life of Origen, to 
name other instances, we remember grate- 
fully that " certain lady, of great wealth and 
distinction," in Alexandria, who showed him 
kindness after his father*s martyrdom « ; that 
other wealthy lady Juliana, in whose house 
he was sheltered in CappadociaS; and his 
friend, Ambrose, himself afterwards a martyr 
for Christ, who, out of his abundance, fur- 
nished him with books, scribes, shorthand 

* Or Tuburbium. ' Eus. vi. 2. 

3 Ibid., vi. 17. Paliadius supplements this notice 
on the authority of an entry in a book by Origen 


writers, and every facility for pursuing his 
Biblical studies.^ 

The name of Origen recalls attention to 
another series of facts intimately bearing on 
our present subject. I refer to the relations 
subsisting between Christianity and the 
Imperial Court These, probably, had never 
quite ceased from the days of the Flavians, 
but we find them renewed towards the close 
of the second century, and perpetuating 
themselves during nearly the whole of the 
third. A commencement is made in the 
reign of Commodus, the unworthy son of 
Marcus Aurelius. Marcia, the favourite 
mistress of this emperor, was the foster- 
daughter of a Christian presbyter, and, even 
in her equivocal position, seems to have re- 
tained her interest in Christianity. On one 
occasion we know of, she was instrumental 
in procuring by her intercession the release 

* Eus. vi. 18, 23. 


of certain Christian confessors from the 
Sardinian mines.' There would seem, in 
fact, to have been in this reign a general move- 
ment in the upper classes towards the new 
faith. Eusebius records that " many of those 
highly distinguished in wealth and family, 
with their whole house and kindred, turned 
to their salvation " » ; and Irenaeus speaks 
freely of the faithful in the Imperial palace.3 
Septimus Severus, the next important em- 
peror, was, as we saw, at first not unfavour- 
ably affected to the Christian religion. His 
Syrian wife, Julia Domna, cultured and 
syncretistic in spirit, seems also to have 
been friendly.4 Their son, Caracalla, had a 

' Hippol3rtus, Phil, ix. 12. 

» Eus. V. 21. To this reign belongs the martyr- 
dom of the senator Apollonius referred to above. 

3 Adv. Hcer. iv. 30. 

♦ Cf . Uhlhorn's Conflict of Christy p. 333 (E.T.) ; 
Baur's History of Church, II. p. 207 (E.T.) ; Bigg's 
Christian Platonists of Alexandria, p. 244. See next 


Christian nurse — was fed, as was said, on 
Christian milk.^ Julia's influence may be 
regarded as propagating itself in the reigns 
of the succeeding emperors. We find 
• Hippolytus addressing a treatise to Julia 
Aquila, the second wife of the infamous 
Elagabulus.2 Julia Mammaea, niece of Julia 
Domna, mother of the next emperor, Alex- 
ander Severus, who exercised a large control 
in the government, was deeply interested in 
Christianity, and sent for Origen to Antioch 
to confer with her. 3 Alexander himself 
honoured Christ by placing His statue in his 
private chapel along with those of other 
sages, and had His Golden Rule inscribed on 
the walls of his palace and public monu- 
ments. 4 A succeeding emperor, Philip the 

' Tert., Ad Scap. 4. 

» See Moeller, Church History, I. pp. 191, 20T, 
(E.T.). 3 Eus. vi. 21. 

* Lampridius, Sev, Alex, Cf. in Gieseler (I. p. 192, 
E.T.) and Neander, I. p. 173 (Bohn). 


Arabian, was so favourable to Christianity 
that he was publicly reputed to be a 
Christian.! Origen is related to have had 
correspondence with him and with his wife 
Severa.3 Dionysius of Alexandria could • 
write of the early years of Valerian, even 
after the Church had passed through the 
fiery trial of the Decian persecution, that 
none of the emperors before him had been 
so favourably and kindly disposed to the 
Christians, " not even those who were openly 
said to be Christians" (Philip) ; and that" his 
house was filled with pious persons, and was, 
indeed, a Church (licicXij<rfa) of the Lord."3 
A new spirit, in fact, began to manifest itself 
in this period towards Christianity, in con- 

* Eus. vi. 34. Cf. in Gieseler, I. p. 192. Some 
modern writers, as Aub6, Moeller, favour this view. 
Cf. MoeUer, Church History, I. p. 192 (E.T.). 

' Eus. vi. 36. 

s Ibid. vii. 10. Valerian subsequently became a 


trast with the spirit of contempt which had 
formerly prevailed, the spirit of eclecticism 
and toleration — the intellectual counterpart 
of which is seen in Neo-Platonism.^ Under 
these circumstances, Origen could boast that 
some addition was made to the numbers of 
the Christians every day, and that in the 
multitude of believers were numbered "not 
only rich men, but persons of rank and delicate 
and high-born ladies."* Eusebius also speaks 

! of " the wealthy and opulent " in the Church 


of Rome at the time of the Novatian schisms 
in the middle of the century. 

It was clearly enough perceived, however, 
by thoughtful men like Origen, that the final 
victory would not come without a terrible 
closing struggle. This time of testing soon 
arrived. The Decian persecution broke over 
the Church, discovering, as before, the 

' See Lect. III. ' Contra Celsum, vii. 26; III. 9. 

3 Eus. vi. 43. 


numbers of persons of wealth and rank 
within Its pale, but proving also the frailty 
of their profession. The well-known passage . 
of Dionysius of Alexandria gives us a vivid 
picture of the behaviour of these apostates. 
When brought to the altar, after the edict 
had actually been promulgated, "all were 
greatly alarmed," he says, " and many of the 
more eminent came immediately forward in 
their fear ; others, holding public offices, were 
drawn on by their duties ; others were urged 
on by those about them. When called by 
name, they approached the impure and 
unholy sacrifices, some pale and trembling, 
not as sacrificers, but as if they were them- 
selves to be sacrifices and victims to the 
idols, so that they were jeered at by the 
large multitude that stood around, as it was 
plain to all that they were afraid either to 
die or to sacrifice ; but some advanced more 
readily to the altars, stoutly asserting that 


they had never before been Christians." ' 
More eloquent than any statement of Church 
historians, however, is the language of the 
persecuting edicts themselves. That of 
Valerian, after he had assumed the rdU of 
a persecutor (A.D. 258), is specially directed 
against office-bearers and persons of high 
rank in the Church. « It ordains " that 
bishops, presbyters, and deacons be imme- 
diately put to death ; that senators and men 
of rank and knights be first of all deprived 
of their rank and property, and then, their 
means being taken away, if they still continue 
to be Christians, be also punished with death ; 
that matrons, after forfeiting their property 
be banished ; that those in Caesar's house- 
hold who have formerly made profession of 
Christianity, or now profess it, be treated as 
Caesar's property, and, being put in chains, 

' Eus. vi. 41. 

^ It is given in Cyprian's Epistle to Successus 
(Ep. 80). 


be distributed among the Imperial estates." 
We are, accordingly, not surprised to learn 
from Dionysius that amongst the victims 
of this persecution were " men and women, 
young and old, young virgins and aged 
matrons, soldiers and private persons of 
every kind and every age." He himself was 
an example of one who had repeatedly had 
experience of "confiscations, proscriptions, 
plunderings of goods, loss of dignities." ^ 

The forty years* peace which elapsed 
between this persecution and the last decisive 
struggle in the Diocletian persecution fur- 
nishes us with few details, yet with sugges- 
tive general notices of the continued growth 
of the churches in numbers, splendour, and 
influence, one marked outward token of this 
prosperity being the number of splendid 

' Eus. vii. II. As respects the order of knights, 
Lanciani mentions that hundreds of inscriptions of 
persons of equestrian rank are found in the Cata- 
combs (p. lo). 


ecclesiasticaf edifices which now began to 
be erected. We read of Christian governors 
of provinces, and of the freedom to profess 
Christianity granted to the members of the 
Imperial household — "wives, and children, 
and servants."^ Mention is made of the 
multitudes crowding in every city to the 
houses of worship — "on whose account," 
says the historian, "not being content with 
the ancient buildings, they erected spacious 
churches from the foundation in all the 
cities." 2 That this is not an exaggeration 
is shown by the great church in Nicomedia, 
which appears to have been one of the 
architectural ornaments of this city — the seat 
of the Court at the time — and by the later 
edicts for the demolition of the churches 
generally .3 Yet instances exist to show that 
Christians were not entirely safe even during 

^ Eus. viii. I. Instances are given. ^ Ibid. 

Lactantius, De Morte Per, 12 ; Eus., Ecc, HisL viii. 2. 


this interval of peace. We know at least of 
one illustrious Roman officer at Caesarea who 
suffered death for his faith ; and we read 
also of how one Astyrius, a Roman of 
senatorial rank, in high favour with the 
Emperor, and well known to all for his 
noble birth and wealth, took the body of 
the martyred man, and, covering it with a 
splendid and costly dress, gave it becoming 

The great accession of members and out- 
wardly prosperous condition of the Church 
at this time is beyond dispute, and the 
incidents of the last and most dreadful of 
the persecutions only furnish new corrobora- 
tions of it. During the first nineteen years 
of his reign, Diocletian had Christians every- 
where about his person. Some of the officers 
of highest rank in his palace were Christians^; 

' Eus. vii. 15, 16. 

