THE PLATONISM OF ARIUS

M Y subject involves two long-standing difficulties. First, there is

no complete agreement about the theological antecedents of
Anus. Recent articles1 by Dr. Pollard and Mr. Wiles have not
closed the debate whether Arianism can be explained as a derivative of
Origenism, or whether some independent influence must be recognized.
No doubt Arius, like most Eastern theologians of his generation, was

Downloaded from http://jts.oxfordjournals.org/ at Nipissing University on October 14, 2014

indebted to Lucian; but we know too little of Lucian to estimate the
debt. We cannot gauge the truth of the traditional charge2 that Arius
drew upon the teaching of Paul of Samosata, the detested opponent of
the Origenists of his day.
Secondly, we know too little about the philosophy current at Alex-
andria during this period. We cannot bridge the gap between the depar-
ture of Plotinus (A.D. 244-6) and the age of Synesius, Hypatia, and
Hierocles. There are slight indications of philosophical writings by
Bishop Dionysius and by the Christian mathematician Anatolius in the
mid-third century; for the end of that century and most of the fourth,
the silence is virtually unbroken. An authoritative article3 by Professor
Marrou has exactly revealed the extent of our ignorance.
In these circumstances I think we may learn something by consider-
ing the philosophical position of Arius himself; especially if there
should prove to be common ground between Arius and his theological
opponents Alexander and Athanasius. Pe"re Henry, in his Sarum
Lectures given in Oxford, spoke of Arius' 'philosophical conformism'.
(I refer with some diffidence to these still unpublished lectures; but my
indebtedness to this distinguished scholar and teacher must be acknow-
ledged.) I think he is right in suggesting that Arius placed more
reliance on philosophical and dialectical techniques than either of his
great critics; and the arresting phrase need not be taken to mean that
there was an established philosophical tradition to which Arius had only
to conform; still less to identify this with the teaching of Plotinus
and Porphyry. As I see it, Arius had little in common with them;

1
T.E. Pollard, "The Origins of Arianism', Jt.r.S.,N.8.,ix (1958), pp. 103-n;
M. F. Wiles, 'In Defence of Arius', ibid, xiii (1962), pp. 339-47.
1
Alexander of Alexandria, in Theodoret, Hist. Eccl. i. 4. 35 = H. G. Opitz,
Urkunden zur Gtschichte des Ariamschen Strata (Athanasius' Werke, Bd. iii,
Teil 1), Urkunde 14, p. 25.
3
'Synesius of Cyrene and Alexandrian Neo-Platoniam', in The Conflict
bettaeen Paganism and Christianity in the Fourth Century, ed. A. Momigliano
(1963). PP- 126-50.
[Journal ctf Theological Studies, N.S., Vol. XV, Ft. 1, April 1964]
 THE PLATONISM OF ARIUS 17
and then as now philosophers had to choose between opposing stand-
points.
On the other hand, we must not pose this alternative: was Arius
influenced mainly by Plato or by Aristotle ?J In this period the influence
of Plato was everywhere predominant. Small groups of empiricists and
sceptics resisted it; but among philosophers whom Christians could
tolerate, the choice lay between Platonists who accepted, and Platonists
who denounced, the contribution of Aristotle or of the Stoics; between
the tradition of Albinus and that of Atticus. As I see it, the purely intel-
lectual opposition to Aristotle was now declining; Plotinus was its

Downloaded from http://jts.oxfordjournals.org/ at Nipissing University on October 14, 2014