^ Ibid. viii. 6. Such was Lucianus, the chief cham- 


his own wife and daughter, Prisca and 
Valeria, were believed to be Christians.^ 
The first persecuting edict was directed 
against the church buildings and the Scrip- 
tures rather than against persons; but it 
ordains also that those holding honourable 
positions were to be degraded, and servants 
in the household, if they persisted in their 
Christianity, were to be made slaves.^ What 
one notices with satisfaction in this per- 
secution is the superior steadfastness of 
believers in the higher orders, as con- 
trasted with the frailty of this class on 
previous occasions. Many of the most 
illustrious martyrs of Diocletian's reign are 
persons of exalted rank. Such were some 
of the great officers of the palace, of whose 
sufferings and constancy a special account 

berlain, to whom Theonas, Bishop of Alexandria, 
wrote a letter of advice. See the account in 
Neander, I. pp. 197-9 (Bohn). 
' Lact. 15. =* Eus. viii. 2. 


is given.* Such were the martyrs of the 
Thebais, many of them, as Eusebius tells, 
"distinguished for wealth, and noble birth 
and honour, and excelling in philosophy and 
learning " « ; such was Adanetus, of Phrygia, 
a man of noble Italian family, "who had 
been advanced through every honour by the 
emperors," and had reputably filled the 
highest offices 3 ; such were certain ladies 
of Antioch, "illustrious above all for wealth, 
for family, for reputation "4 — and many more 
of whom these are but examples. A striking 
instance, referred to in the previous lecture, 
is that of a town in Phrygia which was 
burned with all its inhabitants because its 
whole population, including the governors 
and magistrates, with all the men of rank, 
had confessed themselves Christians, and 
refused to sacrifice.s It may be remem- 

' Eus. viii. 6. ' Ibid. viii. 9. 3 Ibid. viii. 11. 
♦ Ibid. viii. 12. s jbid. viii. 11. 


bered also how the Council of Elvira, in 
Spain, in 306, shows us great landowners 
and persons in the highest magistracies in 
the membership of the Church. 

There is only one other line of evidence to 
which, in closing, I would advert for a moment, 
as bearing on this question of the penetration 
by the gospel of the higher ranks of society. 
It is that furnished by the social station of 
the great teachers of the Church. That 
these, like the earlier Apologists, were men 
of education and refinement is a fact which 
of itself implies a standing sufficiently high 
to secure for them the advantages of a liberal 
training. But we have only to recall the facts 
of their lives to be reminded that many of 
them in reality sprang from families of wealth 
and distinction. Tertullian was the son of 
a proconsular centurion — no very high rank 
perhaps — but enough to obtain for him the 
benefits of a legal and rhetorical education. 


We are probably right in saying that Cle- 
ment of Alexandria was the son of wealthy 
parents. His culture and extensive travels 
would seem to imply as much. Cyprian, we 
know, was of patrician descent, and inherited 
large possessions. Two other distinguished 
teachers of the third century — both pupils 
of Origen — Dionysius of Alexandria and 
Gregory Thaumaturgus, were of wealthy 
and honourable families. So was Pam- 
philus of Caesarea, the friend of Eusebius, 
and founder of the famous library in that 
city. It is going beyond our present limits 
to extend our view to the fourth century, but 
if we do so we have such conspicuous in- 
stances as Basil the Great of Caesarea and 
his brother Gregory of Nyssa, as Ambrose 
of Milan, as Chrysostom of Antioch, and 
many others that might be named. I trust, 
however, I have already said enough to show 
the baselessness of the theory that the bulk 


of the adherents of early Christianity were 
drawn from "the dregs of the populace," 
and to demonstrate that the gospel from 
its earliest beginnings in no slight degree 
affected the higher as well as the humbler 
classes of society. 



The instreaming of Pagan influences on Christianity 
has for its counterpart the outstreaming of 
Christian influences on Pagan society — ^These 
also ordinarily under-estimated — Silence of 
Pagan writers : what it means — Christianity and 
culture in the First Century — New Testament 
Epistles — Seneca and the Gospel — Rise and 
character of Apology in the Second Century — 
The literary attack on Christianity : Celsus — 
Significance and spread of Gnosticism — ^The 
Pagan ethical revival in Second Century — 
Pagan preaching — Influence of Christianity on 
these — ^The Mysteries — The old Catholic Fathers 
— Rise of Neo-Platonism — Effects of Christianity 
on morals and legislation — Conclusion. 




" The Catholic Church is that form 
of Christianity in which every element of 
the ancient world has been successively 
assimilated which Christianity could in any 
way take up into itself without utterly losing 
itself in the world. . . . Christianity has 
throughout sucked the marrow of the 
ancient world, and assimilated it.** ' If 

* Art. on " Research in Early Church History " 
in Cont, Rev,, Aug. 1886, p. 234. 




this dictum of Harnack's is correct, the 
counter thesis must hold good, that Chris- 
tianity must have penetrated deeply into 
the thought and life of the ancient world 
before such assimilation was possible. Be- 
fore, for instance, Christianity could suck 
the marrow out of Greek philosophy, as 
Harnack supposes it did, it must have 
penetrated into minds possessed with the 
spirit and ideas of that philosophy — must 

have entered deeply into the circles and 
schools of culture. I am to ask in the 
present lecture how far this penetrative 
process went, and what traces it has left 
of itself in history. 

Our previous inquiries have an important 
bearing on the subject now to be investi- 
gated. If it were the case that Christianity 
had only an insignificant fraction of the 
population in its following, — if its adherents 
were collected chiefly from the base and 


servile classes, — if it was practically unheeded 
and well-nigh totally despised by persons of 
higher station and better culture for at least 
the first two centuries, it would be natural 
to conclude that traces of its influence on 
society would be scarcely perceptible, and 
that what look like such traces must be ex- 
plained in some other way. We must hold 
with Friedlander that "it is scarcely think- 
able that in the heathen world before the 
time of Severus, the world-historical impor- 
tance of the new religion, so little regarded 
and so contemptuously judged of, was even 
so much as suspected." ' But if, as I have 
endeavoured to show, the case was far dif- 
ferent, — if Christianity had both a larger 
following, and was drawing its adherents 
from the higher and educated classes to a 
much greater extent than is commonly as- 
sumed, — then we are prepared to entertain the 
' Sitiengeschichie Roths, III. p. 536. 


expectation that the traces of its action on 
the Pagan world will be neither few nor 

There is a point of considerable moment 
in this connection to which it is desirable that 
attention should be directed at the outset. 
Much stress is often laid {e,g, by Friedlander^), 
in disproof of any considerable influence of 
Christianity on the thought and life of the 
time, on the silence of Pagan writers respect- 
ing the new religion. How, it is asked, if 
Christianity was so powerful a factor as we 
hold it to have been in the second century, 
should a philosophic writer like Marcus 
Aurelius, for example, pass it by with only 
one contemptuous reference? This silence 

' See his argument, Ibid., p. 533. " Christians and 
Christianity," he says, " till near the end of the second 
century, are, in the classic literature, only very seldom 
and incidentally, indifferently and contemptuously 
mentioned." Similarly Addis, in Christianity and the 
Roman Empire, p. 51, "Epictetus and M. Aurelius 
dismiss it with a scornful phrase," &c. 


of heathen writers is not quite so great as is 
assumed — Pliny was not silent, nor Fronto, 
nor Celsus — but even if the fact were as 
stated, there is one important consideration 
which greatly takes away the point from the 
argument. Nothing is better ascertained 
than that it was the fashion of heathen 
writers, even of those who were best ac- 
quainted with Christianity, to show their 
contempt for it, by deliberately dissembling 
their knowledge of it, and refraining from any 
mention of it in their works. Prof Ramsay 
has noticed this in regard to Dion Cassius, 
who wrote in the third century, when it will 
not be denied that Christianity was a grow- 
ing and formidable force, but who seems 
studiously to have refrained from referring to 
the Christians in his history ; and to iElius 
Aristides, the famous rhetorician, a contem- 
porary of Polycarp under the Antonines, 
who likewise makes a point of not mention- 


ing the Christians (testified to be so numerous 
and influential in, Asia Minor by Pliny), but 
speaks of them generally as " those in Pales- 
tine." " It was apparently a fashion and an 
affectation," Prof. Ramsay says, "among a 
certain class of Greek men of letters about 
160-240 to ignore the existence of the Chris- 
tians, and to pretend to confuse them with 
the Jews." ^ It was not, however, . I would 
observe, a fashion confined to this period, and 
to Greek writers ; and did not apply only to 
the Christians, though in their case it was 
specially noticeable. Boissier warns us against 
being deceived by the grand airs of disdain 
and ignorance which the Romans affected for 
everything which was removed from their 
habits and traditions.^ " The conspiracy of 

* The Church in the Roman Empire, p. 264. Fried- 
l£lnder also has no doubt that the passage in Aristides 
refers to the Christians (III. p. 533). 

' La Religion Romaine, II. p. 59, 4th edit. (bk. ii. 
c h. 5). 


silence," as this writer names it, was main- 
tained, astonishing to say, quite as effectively 
in the fourth and fifth centuries — long after 
Christianity had decisively triumphed in the 
State — as in the second. " Paganism," says 
Dean Merivale, "abstained studiously from 
any allusion to the place which Christianity 
now actually held in public life. It made an 
effort, a laborious effort, to pass over the 
phenomenon in complete silence. Through- 
out the few remains of popular literature of 
the age of Constantine we can trace, it seems, 
no single reference to the existence of the 
Christian Church or Creed. Even at the end 
of the century, the poet Claudian, in versi- 
fying, as is his wont, all the chief events of 
contemporary history, has not one word to 
say of the new religion, which in his day had 
effected a complete revolution both in Church 
and State." ^ And Claudian here was no 
* Epochs of Early Church Hist, p. 6. 


exception. Speaking of Macrobius, by whom 
" the name of Christianity is not even once 
pronounced," Boissier remarks, " Our surprise 
is redoubled when we find the same silence 
preserved by nearly all the Pagan writers of 
this time (fourth and fifth centuries), by 
grammarians, orators, poets, and even his- 
torians, though it appears singular that they 
should omit, in a narrative of the past, such 
an event as the triumph of the Church. 
Neither Aurelius Victor nor Eutropius men- 
tions the conversion of Constantine, and it 
would seem, to read them, that all the princes 
of the fourth century persevered in the prac- 
tice of the ancient worship. It is certainly 
not chance which leads them all to avoid 
mentioning the name of a religion which they 
hate ; it is a plot, a party move, the meaning 
of which can deceive nobody." ' These un- 

* La Fin du Paganisme, II. p. 243. Cf. also Light- 
foot, Philii>pians, pp. 28-29. 


questionable facts do away, I think, in great 
part with the relevancy of any argument 
derived from the mere silence or contempt of 
Pagan writers. If M. Aurelius did not men- 
tion the Christians, it is not, as we shall 
immediately see, because he did not know 
enough about them, but because he did not 
desire to mention them, or willed to ignore 