last great representative, Porphyry established a concordat, the re-
serve of Iamblichus was temperamental and emotional rather than
logical. Nevertheless, there were still anti-Aristotelian Platonists, es-
pecially among Christians, who found Aristotle's logic too dialectical
and his theology too attenuated. But anti-Platonist Aristotelians no
longer existed.
Let us see, then, if we can gain any impression of the philosophical
tradition presupposed by Arius, considering as far as possible his best-
attested pronouncements. The letter of Arius to Alexander1 begins as
follows: OtSa/Liev h>a 6e6v, \u6vov dydwrp-ov, fxovov atSiov, fi6vov avapxov,
fwvov OAT)6IVOV, fiovov adavacrlav l^on-a, yu6vov ao<j>6v, fiovov ayafiov, [i6vov
8uvdon)v. This accumulation of phrases has no precedent, as far as I
know, and one can hardly question its polemical intention of contrasting
the unique position occupied by the 'one God' with the subordinate place
assigned to his vlos novoyevrfs. But in themselves the phrases can hardly
be objected to. The last five are scriptural,3 and the first three are well
established in tradition. Thus avapxos as a theological term goes back at
least to Clement and is used by Alexander. Moreover, the first phrase,
ixovov aydwip-ov, is clearly part of the common ground occupied by
Arius and Alexander. Alexander's own creed begins: TTioreuofiev . . . els
[iovov ayiWqrov naripa, and he defends other phrases by saying that they
do not compromise the principle that there can be only one ayhrv-qrov.*
Alexander does not say, as Athanasius was to say later,5 that Arius makes
1
On the supposed Aristotelian origin of Arianism, see H. A. Wolfson,
'Philosophical Implications of Arianism and Apollinarianism', in Dumbarton
Oaks Papers No. 12 (1958), pp. 5—9 = Religious Philosophy (Cambridge, Mass.,
1961), pp. 126-32. Cf. also J. de Ghellink in Revue d Histoire EccUsiastique,
xxvi (1930), p. 29, who makes a similar point about Eunomius.
1
Athanasius, de Synodis 16 = Opitz, Urkunde 6, § 2.
3
John xvii. 3; 1 Tim. vi. 16, i. 17; Mk. x. 18 || (ouStij ayaBos cl y.^ ets 6
8c6s); 1 Tim. vi. 15.
4
Theodoret, Hist. Eccl. i. 4 — Opitz, Urkunde 14, § 46 and §§ 47-52.
5
DeDecr. 28-31. C. Ar. i. 30-34. Cf., however, de Syn. 47. The spelling of
ayiv[v]TfTos cannot be certain. We may note in passing that the principle that
621/1 C
18 G. C. STEAD
too much of the non-scriptural term ayiwrfrov; indeed such a charge
would have been pointless, since Arius seems to have used Tla-rrip quite
freely; it occurs eleven times in this Arian letter.
We may conveniently pass to the more philosophical and technical-
sounding term fiovds, which occurs twice in this letter. There are three
possible implications at least, all of which (I think) can be paralleled in
Philo. God could be called 'the Monad', (a) as unique, or (b) as the
ultimate origin of things, or (c) as simple and indivisible. At its second
occurrence the sense 'ultimate origin' seems to be required. Having
protested against the notion of two ayhrv^roi dpxai the letter continues:

Downloaded from http://jts.oxfordjournals.org/ at Nipissing University on October 14, 2014

dAA' ws /xovas KCU a.px^l Trdvrwv, OVTWS 6 deos Trpo ndvraiv £o~ri. Could this,
indeed, be rendered, 'But as the monad is the source of all things, so
God is before all things' ? I think this unlikely; first because fiovds has
already occurred in a clearly theological sense, which I shall presently
mention; secondly because it weakens the argument; an excursion into
Platonic cosmology would merely illustrate Arius' contention without
supporting it. I take it, then, that the traditional version is right: 'God
is before all things, as Monad and Beginning of all'. The appeal is from
Movds and 'Aprf, as accepted theological titles used by Philo, Clement
and Origen,1 to the less-familiar notion that God is npo Trdvrcjv.
I have already said that at its first occurrence fu>vds has a theological
sense, meaning that it refers to the Godhead. It brings forward the
notion of indivisibility in an arresting fashion, for Sabellius is criticized
as TTJV /xovdSa Siaipuiv, in that he asserts an vtoTrdropa. This phrase has led
theologians2 to reason as follows: 'Sabellius suppresses distinctions
which he ought to recognize. Since Arius accuses even him of dividing
the monad, Arius himself must conceive the divine simplicity to be a
bare unity excluding all distinctions', as in the notorious passage in
Clement which suggests the comparison of God to a point without
location.3 The mistake here lies in the first sentence; Arius does not, of
course, conceive of the Trinity as a unity within which distinctions
might be recognized; his Trinity arises as the creation of distinct and
subordinate beings by the original monad. The Arian God can delegate
his functions; but any form of distribution would constitute a division.
From this point of view the criticism of Sabellius is logical; though I am
inclined to think that Arius attacks him merely as the most vulnerable
there cannot be two aytyrp-a. is one which Methodius assumes that even a
Valentinian will admit.
1
Mprf: Philo pluries, e.g. de Decal. 52; Clement, Strom, v. 89. 7; Origen,
in Jo. i. 17. 102, &c. Movis: Philo, Leg. All. ii. 3, cf. Clement, Paed. i. 71. 1;
Origen, de Princ. i. 1.6.
2
So, perhaps, J. A. Dorner, Doctrine of the Person of Christ (E.T.), 1862,
3
Div. I, vol. ii, pp. 234-5. Strom, v. 71. 2.
 THE PLATONISM OF ARIUS 19
exponent of economic-trinitarian doctrine, while hoping to discredit its
more orthodox champions; just as he condemns the widely used meta-
phor of a torch kindled from another by associating it with fhe obscure
heretic Hieracas.
As a counterpart to px>vds, Anus seems to have used the term 8vds
for the second Person of the Trinity; in the Thalia1 he is quoted as
saying, Evves, on ij pxivas fy • ij 8vas 8i OVK fy, vplv irnap^j). Avds, of
course, does not mean 'the Two', i.e. Father and Son, as Tpids means
'the Trinity'. Avds means 'the number two', implying both 'the Second'
and 'the Twofold'. It is almost certainly uncomplimentary; in Platonic

Downloaded from http://jts.oxfordjournals.org/ at Nipissing University on October 14, 2014

circles duality implies imperfection, matter, the world of the senses, the
left hand, the female principle. But there are also more respectable
implications. In Philo the term is applied to the spoken word, the Xoyos
Kara npo<f>opdv, as partaking both of the world of thought and the world of
sense, in contrast to the pure monad, the unspoken thought (de gig.
52, cf. quod Deus 82-84). ^n ^e Chaldaic Oracles this conception is
theologized (Kroll, p. 14):
Avas napa rwSe Kafhqrai'
afj/j)6repov yap exe'> VV M*v xaTt^eiv Tt* vor/ra,
aiod-qoiv 8' i-rrdyeiv Koafxois.
And the conception, though not the word, can be paralleled in
Numeniu8, fr. 25 (Eusebius, P.E. xi. 22, p. 544 b): 6 yap Sevrepos, SITTOS
wv, avTOiroict Tqv re ISeav iavrov Kal TOV KOO/XOV, 8rjfi.iovpy6s wv. It may be
that Arius has simply used 8vds as a synonym for the more familiar
Sevrepos (6eos), metrigratia.
This discussion leads me to think that theologians have been too
ready to assume, from Arius' use of the term fiov 's, that he has trans-
formed 'the Living God of the Bible' into 'the Absolute of the philo-
sophical schools' (Pollard, loc. cit., p. 104). Like Philo, Arius is prepared
to argue that God is fiovds while giving full recognition to his biblical
attributes. These are sometimes elbowed out under pressure of con-
troversy ; but in the Letter under discussion God is described in terms
which actually contend for the full biblical perspective in opposition to
Marcion; God is ndvrajv Kpinjv, Stouafrqv, olxovop^ov, . . . SIKOIOV Kal
dyadov, voyxiv Kal Trpo(fn]Ta>v Kal KOIVTJS SiafrqKTjS TOVTOV deov.
The point may be illustrated by a difficult phrase where the Logos
is described as Kal TO l,r\v KOI TO elvai napa TOV FlaTpos €iX-q<j>6Ta /cal
ray Sofas, o-uwnoar-qaavros1 aurw TOV TlaTpos'. 'the Logos has received
1
Athanasius, de Syn. 15.
1
For the transitive use of this word, cf. Proclus, in Tim. (Diehl) iii. 39, of the
demiurge bringing the year into existence simultaneously with the sun; the
simple form uiroanjaajra has already occurred in the letter
20 G. C. STEAD
from the Father his life and being and his dignities, which the Father has
brought into being along with him'. These 86£ai which the Logos has
possessed from the beginning are presumably his titles such as are
given in the Thalia:1 'Emvoetrai yovv nvpiais oaais fnivolais Tlvevfia,
Svvafus, ao(f>ia, 86£a 8eov, aArjBeid re Kal CIKWV, Kal A6yos, OVTOS. But
although the Father has conferred upon him these dignities, he has not
surrendered anything that belongs to himself: ov yap 6 vcn-rip Sous avrui
•naVTWv TTJV KXrjpovofiiav eoreprjoev eavrov <Lv ayewqrws e%ei *** ' a " r V '
The resulting picture is that which is unsympathetically described by
Alexander :z oxne dXrjdivos Kal <f>vaei TOU Trarpos Xoyos iarlv oxne aX-qdivf)

Downloaded from http://jts.oxfordjournals.org/ at Nipissing University on October 14, 2014