That even in the Apostolic Age 
Christianity had entered as a ferment 
into minds possessed of some degree of 
literary and philosophical culture is evi- 
dent from the phenomena met with in 
several of the Apostolic Churches, as well 
as from the cast and character of the 
New Testament writings themselves. In 
Corinth, and Ephesus, and Colosse, e,g'., the 
earlier danger to which the Churches had 
been exposed, and to which the Churches 
in Galatia succumbed, of bejng dr^vsrn 


back into the web of legal bondage, 
had evidently given place to a new and 
subtler peril-^that of the gospel being 
brought into dependence on a philosophy 
foreign to its nature, and spoiled by being 
mixed up with human speculations, and 
set forth in the trickery of an artificial 
rhetoric. It is easy, if we recall the scenes 
of agitation and disputation amidst which 
the gospel was introduced into some of 
these Churches — the conflicts, for instance,' 
around the judgment seat of Gallio,' or 
the two years daily disputation in the 
school of one Tyrannus at Ephesus,^ with 
its sequel in the burning of the magic books 
— to realise how this should be so. In 


Corinth it was the alliance with Greek 
wisdom and heathen rhetoric that was 
sought ; in Colosse it was amalgamation 
with Essenian and incipient Gnostic ele- 
*l ^cts xviii,i>-i7. 'Ibid. xix. 9, io, 19. 


ments that was attempted ' ; but in either 
case the result was the same — a departure 
from the purity and simplicity of the gospel 
— ^an exaltation of knowledge over piety — 
and a straying into various paths of intel- 
lectual heresy. It is interesting to observe 
also how the Apostle, deals with these 
aberrations — not by denying the value of 
knowledge, or the legitimacy of the claim 
of the mind for satisfaction in the sphere 
of intelligence, but by affirming the power 
of Christianity to develop a ao^fa of its 
own, and by setting in their right rela- 
tions knowledge and love. 2 It is not 
wisdom as such, but " the wisdom of the 
world" against which the Apostle's polemic 
is directed. But the New Testament 
writings themselves, in their very form and 
structure, in many instances bear witness to 

* Cf . Lightfoot's Colossians, 
^ I Cor. ii. 6, 7 ; viii. i. 


the intellectual atmosphere in which they 
were produced. The Pauline Epistles, with 
their deep thought, their closely-knit reason- 
ing, and their views of truth reaching out 
into the eternities before and after, were, on 
the face of them, not intended for illiterates 
or weaklings ; the Epistles to the Ephesians 
and Colossians, with their developments of 
the cosmological aspects of redemption and 
their implied references to Gnostic specula- 
tions, discover that they are written in view 
of active heretical tendencies ; an unmistak- 
able Alexandrian stamp rests on the Epistle 
to the Hebrews ; and the Fourth Gospel, 
however profoundly separated in substance 
from Philonism, yet shows, I cannot but 
think, in the shape in which its prologue 
is cast, a desire to create a bridge between 
the current Logos speculations and the truth 
as it is in Jesus. 

Accepting these facts as indications of 



the subtle yet energetic manner in which 
Christianity was engaging the interest, and 
penetrating the thought, of intelligent circles 
in the greater heathen communities, I go on 
to inquire whether evidences of this can be 
discovered outside the New Testament in 
the general Pagan world of the first century. 
There is no a priori reason why they should 
not be, for, as Boissier remarks, "If De 
Rossi is right, it is necessary to assume 
that Christianity was less unknown to the 
rich and lettered in the first century than 
is supposed."' This inquiry has commonly 
been associated with the name of Seneca, 
in the reign of Nero — and not unnaturally, 
for in Seneca's writings we have at once the 
best specimens of the ethical thought of that 
time, and the niost singular approximations 
in sentiment and expression to the new 
ideas introduced by Christianity. Whether, 
' La Rel. Romaine, II. p. 62 (4th edit.). 


or how far, these resemblances are due to 
any measure of acquaintance with the new 
religion — to any direct or indirect influence 
of the gospel spirit — or, again, are an 
independent development from Stoicism, is 
a question on which opinions are, and pro- , 
bably will always be, widely divided, and 
which will tend to be determined according 
to the presuppositions with which the in- 
quirer sets out^ We would not depreciate 
the splendid services which Stoicism, with 
its stern and elevated, yet haughty and im- 
passive, doctrine of virtue, its notion of a 
unity of mankind based on reason, and its 
cosmopolitan ideals, rendered as a prepara- 
tion for Christianity ; and we must not 
overlook the fact that, notwithstanding 

' Among others the question is discussed by 
Fleury, Troplong, Aubertin, Lightfoot, Hasenclever, 
Schmidt, Friedlander, Boissier, and Farrar. The 
fullest discussion in recent writers is by Boissier 
and Lightfoot. 



apparent coincidences with Christian ideas 
and phrases, Seneca's thinking is still, at 
bottom, unchangeably and even crudely 
Stoical.^ At first, too, it must be granted, 
the presumption is strongly against any 
contact of Seneca with Christianity. The 
fictitious correspondence of the philoso- 
pher with St. Paul is long since given 
up ; there is no evidence that Seneca ever 
saw or heard of the Apostle, though the 
possibility of such knowledge cannot be 
denied 2; the fact that it was his brother 
Gallio before whom Paul appeared in 
Corinth affords but a slender ground for 
supposing that the details of this incident 
may have reached Seneca; while the cir- 
cumstances that Seneca, when Paul reached 

* See the convincing evidence of this, e.g,, in 
Lightfoof s dissertation on St Paul and Seneca in 
his Philippians. 

» The possibility is allowed by Friedlander, &c, 

3 Urged by Hasenclever. 



Rome, was already a man of sixty years of 
age, whose philosophical " Weltanschauung " 
may be presumed to have been completed, 
and whose death fell some four years later 
(A.D. 65), is certainly of considerable weight. 
It is not contended, however, except by a 
few, that Seneca's philosophical view ever 
was fundamentally changed. But against 
these negative considerations there are 
others of a more positive character which 
may fairly be placed. Paul was not the 
only channel through which Seneca may 
have derived some knowledge of the 
ethics of the gospel. The Christians, as 
we saw in the first lecture, were by no 
means in his day an obscure party in 
Rome ^ ; numbers of them were found in 
the palace, and among the domestics of 
the great households, including probably 
Seneca's own ; the sj^e was in the habit 
' Cf. Lightfoot, pp. 25, 33. 


of familiar converse with his slaves ^ ; the 
recent case of Pomponia Grsecina must 
have been the subject of much conversa- 
tion in the highest circles 2 ; the bonds, 
and no doubt the preaching of the 
Apostle were bruited throughout the 
Praetorium and widely elsewhere, and, in 
Lightfoot's words, " a marvellous activity " 
was awakened "among the disciples of 
the new faith." 3 The Apostle Paul under- 
went a public trial, at which Seneca may 
have been present 4 ; it is not impossible 
that even the incident of Gallio may have 
come to the philosopher's ears, if not 
otherwise, yet through the mention of it 
in the tales told of this remarkable 

' Ep. 47, Cf. Lightfoot, p. 300. 
' See last lecture. 

3 Lightfoot, p. 32. Not only " throughout the 
Praetorium," but "to all the rest." "In every way 
Christ is preached " (Phil. i. 13, 18). 

4 Thus De Rossi. Friedlander questions his 
argument, III. p. 535. 


prisoner. Dr. Lightfoot also mentions,' 
what his quotations bear out, that it is 
in the. later writings of Seneca that these 
approximations to Christian ideas are 
most apparent^ 

All this, however, does not amount to 
positive proof, and it is on the internal evi- 
dence of Seneca's writings that the deter- 
mination of the probabilities of this question 
must mainly rest. And here, though on 

' Pp. 291, 298. 

' There is a passage in Seneca's Episties in which 
he describes some striking influence which had 
produced a marked change in him. ''I perceive, 
Lucilius/' he says, ''that I am not only amended, 
but transformed. ... I would desire to share with 
you my change so suddenly experienced." ("In- . 
telligo, Lucili, non amendari me tantum, sed trans- 
flgurari . . . cuperem tecum communicare tam 
subitam mei mutationem," i. 6.) He sends his friend 
the books which had wrought this change in him, 
with the passages marked. There is nothing, cer- 
tainly, to connect these books with Christian writings, 
but the words are remarkable. The Epistles to 
Lucilius belong to the last years of his life. 


academic grounds it will always be possible 
to say about as much against as for any 
Christian influence on Seneca, I think the 
reasons for presuming some degree of such 
influence are exceedingly strong. It remains 
the fact, account for it as we may, that about 
the middle of this century a warmer and 
more tender breath begins to enter into 
Stoicism, which, thereafter, continuously 
animates it ; a purer conception of God's 
Fatherly goodness and beneficent Provi- 
dence ; a kindlier and gentler tone towards 
slaves and dependents ; something like a 
religious trust and resignation ; a more 
merciful and gracious spirit generally. This 
is first perceptible, as far as I know, in the 
writings of Seneca, and it is specially per- 
ceptible in his later years. ^ We know of one 

' Troplong remarks, after De Maistre, that Seneca 
has written a fine book on Providence, for which 
there was not even a name at Rome in the time of 
Cicero, and he speaks of the new Stoicism as 


cause which would produce this change, 
while it does not seem to follow naturally 
from the Stoicism of the remaining parts of 
Seneca's system, with which it stands rather 
in striking inconsistency. We are driven 
back, therefore, on an analysis of the sup- 
posed resemblances, and here, after making 
every reasonable deduction, it is difficult not 
to agree with Dr. Lightfoot, as the result of 
his singularly impartial survey, that " a class 
of coincidences still remains . . . which can 
hardly be considered accidental,"' and of 
which some measure of acquaintance with 
Christianity — at least contact with its spirit 
and teaching in some oral form — affords the 

'' enveloped, as it were, in the atmosphere of Chris- 
tianity." — Llnfluence du Christy I. eh, 4. Cf. Meri- 
vale*s Romans under the Empire, ch. liv. 