ao(f>ia avrov ioriv . . . KaraxprjOTiKais 8e Xeyerai \6yos Kal acxf>la, yev6fxevos
Kal avros TU> ISlw Beov \6ya> Kal rfj ev TW 8eS> acxf>la iv fj Kal TOL iravra Kal
avrov nenoirjKev 6 deos. The Son's attributes are, so to speak, parallel to the
Father's; but his wisdom is a generated wisdom and is called wisdom
KaTaxpTjoTiKuJs, since the true and ingenerate wisdom is an inalienable
attribute which belongs to the Father alone.
Just as the Father's attributes are not diminished or alienated at the
generation of the Son, so neither are they transformed into the Son; as
the Letter puts it, oi58^ TOV ovra -nporepov, vorepov yewrfiivra rj eniKTi-
odevra els Ylov; even Alexander admits this, says Arius, presumably re-
ferring to the theory of eternal generation; though I am not sure whether
Arius is attacking the theory of TertuUian and others that the Logos be-
came Son at and for creation, or the theory of Marcellus, that the Logos
became Son at the Incarnation. Arius clearly does not agree with
Alexander's conception ;3 but neither doe9 he hold what Professor Wolf-
son has called a 'two-stage' theory of the Logos.4 There are in a sense
two stages, 'before' and 'after' the generation of the Son, so far as these
terms are applicable before time began; but not two stages in the history
of the same being. Arius holds rather what we might call a 'two-level'
theory, by which wisdom, power, &c, remain as eternal attributes of the
Father; but the Son—as Arius habitually calls him—being a perfect
creature, learns wisdom (God being rfjs aortas StSacr/caAos avros, Thalia,
loc. cit.) and has all things committed to him (Ta travra aura) TraptSofh),
Letter, loc. cit.) and is called aMjdeid re KO! CIKWV KCU Xoyos (Thalia, loc.
cit.). In this system terms such as 'wisdom' and 'logos' have two mean-
ings, whereas 'Son* is presumably distinctive; the generated Logos is the
Son, the ingenerate Logos is the inalienable possession of the Father.
1
1
Athanasius, de Syn. 15.
Depositio Arii, Opitz, Urkunde 4b.
J
Cf. his objection to ad vtts, Letter to Eusebius of Nicomedia, in Theodoret,
H.E. i. 5. 1 ff. = Opitz, Urkunde 1.
* The Philosophy of the Church Fathers (1956), pp. 258 ff.
 THE PLATONISM OF ARIUS 21
Such a system obviously lends itself to misunderstanding or misrepre-
sentation, as Arius' critics quickly demonstrated. 'Utterly illogical'1 it is
not.
But if Arius differs radically from Alexander and from Tertullian and
from Marcellus, he differs no less radically from Paul of Samosata, who
contrasts the Wisdom of God with the human saviour Jesus, where-
as for Arius even the lower, generated Wisdom has existed from before
creation. It seems unlikely that he should have admired or sympathized
with Paul; moreover, an origin for Arius' theory can be found within
Alexandrian tradition,1 and in a fairly familiar context, namely those

Downloaded from http://jts.oxfordjournals.org/ at Nipissing University on October 14, 2014

passages in which Origen relates the Logos to the Father by means of
the Platonic conceptions of idea and participant. Philo, as we know, had
interpreted the ideas as thoughts of God, which being God's thoughts
are eternal realities and prototypes of the physical world; and had held
that the Logos is the place of those thoughts, or alternatively the sum-
total of them.2 In the main Origen keeps to Philo's picture; thus it is the
Logos whom he characterizes as rj avroaX^Oeia 17 ovauoSrjs (in Jo. vi. 6.
38), auroSiKCUocwvT] . . . Kal avroayiaofAos Kal avroaTToXvTpwais (ibid. i. 9.
59), 6 ainoXoyos xal 17 avroacxf>la KCU TJ avroaXrjdeia (c. Cels. iii. 41); again,
the Logos is God not by participation, fieTovaia, but in essence, ovolq.
(Sel. in Ps. 135, Lomm. xiii. 134). On the other hand, notoriously, the
Son is 9e6s but not avrode6s; and apparently not avroayadov, but CIKOJV
rrjs ayaOoTrjTos TOV dzov. Now admittedly there is good Platonic pre-
cedent for placing the idea of the good on a different level from the other
ideas, conferring upon all things that are known not only knowledge
but their very being (Plato, Rep. 509 b); but the divine attributes are
naturally associated (as justice with goodness), and once the Father is des-
cribed as avroayaOov in contrast with the Son, it is easy to make him also
avroacxf>la, avroXoyos. This pattern certainly appears in Eusebius, whose
usage of these forms is quite inconsistent, but who in controversy with
Marcellus can permit himself to distinguish between the ultimate
ev TI detov apprjTOV who is ainodeov, ainovovv, avroXoyov, avroacxfriav,
avrcxfnos, avro^urqv, auroKaXov, avroayadov and, on the Other hand, TOV
1
I differ on both these points from Dr. Pollard (art. cit.), who argues that
Arianism cannot be explained as a wholly Alexandrian product and must,
therefore, derive from Antiochene tradition. Perhaps even to speak of 'An-
tiochene tradition' in the third century is to underestimate the theological
turmoil which Antioch must have suffered when its patriarch was excom-
municated by representatives of a tradition which was itself beginning to be
suspected. I believe the comparison of Arius with Paul really rests on Origenistic
accounts and interpretations of Paul which could be adapted for use against
Arius.
2
Op. mind., §§ 20, 04; cf. also § 12 (eternity is characteristic of T<J VOT]T6V) and
§ 25 (the rorjrds Ktiofios is the divine Logos).
22 G. C. STEAD
Si TOVTOV /jMvoyevfj viov . . . Kal ainov 9eov Kal VOW KOX \6yov Kal acxfaiav
Kal Z,OJT)V Kal <f><os . . . ainov re TOV KOXOV Kal dyaOov etKoVa (Eccl. Theol.
ii. 14. 6-7). This, as I see it, is also the position reached by Arius,1 for
whom also the Father is the perfect prototype of goodness, wisdom, and
all excellence, while the Son is merely a perfect instance of these, not
even necessarily unique (according to the Thalia),z created by the Father
to be his agent in making the world.
I have traced this development back to Origen; but it is possible that
it may be older. Origen's distinction between dyaSov and ainoayadov
is probably taken from Numenius, who is quoted by Eusebius as saying

Downloaded from http://jts.oxfordjournals.org/ at Nipissing University on October 14, 2014

6 Srjfuovpyos 6 rfjS yevioews « m v dyaOos' T) TTOV earat Kal 6 TTJS ovolas
8-qfj.iovpyos avroaya66v (P.E. xi. 22); the first principle, the avroayadov
is the source of unchanging reality, ovaia, the second is merely the
source of yeveais, of change and time. Again St. Clement may have
had a hand in the process; at any rate Photius thought that in his
Hypotyposes he had suggested the notion of two Aoyoi; see cod. 109, the
passage cited as Clement, fr. 23: Xeyerai fiev KO! 6 vlos \6yos, o/xaW/xtor
TO> na.Tpt.Ku> Xoyco, dXX' oi>x ofrros io-Tiv 6 a d p f yev6fi€vos.
Now theologically, of course, there is a crucial difference between
Alexander's insistence that the Son is in all respects save one exactly
like the Father, and Arius' insistence that they differ in one all-important
respect. But behind this theological difference there is a certain similarity:
the Father has all the fullness of being and glory; the Son derives his
being from the Father; the point of difference is the ayewrjTov. Moreover,
the thought of Arius and Alexander, and indeed Athanasius, is strongly
dualistic, controlled by the antithesis between the one ultimate source
and every form of derived reality. In the light of this antithesis Alex-
ander in particular finds it difficult to define a place for the Logos which
is both theologically adequate and consistent with his philosophical
scheme; this appears from such unsatisfactory phrases as ixeoiTtvovoa
<j>vais novoyevTjs—for how can there be a nature intermediate between the
originate and the unoriginate ?—and again the paradoxical dyewrp-oyevris.
Athanasius, as we know, was to solve the problem by the masterly
distinction between yewdaOai and yeviodai, incorporated in the yemnj-
Bevra oil Trovrfitvra of Nicaea, which reflects Origen's doctrine of eternal
generation, but also goes back to a piece of Platonic scholarship, which
distinguished between God's attributes as IJarqp in relation to rational
1
This cannot be a certain conclusion, since the characteristic compounds of
auro- are not attested in the earliest Arian documents. Mniffiw, peroxf\ are
hardly technical enough to be significant; however, they are attributed to Arius
by Athanasius in what purports to be a summary of the Thalia in connexion
with the three terms EotfUa, A6yos, and 0(6s: c. Ar. i. 5.
1
Athanasius, de Syn. 15.
 THE PLATONISM OF ARIUS 23
1
beings but IIonjTTJs of the rest {Timaeus 28 c). Arius keeps the con-
sistency of his philosophical scheme by treating the Son and Holy Spirit
simply as the foremost among dependent beings. But although he differs
so sharply from Athanasius on the status of the Son and Spirit, one can
still see traces of the same basic dualism in Athanasius' radical distinction
between God and creatures. Athanasius has little use for the doctrine,
still important for Origen, that the vovs or Xoyos in man is naturally akin
to the divine.1 He does indeed follow the Platonic ethic which depreciates
the senses and holds that the Xoyos gives knowledge of God and of man's
ultimate good. He calls God dawftaros. But the soul in man is created