' P. 298. Prof. Ramsay, in his Church in the Roman 
Empire, said, '' that Seneca had some slight acquaint- 
ance with Christian teaching appears to be plain 
from his writings" (p. 273). His statement in St, 
Paul the Traveller is less positive (ch. xv.). 


most natural explanation. I do not attach 
much importance to the fact, but it is worth 
mentioning, that a tomb was discovered at 
Ostia bearing the inscription, " Annaeus 
Paulus Petrus," showing that at a later period 
(third century) persons belonging to the family 
of Seneca, possibly descendants of freedmen, 
were Christians.^ At a later period, the 
evidence of the influence of Christianity on 
this transformed Stoicism is clearer. Epictetus, 
the lame slave, and noblest representative 
of second-century Stoicism, refers, indeed, 
but once to the Christians under the con- 
temptuous name of " Galileans," yet his 
discourses breathe a remarkable spirit of 
elevated piety, and Dr. Lightfoot finds in 
them parallels with the Gospels and writings 

' The exact words are, " ANNiEO. Paulo. Petro. 
ANNi£US. Paulus." — De 1 Rossi in Bull, di ArcheoL 
crist, 1867. Cf. Harnack in Princeton Review, July, 
1878, p. 261 ; Friedlander, III. p. 535. Boissier, 
Lightfoot, Renan, &c., refer to the inscription. 


of Paul, which he can hardly believe to be 
accidental. On one such coincidence he 
remarks that, " combined with the numerous 
parallels in Seneca's writings collected above, 
it favours the supposition that our Lord's 
discourses, in some form or another, were 
early known to heathen writers."^ 

In this second century, to which we now 
come, we reach a period in which the in- 
fluence of Christianity on general contem- 
porary thought is no longer a matter of 
precarious inference, but is attested by a 
wide range of interesting facts. The prin- 
cipal of these within the Church are the rise 
of a vigorous and learned Christian Apology, 
and the development in every form and 
variety of the heterogeneous systems which 
we group under the name of Gnosticism ; 
while, in the Empire itself, phenomena pre- 
sent themselves, which, as I believe, are inex- 

» P. 316. 


plicable save through the powerful and still 
under-estimated penetration by Christianity 
of the Pagan world of religion and culture. 
The second century is peculiarly the age 
of the Christian Apology. It was an age 
intensely literary, and, as we shall see, was 
marked by a powerful religious and ethical 
revival. The rhetorician, the philosopher, the 
preacher, the teacher, the declaimer, were 
everywhere. Under the arrangements in- 
stituted by Vespasian for the support of 
lecturers throughout the provinces and cities, 
literature took on a new refinement, schools 
and universities flourished, and thought and 
speech ran naturally into the forms of 
rhetorical and philosophical discourse and 
argument' In harmony with this spirit, 

' See the sketches of this age in Merivale's Romans 
under the Empire, chs. lx.,lxvi.; and in Renan's Marc- 
Aurele, ch. iii., "The Reign of the Philosophers." 
Cf. also Hatch's Hibbert Lectures, Lects. II., IV., 
" Greek Education," " Greek and Christian Rhetoric." 


fostered by the patronage of Hadrian and 
the Antonines, there now began what may 
be succinctly described as the set literary 
defence of Christianity. I do not concern 
myself here with the theology of the Apo- 
logists, which, in my view, has had scant 
enough justice done to it by Engelhardt, 
Hamack, and their followers,^ but confine 
myself to what is implied in the very exis- 
tence of such an Apology. It needs no 
elaborate proof to show that the character 
of the age, as I have just described it, power- 
fully affected the form of the Apology. 
It is conceded that Justin and the rest 
who represent this phase of Christian litera- 
ture treat Christianity predominatingly as 
a "new philosophy "^ — a fact which goes 

' To this school the Apologists have lost the real 
meaning of Christianity, and reduced it to a Moral- 
ismus, or rational natural theology — a very unfair 

' Cf. e.g.y Justin, Dialogue, 8. 


with the other, that most of these writers 
were philosophers or rhetoricians by train- 
ing and profession. The literary and rhe- 
torical stamp is, therefore, on all they 
write; the learning, the arts, the dialectic 
of the schools, the skill of the forensic 
pleader, are brought into play by them 
without stint or disguise. This is the side 
of the Apology commonly dwelt on, but 
there is another. The very appearance of 
such an Apology marks a great step in 
advance. It shows not only that the spirit 
of the age had affected Christianity, but also 
that Christianity had pushed its way into 
literary circles, and was attracting their 
attention. It makes clear that the Christians 
were beginning to have confidence in them- 
selves, felt their growing power, were no 
longer content to be " a dumb folk, mutter- 
ing in corners,"* as their enemies scornfully 

' Min. Felix, 8, 31. 


described them, but were emboldened to 
present their case in the open court of public 
opinion, and to challenge a verdict in their 
favour on the ground of its inherent reason- 
ableness. There is a high tone in the writers 
of these Apologies which the reader cannot 
mistake. " They are always more or less 
conscious," as Baur says, " that they are the 
soul of the world, the substantial centre hold- 
ing everything together, the pivot on which 
the world's history revolved, and those who 
alone have a future to look to. . . . When 
there are men," he adds, "who feel them- 
selves in this way to be the soul of the world, 
the time is indisputably approaching when 
the reins of the government of the world will 
fall unasked into their hands." » The point 
of special interest to us in this connection is 
that, as I have already said, these writers — 
one and all — were men of liberal culture, 
History of Church, II. pp. 129, 131 (E.T.). 


of wide and varied learning, several of 
them philosophers by profession ; and they 
appear at a great variety of points scattered 
over the surface of the Church. Aristides, 
the author of the earliest complete Apology 
we possess — only the other year recovered — 
was a philosopher of Athens ; Athenagoras, 
in thp reign of M. Aufelius, was also a 
philosopher of Athens ; from Athens, too, 
is said (though this is doubtful) to have 
come the oldest of all the Apologies — 
that of Quadratus. Justin Martyr passed 
through the Platonic and other schools of 
philosophy in .his search for the truth, 
and after his conversion, continued to wear 
his philosopher's mantle, and to dispute 
in public places in Ephesus and Rome 
with any who would hear him. A man of 
learning like himself, though of widely diffe- 
rent spirit, was his disciple Tatian — the 
author of the recently discovered Diatessaron. 


Theophilus of Antioch, Melito of Sardis, 
Apolinarius of Hierapolis, were bishops, but 
men of culture and philosophic training, well 
acquainted with heathen systems. Minucius 
Felix, author of what Renan calls "the pearl 
of the apologetic literature of the reign of 
M. Aurelius,"^ was a Roman advocate. Ter- 
tullian's learning, and legal and rhetorical 
gifts, I need not speak of. All this implies 
that Christianity had penetrated in no slight 
degree into the schools, and was exercising a 
powerful attraction on minds athirst for 
truth and certainty on the great questions 
of existence, as well as drawing into its 
service not a few of the gifted and earnest 
men of culture of the time. 

With this rise of a literary Apology for 
Christianity must be connected a yet more 
significant phenomenon in the Pagan world 
— the rise of a formal literary attack on 

' Marc-Aurele, ch. xxii. 


Christianity. It may be taken for granted 
that a religion must already have attracted 
considerable attention before the ablest lite- 
rary men of the time sit down to write 
elaborate refutations of it. I remarked before 
that if M. Aurelius kept silence about the 
Christians it was not because he did not 
know enough regarding them. He was sur- 
rounded with people who knew them well. 
Fronto of Cirta, the celebrated rhetorician, 
one of his tutors, and an intimate correspon- 
dent and friend, wrote a bitter attack on the 
Christians, which Renan thinks is repro- 
duced in the Octavius of Minucius Felix, ^ 
Diognetus, another of the tutors of Marcus, 
is probably the same to whom the beautiful 
Epistle to Diognetus is addressed. Herodes 
Atticus, yet another of his tutors, the 
wealthiest man and most famous orator of 
his time, had, as on the ground of a Cata- 

» Marc-Aurele, ch. xxii. 


comb inscription we have seen reason to 
believe, a daughter who was a Christian. 
Rusticus, the prefect of Marcus, presided 
at the trial of Justin and his companions. 
The policy of silence was, besides, no 
longer observed. I have just mentioned 
Fronto's written attack. Luclan satirised 
the Christians in his witty Peregrinus 
Proteus, which is in truth an honourable 
tribute to their charity. Celsus wrote his 
True Word (Aoyoc «^l^^c) '" refutation of 
their opinions.^ It is this work of Celsus 
which, above all, shows how important 
a phenomenon Christianity was now felt 
to be, and how carefully the writings of 
the Christians were being studied by some 
of their opponents. Here is a man of 
undeniable acuteness, of wide reading, of 

' Celsus' book has been largely reconstructed by 
Keim on the basis of the extracts and notices in 
igen. There is a good sketch of it in Pressense's 
^ Martyrs and Apologists, bk. iii. 


philosophic culture, of exceptional literary 
ability, who of deliberate purpose sets him- 
self down to assail, undermine, and over- 
throw Christianity by all the resources of 
knowledge, argument, and raillery, at his 
command. He sets about his work in no 
light spirit, but as one who feels that he must 
bend all his powers to attain his end. To 
fit himself for this task he makes a minute 
study of the Christian writings, keenly notes 
every assailable point, makes himself ac- 
quainted with the Christian beliefs, then, 
passing to the synagogue, gathers up all the 
slanders which Jewish malice could invent. 
It is fair to say that, like Pliny, he acquits 
the Christians of the grosser calumnies which 
were urged against them, but, short of this, 
he spares no pains to damage and discredit 
the sect And we may be sure that in this 
desire to know something about the Chris- 
tians and their literature Celsus did not 



comb inscriptior 

believe, a daug' 

Rusticus, the I 

at the trial of 

The policy < 

longer observ 

Fronto's writ 

the Christia 

Proteus, wh' 

tribute to t 

True Wora 

their opini' ^ 

which, at //,. 

a phenor .i„ 

to be, a ''/ ,1 

the ChT "'/,/ ^ '■■-. 

of the- I'K,,^ "'■■-.',_ 

unden , ''/ , 

Keim '*"*,/ "-' ^^_ 

Oriff '',, 

Mar "* 

» I 


"''''• /^^rnosticism is peculiarly the heresy of 

*"*''•'} -*. ad century. We can best judge of 

* •> '.....: of its influence, and the acuteness of 

*'* ' ' ij^, jS it evoked, by observing the extent 

•«»••"/«/./ 4,» ,h it bulks in the existing literature of 

'** '»'/«;,„ *'iod. The whole of Irenaeus, a great 

Ilk .jj^J Tertullian, the whole of Hippolytus 

"*'* * ''/ //y . t and not a little of Clement of Alex- 

* ** *'''* (tni^i ^^ devoted to its refutation. This 

•WL ^if^,^^ not take account of lost treatises. But 

'•» »\ ,j^ III ji^ -^ave only to consider the nature of this 

Jw ^,fn.f,^yt jlar appearance to see that it is one of 

" • ♦■ '*ii{iic -"^^^^ convincing testimonies we possess 

• --'w fijim fl^ -^^ power with which Christianity was 

"fluc '. letrating the innermost regions of thought 

^ ui (h* fk ^ speculation in the second century. This 

:.,,/ /^^ A A /^'^^ ^^^^ of the.subsect to which, as it seems 

'. ,,, .. . me, justice has not always been done. 

'*t* orl' 7 ^^^^^^y ^'S-y properly lays stress on Gnosti- 

ism as a phenomenon of the first importance 

' ^'' -'^Z a mn the early Church. He hits off its charac- 


teristic by describing it as an acute stage of 
that Hellenising of Christianity which after- 
wards was accomplished more gradually in 
the development of the Catholic dogma.' 
But, without discussing at present the justice 
of this view, it is surely obvious that if Gnos- 
ticism was on the one side an acute Hellen- 
ising — I should prefer to say Orientalising — 
of Christianity, it was not less on the other an 
acute Christianising of Hellenic and Oriental 
speculations. Gnosticism has this peculiarity, 
that it is the result of a blending of Christian 
ideas with the floating religious and theosophi- 
cal speculations of the time, especially those 
derived from an Oriental, or a mixed Greek 
and Oriental, source. It was a product which 
did not spring up spontaneously in the minds 
of the mechanics and slaves and women and 
children, whom most, like Celsus, suppose 
to have formed the bulk of the Christian 
' Cf. his Hist, of Dogma, I. p. 223 ff. (E.T.). 


communities, but could only have taken its 
rise in minds of a more cultured and specula- 
tive cast This, indeed, was its claim — to be 
a religion of "Gnosis," or knowledge, for the 
more highly trained or dite. It could only 
exist at all, therefore, as the result of a 
Christian ferment which had entered these 
speculative circles, and was there powerfully 
at work. Baur rightly appreciates the situa- 
tion when he says : — ^** Gnosticism gives 
the clearest proof that Christianity had now 
come to be one of the most important factors 
in the history of the time, and it shows 
especially what a mighty power of attraction 
the new Christian principles possessed for the 
highest intellectual life then to be found 
either in the Pagan or in the Jewish world." ^ 
Above all, these systems are a striking 
witness to the impression produced on the 
heathen mind by the great Christian idea 
« Ui$t of Church, II. p. i (E.T.). 


of Redemption. "When the Gnostic sys- 
tems," says Neander, "describe the move- 
ment which was produced in the kingdom 
of the Demiurge by the appearance of 
Christ as the manifestation of a new and 
mighty principle which had entered the 
precincts of this lower world, they give us to 
understand how powerful was the impression 
which the contemplation of the life of Christ, 
and of His influence on humanity, had left 
on the minds of the founders of these sys- 
tems, making all earlier institutions seem to 
them as nothing in comparison with Chris- 
tianity."^ We must beware, therefore, of 
underestimating either the extent or the in- 
tensity of this great intellectual ferment set 
up by the gospel in the heart of. heathenism. 
The Gnostic sects multiplied with extraordi- 
nary rapidity, and the influence exercised by 
their most renowned teachers, as Basilides 
' Hist of Church, II. p. 8 (Bohn). 


and Valentinus, was exceptionally great. 
The Church of the Marcio^ites — only, how- 
ever, partially Gnostic — long maintained its 
ground as an independent ecclesiastical 
organisation. ^ 

From the phenomena just considered in the 
sphere of the Church, we turn now to survey 
briefly certain scarcely less striking facts 
which meet us on the ground of Paganism. 
It is well understood that the second century 
was an age of ethical and religious revival ; 
but it is not always realised how powerful 
this current of revival was, and how remark- 
able were some of the forms which it as- 
sumed. I have said that this age of the 
Antonines was an age of lecturing, preach- 
ing, teaching, and declaiming, beyond all 
precedent. From the time of Vespasian the 
Empire had been provided with a hierarchy of 
rhetoricans and grammarians, whose business 
« Cf. Diet of Christ Biog., III. p. 819. 


it was to instruct the people in all liberal 
arts ; and society was overrun with profes- 
sional talkers, debaters, moralists, ready to 
supply oratory on any subject to whoever 
cared to pay for it There was little in this 
sophistic declamation to make the world 
wiser and better ; yet it is undeniable that 
towards the end of the first, and during the 
course of the second century, a certain glow 
of moral enthusiasm b^an to spread itself 
through the Empire, accompanied by a 
manifest revival of religious faith and earnest- 
ness.' In some of its representatives this 
fervour rose almost to a kind of Apostolic 
zeal, "It is too often foi|[otten," says Renan, 
" that the second century had a veritable Pi^an 
preaching similar to that of Christianity, and 
in many respects in accord with the latter," " 

' On tbis religious revival in the second centui;, 
see Friediander, Sitiengeschichte, III. pp. 430-33 ; 
Bigg's Ckrislian Plalonisls of Alex., pp. 23 ft. 

' Mare-Aur'de, chap. iii. p. 45. Cf. Lightfoot, 
Ignatius, p. 449. 


An early type of this species of " itinerant 
homilists," as Merivale names them, " who 
b^an from the Flavian period to go about 
proclaiming moral truths, collecting groups 
of hearers, and sowing the seeds of spiritual 
wisdom and knowledge on every soil that 
could receive it,"* was Apollonius of Tyana, 
to whose gifts of teaching was added the 
repute of miraculous powers. » Other and 
loftier types of this Pagan ministry are the 
celebrated Dion Chrysostom.s in the reigns 
of Nerva and Trajan, and Maximus of Tyre 

' Romans under the Empire, chap. Ixvi. 

' His life, with romantic embellishments, was 
written by Philostratus at the request of the Empress 
Julia Domna, a.d. 217. See Newman's sketch of it in 
his Life of Apollonius Tyanaus, and Bi^s Christian 
Platonists, pp. 243-247. 

' On Dion, eighty of whose orations remain to us, 
see the interesting sketch in Merivale, Romans under 
the Empire, chap. Iwi, " The name of Clirysostom," 
he says, " may have already reminded us of the most 
illustrious fllittHttUAll^^bi^W orators, and his 


under the Antonines. ^ Epictetus, the greatest 
name in the history of Stoicism after Seneca, 
is the noblest representative of the movement 
on its earnest philosophic side. With all this 
went on, as the accompaniment and counter- 
part of these better features, a vast develop- 
ment of superstition, an inrush of Oriental 
cults, a craving for theurgy and mysteries, a 
general susceptibility to dupery, giving rise 
to such characters, as Alexander of Abono- 
tichus, the most stupendous example, per- 
haps, of successful charlatanry in history. « 

may be compared, with little disadvantage, with the 
sermons of the Bishop of Constantinople, for their 
warm appeals both to the heart and the conscience 
of their hearers." 

' Forty-one orations of Maximus are preserved. 
On him and the others see Hatch, Hihbert Lectures ^ 
pp. 6-242, &c. The sketch of iElius in Friedlander, 
III. p. 440, may also be consulted. 

» Cf . Froude's " A Cagliostro of the Second Century " 
in Short Studies j vol. iv. In this and other sketches 
in vols. iii. and iv., Froude gives admirable charac- 
terisations of the period. 



What, now, are we to say of this remark- 
able revival movement in second century 
heathenism, and, in particular, can it be 
affirmed that Christianity had anything to 
do with it ? The majority of writers would 
probably answer — No. I cannot, however, 
share this view. It seems to xx\<t primd facte 
unreasonable that, in summing up the forces 
which helped to give the age its character, 
we should take account of every stray in- 
fluence from East to West — of Epicureanism, 
of Stoicism, of Pythagoreanism, of Isis- and 
Mithras-worship, of an Apollonius of Tyana, 
of a Dion Chrysostom, of charlatans even 
like Alexander of Abonotichus ; but that no 
influence whatever should be attributed, or 
allowed, to this constantly present and in- 
tensely active force of the Christian religion. 
It is true that Christianity was persecuted, 
was regarded with contempt and scorn, but 
we must not be deceived by this into sup- 


posing that its influence was not telling 
silently and secretly on multitudes in the 
Empire, and that it was not affecting Pagan- 
ism in many indirect ways, even where the 
obligation to it was not openly acknowledged. 
We saw before that Epictetus alludes to it 
but once, and with contempt, but there is 
good reason for believing that he was not 
unacquainted with its Scriptures or unin- 
fluenced by its teaching. 