Downloaded from http://jts.oxfordjournals.org/ at Nipissing University on October 14, 2014

and perishable; it is a Hellenic error to speak ofvouj or <pvxv as dyevtyro.;*
and adjectives like pevaros are generally attached, not to the body in con-
trast with the soul, as in Clement, but to the created nature as such.
I used to think that this dualism in Athanasius was a result of the
Arian controversy. But if it M shared by both parties, it is more likely
to reflect a common background of thought which differs perceptibly
from the Stoicized Platonism of Clement and Origen, no less than from
classical Aristotelianism and from Plotinus.
Arius' divergence from Alexander and Athanasius may be indicated
as follows; the latter are prepared to do violence to their philosophic a
dualism in order to establish a position for the Son which is theologically
and devotionally adequate. Arius does his best (at least initially) to
establish such a position while keeping his basic dualism intact. Thus
1
Opinions differed on whether Tra-rqp and wowpTj? referred to the same being.
Some, like Atticus, identified them; see Proclus, in Tim. (Diehl), i. 305,
Eusebius, P.E. xv. 13. 5 (p. 815 d), and cf. C. Andresen in Z.N.T.W. xliv (1952),
pp. 167-8, on Sfi/uovpyds as a recognized variant for ironjTrfs, citing Justin, Apol.
ii. 10. 6. Others distinguished; e.g. Numenius (fr. 36 = Proclus, op. cit. i. 303)
and Porphyry (ibid. 300: Trarf/p \iiv iariv 6 &4>' iavroC ytwwv TO OAOI>, iroiijn^j Zk 6
•nap' dAAou rfjv vXr/v Xafifidvajv, with which cf. Methodius, de Creatis ix, cited
below). Christians could use the first opinion as a counter to Marcion's doctrine
of an imperfect creator, but the second to define a function for the Logos.
Plutarch suggests, inter alia, that the same God is wanfc of animate beings but
merely Troirjrfjs of the rest; see Plat. Quaest. ii, cited below. Origen rejects a
cruder variant propounded by Celsus (God creates only souls, not bodies, c.
Cell. iv. 52 ff.), but adopts a not dissimilar position: God stands in an eternal
relation to all XoyiKoi, who indeed are umus substantiae with the divine nature
(de Princ. iv. 4, 9 and parallels cited by Koetschau, p. 362). The rejection of
Origen's doctrine of pre-existent souls leaves 'eternal generation' as a prerogative
of the Logos in contrast to rd Tronfiivra.
1
A notable exception is c. Gentes 30—33. The de Incarnatione gives a very
different impression; even at its first creation human nature is weak and un-
stable (3. 3-4), needs divine pity (ibid.), is subject not only to corruption but to
total destruction, the 'second death' (3. 5). As it is, the Xoyixot are corrupted
(6. 7, 8. 2), the soul defiled (11. 4) and deceived (14. 4), Sec.
3
De Deer. 28, cf. Vita Antonii 74.
24 G. C. STEAD
in principle, and in contrast with the one ingenerate Father, the
Son has to be ranked with the creatures; and their status is defined
by three characteristics: they are e'£ OVK OVTWV; they are created by a
sovereign act of divine will; and created at a definite moment, though
not in time.
Arius' assimilation of the Son to the creatures provides a clue to some
of his polemical statements. Thus his contention that the Son is ourc
fiepos 6eov oxne ef imoxeifidvov TIVOV has no properly christological set-
ting that I can discover; but it can be readily connected with the Stoic
opinion that rational man as such is fjJpos 6eov, and with the opinion

Downloaded from http://jts.oxfordjournals.org/ at Nipissing University on October 14, 2014