I believe that we profoundly err in assum- 
ing that the borrowing of ideas and moulding 
of institutions in this age was all on the 
part of the Christian Church, and that a very 
considerable influence was not going out also 
from the Christian Church on the religion 
and life of Paganism. Dr. Hatch, for in- 
stance, would see in the lecturing and de- 
claiming of this rhetorical age the origin of the 
Christian sermon.' But might we not, with 

* Hibbert Lectures, p. iij. 


equal reason, reverse the supposition ? Is it 
not, at least, as likely that the example of the 

Christian Church, its unceasing and intensely 
zealous propaganda, extending now over 
more than a century, and presenting so 
splendid an example of success, had some- 
thing to do with kindling the enthusiasm 
and quickening the Apostolic zeal of such 
itinerant preachers as Dion and Maximus? 
Take the picture of that Christian propa- 
ganda as furnished by so sober a pen as 
Friedlander's. "The example of the first 
Apostles," he says, "unceasingly stirred up 
imitators in constantly increasing number, 
who, according to the doctrine of the gospel, 
shared their possessions with the poor, and 
grasped the travelling-staff in order to carry 
the Word of God from people to people, 
and whose zeal neither wearied nor grew 
cold under the greatest difficulties and 
dangers. The Christians were zealous (says 


Origen) to sow the seed of the Word in the 
whole world* The messengers of the new 
doctrine visited not only cities, but also 
villages and farms; nay, did not shun to 
force themselves into the interior of families, 
and to place themselves between those re- 
lated by blood." ' The success which 
attended this zealous gospel preaching in 
Rome, in Bithynia, in Carthage, in Antioch, 
everywhere, we have already seen, and it was 
a constant object-lesson to the Pagans, who 
felt their own faith crumbling, and were 
looking round for means with which to 
combat the victorious progfress of the new 
religion which emptied their temples, and 
made even the purchase of sacrifices to 
cease. Can we believe, then, that it had 
nothing to do with awakening their emula- 
tion, and inciting them to a similar propa- 

* Sittengeschicktet III. p. 517. 


gandism ? ' Their silence and contempt go 
for nothing. When Maximin = and Julian 
conceived the idea of re-modelling the Pa^n 
priesthood as a set-off to the Christian 
hierarchy, they did not proclaim in so many 
words that it was this hated sect they were 
imitating, any more than the Anglican 
Church, when the Evangelical Revival was 
pouring new life into its veins, made public 
acknowledgment of its indebtedness to 
Wesley and Whitefield. In a similar way, 
it is no disproof of the manifold influences 
with which Christianity was bathing the 
Paganism of the second century, that the 
recipients of the benefit do not acknowledge 

■ Merivale points out that Dion Chrysostom had 
probably a connection with Flavius Clemens, the 
consul, who suffered for his faith under Domitian. 

" Cf. Euseb, ix. 4. " Maximin perceived the 
power tiiat existed in the Catholic Church with its 
wonderful organisation, and conceived the stupendous 
idea of rejuvenating Paganism by creating a Pagan 
Catholic Church " (McGiffert's note .. 


the source from which it comes. We see the 
change that is in process ; we mark the new 
spirit and the energetic propaganda ; we 
know that this has come into existence with 
the example of the Christian Church before 
it, and the influences of the Christian faith 
permeating every pore of the old system ; 
and we think it not unreasonable to suppose 
that there is a connection between the facts. 

This influence of Christianity which we 
indicate is probable in itself, and the pre- 
sumption in its favour is strengthened when 
we consider certain other features in the 
religious condition of the age, which it is 
difficult to avoid tracing in some measure 
to Christian influence. It has been hinted 
above that the second century was an age 
not only of ethical, but of religious revival. 
It was an age characterised by a new sense 
of sin and weakness, by a longing for re- 
demption from these evils, by a yearning for 


immediate communion with the Deity, by 
the craving for the assurance of a blessed 
life hereafter. Outwardly, it was marked by 
a great influx of foreign cults, and specially 
by the introduction of new forms of heathen 
mysteries, and the extraordinary develop- 
ment and rapid spread of the latter. The 
chief were those of the Phrygian Cybele, of 
the Egyptian Isis . and Osiris, and of the 
Persian Mithras — the types which promised 
most satisfaction to the cravings referred 
to.i The effects of these mysteries on 
the ideas and usages of the Christian 
Church have been traced by Dr. Hatch and 
others, not without some exaggeration,' but 

' Cf. on this subject of the mysteries, Boissier's La 
Religion Romaine, bk. ii., chap. ii. ; Anrich's Die 
Antike Mysterienwesen ; Hatch's Hibbert Lectures, x. ; 
Cheetham's Mysteries : Pagan and Christian ; Bigg's 
Christian Platonists. The older and newer literature 
may be seen in Cheetham. 

° Cheetham's book deals with some instances of 



the counter-question of a possible influence 
of Christianity on the mysteries has received 
but scant attention. Yet I believe there is 
an important field to be Worked here also. 
The mysteries, especially those which sprang 
up in the time of the Empire, deserve the 
closest study we can give them. They 
represent, as Renan has said, the most 
serious phase of Pagan religion — are, in a 
sense, the underground Church of Heathen- 
ism, But it is necessary also to study them 
with discrimination, and to distinguish 
carefully times and seasons, and the suc- 
cessive stages of development. When we 
do this, we discover that the special period 
of growth of these new cults is from the 
second century onwards ' ; that there are 

' " This new development of the mysteries,'' says 
Anrich, '' is conditioned by the re-awakening of the 
religious life which exhibits its slight beginnings in 
the first century, in order from that steadily to in- 
crease, till in the third century this new type of 


certain ground-features in which they all 
agree; but that their distinctive forms, 
rites, and terminology are often a later 
formation, and are strongly affected by the 
conditions of the age. Among these con- 
ditions we do not think it unreasonable that 
the Christian Church — the most formidable 
rival of the mysteries in the Empire — should 
be included. It is certain that the Fathers 
of the Church held the mysteries in abhor- 
rence, and that whatever borrowing took 
place from these on the Christian side was 
unconscious, and in a sense involuntary.' 


religious tendency, essentially difFerent from the piety 
of earlier times, has become the all-controlling power 
of the age " (p. 35). He observes how the Isis-cult had 
a rapid development after the middle of the second 
century, and in the third century "was perhaps most 
widely spread, and at all events the most important 
religion of the Roman Empire," and how the Mithras- 
cult, which begins to spread in the first century 
** reached, however, first in the age of Diocletian and 
Constantine its highest bloom " (pp. 43-45). 
' Anrich, p. 235 ; Cheetham, p. 78. 


But it is not so unlikely that the patrons 
of the mysteries should adopt terms and 
features from the language and worship of 
the Christian Church. And that they 
actually did so seems the simplest expla- 
nation of various striking facts. I may 
refer to the prominence given in the later 
mysteries to the idea of the awrripy and the 
description of the promised blessing as 
(Tuyrripla^ — for though these terms are not 
new in Paganism, they are brought into new 
connections, and acquire a deeper significance 
in the age we are speaking of. So again, 
we have the use of such terms as renatus^ or 
renatus in eternum? to designate the initiated 
person ; we have new expiatory rites, cul- 
minating in the hideous Taurobolium 3 ; we 

* Cf. Anrich, pp. 47, 49. Mithras came to be 
called <T(ur4p. On the older usage, see Cheetham, 

p. 15. 
' See passages in Anricb, pp. 47, 53, &c. 

3 Ibid. pp. 51-2 ; cf. Bigg, pp. 238. 


have curious resemblances to the Christian 
Sacraments,^ which the early Christian 
writers could only explain by supposing that 
the demons had invented a caricature of 
Christian ordinances with the view of 
throwing discredit on the latter * ; we have 
the further developments of the initiation 
of very young children — a sort of infant 
baptism 3 — and, in general, the growth of 
something like a Church idea in these secret 
celebrations.4 When we remember the hold 
taken on Gnostic minds by the ideas of the 
(Tbtfrnp and of redemption, and reflect on 
the half-pagan character and wide diffusion 
of many of the Gnostic sects, we may 

* Cf. Harnack, as below. These resemblances 
were chiefly in the Mithras- cult. 

' Cf . Justin, Apol. i. 66 ; Tert. De Press. Hcsr,, 40 ; 
and see Harnack, Hist, of Dogma, I. p. 118 (E.T.). 
3 Anrich, p. 55. 

* Ibid., p. 56. " The worship of Mithras in the 
third century," says Dr. Harnack, ** became the most 
powerful rival of Christianity " (I, p. 118, E.T.), 



perhaps — even apart from the direct influence 
of the Church — ^see a channel through which 
the ideas of Christianity might filter into 
purely Pagan circles. Dr. Bigg, in his 
lectures on The Christian Ptatonisfs of Alex- 
andria has some suggestive remarks on this 
point, which are, I think, in the main correct 
" The disciples of Mithra," he says, " formed 
an organised hierarchy. They possessed 
the ideas of mediation, atonement, and a 
Saviour, who is human and yet divine, and 
not only the idea, but a doctrine of the 
future life. They had a Eucharist and a 
baptism, and other curious analogies might 
be pointed out between their system and 
the Church of Christ Most of these con- 
ceptions, no doubt, are integral parts of a 
religion much older than Christianity [some 
of them on the other hand do not seem to go 
beyond the second or third century]. But 
when we consider how strange they are to 

I w. 


the older polytheism of Greece and Rome, 
and when we observe further that Mithra- 
ism did not come into vogue till the time 
of Hadrian, that is to say, till the age of 
Gnosticism, we shall hardly be wrong in 
judging that resemblances were pushed 
forward, exaggerated, modified with a special 
view to the necessities of the conflict 
with the new faith, and that differences, 
such as the barbarous superstitions of the 
Avesta, were kept sedulously in the back- 
ground with the same object. Paganism 
was copying Christianity, and by that very 
act was lowering her arms." ' 

Fully to estimate the force of these con- 
siderations, it is necessary to bear in mind 
the evidence which has already been adduced 
as to the extent to which Christianity had 
penetrated literary circles, and had become 
known to learned opponents, who employed 

* P. 240 ; cf. Cheetham, pp. 77, 146. 


their best abilities to discredit and destroy it 
— ^in vain. 