of many Platonists that God made the world c£ xmoKeifjAvov TWOS. The
first can be found in Epictetus and Marcus Aurelius (ii. 8. 10; iv. 4
respectively) and is sharply disowned by St. Clement: el fir/ ns fi^pos
airrov KOLI 6/xoovolovs 17/xas TOI 6eu> roXfirjaei. \e~yeiv {Strom, ii. 74. 1),
where two of Arius' aversions appear in the same clause. Furthermore
Arius' dislike of ^tipos Oeov can be connected with his objection to «£
avrov TOV Oeov,2 since Plutarch associates these two notions in a dis-
cussion of the soul: 17 8e ^fy; . . . OVK Hpyov tori TOV 9eov fiAvov aXka.
Kal fjjpos, oi)S' inr' avrov aXXa. /cat dv' avrov KCU e£ avrov yeyovev
{Plat. Quaest. ii. 2). In a contrary sense Hermogenes argued that
partition of the Godhead would result if God made anything 'from him-
self'; Tertullian writes 'Negat ilium (sc. deum) ex semetipso facere
potuisse, quia partes ipsius fuissent quaecumque ex semetipso fecisset
dominus; porro in partes non devenire ut indivisibilem et indemuta-
bilem et eundem semper, qua dominus' {adv. Herm. ii. 2). In these
oppositions we can trace a debate between opposing schools of Platonists,
in which the great Christian Alexandrians, like Philo before them,
tended to stress the unity and indivisibility of God, in opposition not
only to Valentinian theories of prolation and Stoic conceptions of a
material deity, but to Stoicizing Platonists like Plutarch; though some,
like Origen and Alexander, made concessions which Arius considered
unwise.
Plutarch provides a particularly good example of the freedom with
which some Platonists could speak of the deity as physically disseminated
in the world, in a comment on the Timaeus 28 c, a passage already re-
ferred to. Are we to think, he says, that God is 'Father' of divine off-
spring3 and of men, but 'Maker' of irrational creatures? Or is 'Father' a
1
Letter to Eusebius of Nicomedia, Opitz, Urkunde 1, §5 (p. 3, 1. 6).
1
Ibid., §2. P. Nautin, in Analecta Bollandiana 67 (1949), pp. 131—41, has
argued that this phrase is an orthodox interpolation, designed, of course, to
clarify Arius' position. However, Candidus' Latin version includes the phrase;
in any case Arian criticism of it is independently attested by George of Laodicea,
J
Urkunde 13 = Athanasius, de Syn. 17. TCUK Btwv T
 THE PLATONISM OF ARIUS 25
1
mere metaphor, as poets are called 'fathers' of their writings? Or,
thirdly, is God called 'Father' because the world itself is his living off-
spring?
*H 8ia<ftep(i TTaT-qp re TTOITJTOV Kal yewrjaews TToir}Ois; (Ls yap TO
xal TreTrolrjTai, ov fxr)v avanaXiv, OVTWS 6 yevyrjaas Kal
fj^ni^ov yap Ttoirjais rj y^wrjois 4<rrim Kal TTOITJTOV ftev, 010s
OIKO86[IOS r) ixfxivT-qs rj Xvpas Srj/xiovpyos rj avSpidvros, aTrrjXXaKTai yevo-
fievov rovp-yov3 r) 8' OTTO TOV yewqaavros apx?] Kal Suvafiis iyKeKparai TS>
TeKVwOevTi Kal awe^ei T?JV <f>vaiv, a-rrooTTaofia Kal fiopiov* ovaav TOV

Downloaded from http://jts.oxfordjournals.org/ at Nipissing University on October 14, 2014

T€Kvd)oavros. eTrel TOIWV ov TreTrXaapAvois 6 Koofj&s ov8e awqpixoujxivois
TTOirffxaaiv eoiKev, aAA' tWcmv aurai fioipa noWrj ^WOTTJTOS KOI deion/ros,
rjv 6 Oeos eyKaTJoiTiipev axf>' iavrov Tjj vXjj Kal KaT^/xi^ev, eiKOTCos
dfza Trarfip re TOV Kocrpxiv, £,a>ov yeyovoros, Kal TTonjrrjS errovofid^eTai
(Plat. Quaest. ii).
The contention that the Son is not cf imoKtifievov nvos has to be
related to contemporary Platonist opinions about the cosmos. There were
three views, all linked with traditional interpretations of the Timaeus:
(i) The cosmos has always existed, the story of its creation being a
parable of its eternal dependence on the Godhead. This view emerges
as early as Aristotle and Xenocrates, and was probably the majority
opinion among Platonists in our period; it is noted as such by Calcidius5
and was held by Plotinus and his successors.
(ii) The cosmos was created by a definite act out of formless matter,
which is ov8e7TOT€ 6v in the sense of having no character of its own, which
comes to be all things but is not itself anything (Timaeus 27 d). This
view was held by Plutarch and Atticus and 'many other Platonists',
according to Proclus;6 among Christians, Justin and Athenagoras appear
not to dissent from it.7
(iii) The cosmos was created strictly ex nihilo. This opinion is not
very easy to trace outside Christian circles. However, Lactantius quotes
an otherwise unknown passage from Cicero's de Natura Deorum which
1
Cf. Dionysius of Alexandria, ed. Feltoe, p. 195, 11. 4-6 = Athanasius, de
Sent. §21.
1
Cf. Tertullian, adv. Marc. v. 4, quit parent nisi qui et factor?, and adv.
Hermog. 32. 2, quia et filios facimus licet generemus; Athanasius, c. Ar. ii. 4.
1
The same contrast is echoed in a christological context by Origen, in Jer.
ix. 4 (p. 70, 1. 15); Athanasius, de Deer. 20. 4.
4
Cf. Epictetus, i. 14. 6: dAA* ax <pvx°u- H^" ovra>s tlolv ivSfhtijUvai Kal owaijifis
Tip $cq> are avroC fi6pia ofiaai Kal dTrooTrdafuiTa. j^TT6a-naa\u3. is regularly used of
human seed by the Stoics, see S.V.F. index; but is also applied by Philo to the
5 6
soul. In Tim. 23-25 and 300. In Tim. (Diehl) i. 276. 31 ff.
7
It was no doubt recommended by Wisdom xi. 17, x«P (dtoC) KrCaaaa rdv
K6OHOV ii afwpjnv VXTJS; cf. Justin, Apol. i. 10. 2; 59. 2; 67. 8; but also Dial. 5;
Athenagoras, Suppl. 19.
26 G. C. STEAD
is directed against it; Basilides probably adopted it;2 and among
1