Much might be said of the writings of the 
old Catholic Fathers, and especially of the 
famed Alexandrian school, in illustration of 
the subject we are considering. For how 
clear is the evidence in the writings of these 
Fathers of the hold that Christianity had 
taken of men of the most powerful intelli- 
gence and widest learning ; how plain the 
indications that it had become a subject of 
the profoundest theological reflection ; how 
complete its victory over the brilliant Auf- 
kldrung of Gnosticism ; how evident the 
alliance which had been effected between it 
and the best elements of the Greek wisdom — 
that revelation in reason which the Alex- 
andrian Fathers, with Justin, traced to the 
illumination of the pre-incamate Logos ! If 
the school of Harnack sees in the theologica 
movement of this period an amalgamation of 

• V- ■ --^ i 


Greek intellectualism with Christianity, it 
must in consistency recognise that the pre- 
vailing Greek spirit had been seized and was 
being led captive by the new faith. And 
that of itself speaks to a mighty internal force 
of assimilation. Without, however, dwelling 
on this, I hasten to speak of what is, after all, 
perhaps the most striking proof of the influ- 
ence of these Fathers of the Early Church on 
contemporary religious thought — I mean the 
rise of Neo-Platonism in the third century. 
In this century, as we previously saw, the 
river whose course we have been tracing flows 
no longer underground, but comes to the 
light of open day. The fact of a Christian 
influence on the intellectual currents of the 
age is all too patent to be further denied. I 
referred in the previous lecture to the eclectic 
temper of this age, and to its characteristic 
embodiments in Julia Domna, the talented 
wife of the Emperor Septimus Severus, and 


in Alexander Severus, a succeeding emperor. 
Julia Domna gathered round her at her court 
a brilliant literary circle. It was at her com- 
mand that Philostratus wrote the "Life of 
ApoUonius *' ' — partly, there is reason to 
believe, as a parallel to the representation of 
the life of Jesus in the Gospels. Alexander 
Severus went further, and, as formerly nar- 
rated, placed the statue of Christ along with 
those of Abraham, Pythagoras, and others in 
his lararium, besides inscribing the Golden 
Rule on his walls and monuments. "Men 
sought," says Dr. Bigg, "to distil an elixir 
from all religions, from all, that is, except 
Christianity, which they never name" — a 
statement which needs to be slightly quali- 
fied. " Yet," he goes on, " the church from 
which they avert their eyes as from the angel 

^ Cf. Classical Dictionary, Art. " Philostratus" ; 
Newman's Apollonius Tyanceus ; Bigg's Christian 
Plaionists, p. 246. 



of doom, is really the prompter and guide of 
all their efforts." ^ Under these intellectual 
and spiritual conditions, arose the new form 
of opposition to Christianity which we de- 
nominate Neo-Platonism. The founder of 
this school, Ammonius Saccas — ^whose lec- 
tures Origen for a time attended at Alex- 
andria — was born of Christian parents, and, 
indeed, for a time himself professed Chris- 
tianity.2 Here is proof, if such were needed, 
of a strain of Christian influence entering 
into Neo-Platonism at the commencement. 
Ammonius, it should be remarked, had an 
important precursor in the second century, 
Numenius, who likewise was moulded by 
Jewish and Christian influences.3 The ideas 
of the founder were developed by his more 

* Christian Platonists, p. 242. 

" Eusebius, Ecc, Hist. vi. 19. Eusebius will not 
admit that he ever apostatised, but this is evidently 
a mistake. 

3 Cf. Bigg, pp. 251-3. 


famous pupil Plotinus, and carried still 
further by the third great teacher Porphyry. 
The Neo-Platonic system thus developed, 
while bitterly hostile to Christianity, is really 
the strongest testimony to its power. It not 
only shows upon itself the distinct mark of 
Christian ideas — e.g, in its doctrine of the 
Trinity, respecting which Bigg truly remarks 
— " It may be confidently affirmed that no 
Trinity is to be found in any Pagan philo- 
sopher who was not well acquainted with 
Christianity " ^ ; it was not only, as Schaff 
has observed, " a direct attempt of the more 
intelligent and earnest heathenism to rally all 
its nobler energies, especially the forces of 
Hellenic and Oriental Mysticism, and to 
found a universal religion, a Pagan counter- 
part to the Christian " * ; but it testifies to the 
changed attitude towards Christianity in the 

' Cf. Bigg, p. 250. 

^ Church Hist, (Ante-Nic), p. 99. 


fact that it no longer poured unqualified 
ridicule on the new religion as Celsus had 
done, but dealt with it rather in the philoso- 
phical eclectic spirit characteristic of the time, 
condemning only its exclusive claims. It 
reckoned Christ among the sages ; professed 
respect for His personal teaching, as con- 
trasted with the corrupted doctrines of His 
Apostles, and sought to appropriate its 
spiritual elements to itself. " The Neo- 
Platonists," says Augustine, "praised Christ, 
while they disparaged Christianity." "We 
must not," said Porphyry himself, "calum- 
niate Christ, but only those who worship 
Him as God." ' But the battle was a hope- 
less one. " Under the banner of Neo- 
Platonism," says Dr. Lightfoot, "and with 
weapons forged in the armoury of Chris- 
tianity itself, the contest is renewed. But 
the day of heathenism is past This new 
' Augustine, City of God, xix. 23. 

_«..^j.. ■ . 


champion retires from the field of conflict in 
confusion, and the gospel remains in posses- 
sion of the field." ' 

Here I must close. Other parts of the 
field I am compelled to leave well-nigh 
untouched, especially that relating to the 
influence of Christianity on social life and 
legislation. This, at the same time, is the 
part of the subject which has been least 
neglected. It has often been shown with 
abundance of illustration how revolutionary 
were the ideas and principles of the holy and 
spiritual religion which had its birth in Judaea 
when introduced into the unspeakably corrupt 
society of the Graeco-Roman Empire.^ To 
the profligacy of that effete heathen world, 
Christianity opposed its own fresh, young 
life, and glowing spiritual ideals ; to its pride, 

' Philippians, p. 319. 

' Cf. the works of Troplong, Schmidt, Uhlhorn, 
Lecky, Loring Brace, with the histories of Milman, 
Pressense, Schafif, &c. 


the proclamation of a common fall and a 
common salvation ; to its selfish egoism, the 
demand for a universal charity ; to its denial 
of the rights of humanity, the doctrine of the 
love of God, and of the spiritual dignity of 
man as made in the image of God ; to its 
degradation of woman, the assertion that 
in Christ Jesus there is neither Jew nor 
Greek, male nor female, bond nor free ' ; 
to its contempt for labour, the recollection 
of the Carpenter, and the injunction " Take 
thought for things honourable in the sight 
of all men."* Opposed at nearly every 
point to the existing Pagan order, it yet 
gave to the world of that time exactly 
what it needed, implanted within it the 
seeds of emancipation and renewal. If it 
could not save the old Roman Empire, it at 
least laid within it the foundations on which 
the rearing of a new order could proceed — 
Gal. iii. 28. ' Rom. xii. 17. 


rendered possible the rise of a rejuvenated 
and progressive Europe. The pure morals 
and blameless, self-denying lives of the Chris- 
tians, were the strongest points the Apologists 
for the new religion could urge in its favour. 
Thus Tertullian powerfully contrasts the 
private virtues and public morality of his 
fellow-believers with the foul conduct of the 
Pagans, and challenges his opponents to 
produce instances of Christians in the long 
list of those committed to prison for their 
crimes.' If there were exceptions, it was 
only as it must happen to the healthiest and 
purest body, that a mole should grow, or a 
wart arise on it, or freckles disfigure it* The 
heathen themselves bore involuntary testi- 
mony to the superior excellence of the 
Christian character by appealing to it in 
rebuke of the lack of virtue in one another.3 

' Apol., 42-46 ; Ad Nat. i. 4. * Ad Nat, i. 5. 

3 " You are accustomed in conversation yourselves 


As respects legislation, naturally little could 
be done till the Empire had become publicly 
Christian, but with Constantine we have 
already numerous enactments which show 
the new spirit that had entered society,^ and 
under the succeeding emperors these evi- 
dences of Christian influence are multiplied.^ 
The Theodosian code is little more than a 
compilation of the decisions of the Christian 
emperors. Even in the earlier period, it 
is not wholly unreasonable to see in the 
gradual ameliorations introduced into many 
of the laws under the influence of the newer 
Stoicism an indirect result, at least in part, 

to say, why is so-and-so so deceitful, when the Chris- 
tians are so self-denying ? why merciless, when they 
are so merciful ? " &c. — Ad Nat. i. 5. 

* Julian termed Constantine, "Novator turbatorque 
priscarum legum et moris antiquitus recepti " (Amm. 
Marc. xxi. 10). See a sketch of his reforms in the 
laws relating to women, children, slaves, &c., in Did, 
of Christ Biog, I. pp. 636-7. 

> See Loring Brace's Gesta Chrisli, passim. 



of that atmosphere of mercy with which 
the Christian Church was already bathing 

In leaving the subject, I can only express 
the hope that these lectures, however im- 
perfect, may have done something to intensify 
our sense of the mighty power which, as 
the Divine Leaven introduced into humanity, 
Christianity from its first entrance into the 
world exercised on everything it touched, 
and to guard against the tendency, still too 
prevalent, unduly to minimise its influence. 

* Cf, Troplong, L Influence du Christ, p. 83, &c. 




{in Frontispiece). 

FOR the drawings of these inscriptions, I am 
indebted to the Rev. Archibald Paterson, 
B.D., Rosslyn, who from his first-hand researches 
kindly furnished me with information regarding 
them. They should, he thinks, probably be 
restored and identified as follows : — 


AciLio Glabrioni Acilio Glabrioni 

FiLio or FiLio 

M' AciLii Glabrionis M' Acilius Glabrio 

Cos. Pater. 



Deceased may have been the son of Manius 
Acilius Glabrio, consul in a.d. 124, the latter 
probably being the son of the consul of a.d. 
91, who suffered under Domitian, in a.d. 95. 


Manius Acilius Verus 
Clarissimus Vir, 

1 PUELLA (?) 

ST (?) Priscilla Clarissima < 

^ ( Femina (?) 

May be the children of Manius Acilius Glabrio, 
consul in a.d. 152 [son of the consul of a.d. 
124], and Vera Priscilla, who is known from an 
inscription to have been the wife of a Manius 
Acilius Glabrio. The Manius Acilius Glabrio 
who was her husband may, however, have been 
the consul of this name in a.d. 186 [son of 
the consul of a.d. 152]. In this case the children 
will be their offspring. 


A fragment probably of— 


M[arci] Acilii, 
belonging to the family of Marcus Acilius Vibius 


Faustinus, who was one of the Salii before a.d. 
170; or to the family of Marcus Acilius Priscus 
Egrilius Plarianus, who lived at the same time. 


KXavSmv Axeiktov OvaXiptov IXa/irrporarov'] vBavuTKov, 

It is known that Claudius Acilius Cleoboles 
(grandson of the consul of a.d. 186) derived 
his name Claudius from adoption by Tiberius 
Claudius Cleoboles, consul suffectus (year un- 
certain). The name Valerius is from the mother's 
side. The inscription cannot be earlier than the 
third century. 



Abercius, of Hieropolis, 54 
Acilius Glabrio, consul, 125 
fF. ; inscriptions of Acilii, 
frontispiece, and appen- 
dix, 227 
iElius Aristides, 167, 202 
Alexandria, Church of, 31, 
60 ; mixed state of, 66-7 ; 
school in, 68, 216 ; wealth 
of, 140 ff. 
Alexander Severus, emperor, 

147, 218 
Alexander of Abonotichus, 

51, 202-3 
Ambrose, of Milan, 158 
Ammonius Saccas, 219 
Anrich, on Mysteries, 209- 

13, passim 
Antioch, Church of, its num- 
bers, 73-83 
ApoUonius, senator, 137, 

Apology, of 2nd century, 34, 

58 ; its significance, 185- 

90; 224 
Aristides, apology of, 45 
Armenia, conversion of, 88 

Baur, 15 ; on apologists, 
188 ; on Gnosticism, 197 

Bigg, C, 146, 194-5. 209- 
20, passim 

Bithynia - Pontus, Pliny's 
testimony, 48 ff. ; 129, 

Blandina, martyr, 137 

Boissier, G., on spread of 
Christianity, 27ff.; ignoring 
of Christianity by pagans, 
168, 170 ; influence of 
Christianity, 175, 183 

Brace, Loring, on influence 
of Christianity, 19, 225 




Catacombs, character and 




extent of, 35 ff. ; testi- 
mony to numbers of 
Christians, 33-5, 39-41 ; 
inscriptions, 98 ; testimony 
to wealth of Christians, 
113 ff., 132 ff. ; cemeteries 
of Lucina, 118; of Domi- 
tilla, 123 ; of the Acilii, 
126 (with frontispiece 
and appendix) ; of Prse- 
textatus, 132 ; of Caedlia, 
134 ff. 

Cappadocia, Church in, 56 

Caracalla, emperor, 146 

Carthage, Church of, 30 ; 
numbers, 61 ff. ; rank and 
wealth, 139-43; martyr- 
dom of Perpetua, 143-4 ; 

Celsus, his True Wordy 59, 
71, 99, 167, 192-4 ; 196, 

Chastel, on nos. of Chris- 
tians, 24 

Cheetham, on Mysteries, 

Christianity, influence of 
paganism on, 18, 163, 
204, 209 ; effects of 
Christianity on paganism, 
20 ff. ; on pagan preach- 
ing, 204 ff. ; on Mysteries, 
210 ff. ; on morals and 
legislation, 222 ff. 

Chrysostom, on Church of 
Antioch, 75 ff.; 158 

Churches in Rome, Corinth, 
Ephesus, Galatia, An- 
tioch, Bithynia, Pontus, 
C3nrene, Alexandria, Car- 
thage, Gaul, Spain, &c. — 
See under these heads. 

Claudius, emperor, his edict, 

Clement, of Rome, 43 

Clement, of Alexandria, 
on luxury of Christians, 
141-2 ; Gnosticism, 195 ; 

Commodus, emperor, 145 

Corinth, Church of, 32, 107, 
108-9, 130* 171-2 

Cyprian, of Carthage, 86, 
151, 158 

Cyrene, Church in, 31 

Decian Persecution, 86, 
149 ff. 

De Rossi, G. B., on Cata- 
combs, 34, 117, 123, 135, 

i75» I79» 183 

De Rossi, Michele, measure- 
ments of, 37 ff. 

Diocletian, emperor,* perse- 
cution of, 87 ff. ; 152 ff. 

Dion Chrysostom, 201, 205 

Dionysius, of Alexandria, 
148, 150, 158 

Domitilla, Flavia, 122 ff.; 
cemetery of, 123 

Domitian, emperor, perse- 
cution of, 121 ff. 



Elagabulus, emperor, 147 
Elvira, Council of, 86, 157 
Emperors, see under names. 
Ephesus, Church of, 44, 108, 

1 7 1-2 
Epictetus, 183, 204 
Eusebius, references to, 48, 

5i» 54, 56, 59, 73-4, 78, 
82, 87-90, 123, 130, 137, 
145-^, 148, 152-3, 154-^, 

Farrar, F. W., 141, 177 

Flavius Clemens, consul, 
his martyrdom, 122 ff. 

Friedlander, on numbers of 
Christians, 24, 116 ; on 
obscurity of Church, 165- 
6; 177, 179, 183, 200, 
202 ; on the Christian 
propaganda, 205 

Froude, J. A. 202 

Galatia, Church in, 55, 

57, 171 
Gaul, Christianity in, 30, 

81 ff. 
Gibbon, on numbers of Chris- 
tians, 24, 39, (i&, *J% ff. ; 

47, 51, 122 
Gnosticism, 58, 184, 194 ff., 

213, 215-16 
Greek Spirit and Christianity, 

17, 164, 187, 196 
Gregory Thaumaturgus, 52, 


Hadrian, emperor, 66-8 
Hamack, quoted, 116, 124, 

133, 163, 213 ; referred to, 

118, 123, 128, 164, 195, 

Hasenclever, on rank of 

Christians, 64, 118; 177 
Hatch, E., 16, 185, 202, 

204, 209 
Hefele, on Councils, 84, 86 
Hermas, on Roman Church, 

Herod Atticus, 133, 19 1-2 

Ignatius, 59, 73, 129 
Imperial Court, Christianity 

in the, 116, 124, 145 fF. 
Irenaeus, of Lyons, 59, 64, 

82, 146, 195 

Julia Domna, her influence, 
146, 217 

Julian, emperor, 74, 77, 
207, 225 

Justin Martyr, on spread of 
Christianity, 47 ; on wealth 
of Church, 132 ; as apolo- 
gist, 186-9 5 192, 213, 2 16 

Kbim, on progress of Chris' 
tianity, 27 

Lanciani, on Catacombs, 
36 ff., 115, 123, 133, 
134, 152 ; on Hadrian, 68 

Lightfoot, Bishop, referred 
to or quoted, 16, 43, 45, 




48» SO. 54, 55. S8, 67. 
116, 118, 121-3, 126-7, 

129» 130, 134-S. 173. 
177^80, 182, 221 

Lacian, of Antioch, martyr, 

on difiiision of Christianity, 

Lucian, on Pontus, 51 ; his 

Peregrtnus Proteus^ 192 
Lyons, its importance, 82-3 ; 

martyrdoms, 82, 136 

Marcus Aurelius, em- 
peror, 133, 189; relations 
to Christianity, 166, 171, 

Martyrdoms, of Ignatius, 
Polycarp, Justin, at 
Vienne and Lyons, Per- 
petua, &c., see under 

Maximin, emperor, his per- 
secution and letter, 89, 207 

Maximus, of Tyre, 201-2, 

Merivale, referred to or 
quoted, 32, 1 12-13, 133, 
169, 185, 201 

Milman, 82, 90, 142 

Mithras-cult, 209, 211-13, 

Mommsen, 50, 55, 85, 116 

Mosheim, 47 

M3rsteries, in Roman Empire, 
2Q9 ; possible influence of 
Christianity on, 210 if. 

Neandbr, 155, 198 
Neo-Platonism, 149, 217 if. 
Nero, emperor, persecution 
of, 43 ; his palace, 116, 

Neumann, 17, 42, 64, 138 

Northcote and Brownlow, 

on Catacombs, 36 ff., 

114 ff., 132 ff. 

ORiGBN,on numbers of Chris- 
tians, 68 ff. ; intelligence 
of, 98-9, 144 ; relations 
with Court, 145-9, 158, 

Paganism, ethical revival 
in, 199 ff. ; pagan preach- 
ing, 200-2 ; religious re- 
vival, 208 ff. 

Pamphilus, of Caesarea, 158 

Paul and Thecla, Acts of, 128 

Perpetua and her com- 
panions, 143 ff. 

Persecutions under Nero, 
Domitian, Trajan, Mar- 
cus Aurelius, Severus, 
Decius, Valerian, Diocle- 
tian, Maximin, see these 

Phrygia, Christianity in, 53, 

Pliny, correspondence with 
Trajan, 28, 48 ff., 129, 

167-8, 193 
Pomponia Graecina, 117, 136, 



Pontus, 51 

Porphyry, 220-1 

Pudens and Claudia, 1 19-21 

Ramsay, W. M., referred to 
or quoted, 17, 31, 42, 48, 
53-5» 96, 105, 116, 128, 
167-8, 182 

Renan, 53, 83, 191, 200, 210 

Ritschl, 16 

Robertson, Canon, 27 

Roman Empire, population 
of, 24-5, 29 

Rome, Church of, 4i-44» 
78-9, 129-36, 137, 139, 
206. See Catacombs. 

ScHAFP, Philip, 34, 38, 220 

Schultze, Vict., 31, 32 ; on 

numbers of Christians, 

24-5. 43> 5i» 56-7, 65, 75, 
83-6; on rank of Chris- 
tians, 99, 112 

Seneca, relation to Chris- 
tianity, 175 ff. ; Epistle to 
Lucilius, 180 

Septimus Severus, emperor, 
persecution of, 138 ff. ; 146 

Silence of Pagan writers 
on Christianity : how ac- 
counted for, 166 ff. 

Social rank of Christ's per- 
sonal disciples, 100 ff. ; of 
early converts, 104 ff. 

Spain, Church in, 85 ff., 157 

Tacitus, 28, 43, 117 

Tertullian, on numbers of 
Christians, 28,61 ff.; rank of, 
139,141-2 ; on Gnosticism, 
195 ; the Mysteries, 213, 
31, 48, 84, 86, 147, 224 

Trajan, emperor, persecution 
under, 48, 129 

Tryphsena, Queen, 128 

Uhlhorn, 26, 59, 146, 194 
Urania, daughter of Herod 
Atticus, 133, 192 

Valerian, emperor, perse- 
cution of, 148, 151 

Vienne and Lyons, martyr- 
doms, 82, 136 

WiTHROW, 38, 116