Christians it was unambiguously upheld by Theophilus who, like

Tertullian, wrote against Hermogenes.3
It should I think be noted that the mere phrase «| OVK OVTWV is not
entirely adequate to define the doctrine of creation ex nihilo as we
understand it, and the early uses of the phrase do not necessarily imply
it.4 For TO. OVK oWa can mean what is indeterminate, or bad, or indeed
anything which is distinguishable from 6 a>v, the One ultimate reality.5
Again it is difficult to avoid the suggestion that some process of change
is involved in creation, as if not-being were some form of attenuated

Downloaded from http://jts.oxfordjournals.org/ at Nipissing University on October 14, 2014

being which can be transformed into real being. Thirdly, there is the
argument that all change is in a sense eg OVK OVTWV, since any x mu9t origi-
nate from not-*; this argument is actually used by Atticus, as cited by
Eusebius (P. E. xv. 6. 803 a, b). Nevertheless, there is no doubt that
both Anus and Athanasius adopt the third position, and adopt it in
conscious opposition to the second. Athanasius never charges Platonists
with teaching that the world is without beginning; their mistake is to
think that God needed matter to make it. Anus likewise, with his
reference to the vvoKclfievov, seems to oppose the second form of
cosmology, though there is no doubt that his emphasis on the priority
of the Father entirely excludes the first. It is at this point that Arius'
divergence from Origen is most strongly marked. Origen had placed the
Father in an eternal relationship, not only with the Son, but even, in
principle, with the world. Arius asserts the Father's priority not only to
the world, but to the Son.
Arius thus criticizes Alexander's Origenistic doctrine of the Son's
eternal generation; he is, however, prepared to concede that the Son's
generation is not in time; he is dxpovws yewrjdfls, dxpovws npo TTOVTWV
&c. On the other hand he clearly conceives of the generation as in some
sense a momentary event, as is shown by his frequent use of the aorist
vTTTJpgev. It might seem disingenuous to speak of 'before' and 'after'
while professing to exclude the notion of time. But Arius1 position has
an obvious basis in Plato's Timaeus; see 38 b, xp^vos 8' oSv /xer' ovpavov
yiyovevf a n d 39 e, KOL TO. fikv dXXa 17877 ft^XP1 XP°V0V y*v*oe<*>s dneipydoaTo.
For Plato, of course, time depends upon the heavenly bodies, a view
which can conveniently be harmonized with Gen. i. 14, but leaves the
1
Divine Institutes, ii. 8. 10. * Hippolytus, Ref. vii. 21. 4.
1
Sec J. H. Waszink, Tertullian, The Treatise against Hermogenes (Ancient
4
Christian Writers, no. 24), pp. 9-12. E.g. 2 Mace. vii. 28.
' So even of the Son, in the Arian statement reported by Alexander. (Opitz,
Urkunde 4b, §7): 6 yap uiv 6c6s rdv fiij Svra ix TOS /t^ OKTOS
6
Cited by Methodius, de Res. i. 8. 8.
 THE PLATONISM OF ARIUS 27
obvious difficulty that it is possible to conceive of events 'before' these
bodies existed.1 Alexander is struggling with this difficulty when he
says that temporal-sounding phrases like rrpo auovwv do not do justice
to the Son's divinity and 'quasi-antiquity' {otov dpxauirrp-a), which
nevertheless do not make him dyewr/rov; we have therefore to say that
the Son is not merely co-temporal with the Father, but co-eternal; if we
admitted Jjv ITOTC ore OVK fy, the OVK TJV would refer r) xpovois ... r) alwvos
TIVI StaonjfiaTi,either to time or to some period beyond time, both of
which suggestions must be excluded.2 It follows that, on contemporary
assumptions, Athanasius is justified in observing (c. Ar. i. 14) that the

Downloaded from http://jts.oxfordjournals.org/ at Nipissing University on October 14, 2014

Arians postulate a state of being 'when' the Father 'was' without his
(distinct and generated) Word; but to say that despite their disclaimers
they nevertheless imply periods of time (ovSev JJTTOV ypovovs crrj/juuvovres)
is no more than a rhetorical device.
Arius maintained that the Son was generated by an act of divine will.
This again has a basis in the Timaeus (41 a, b) where the supreme power
declares that his works are dAin-a ifiov ye fiJ) iOtXovros; therefore the
subordinate gods, who have been brought into being (yeyeirqode), will be
preserved by the divine will, rrjs efj.ijs ^ovXr/aews jxeit,ovos en Beafxov . . .
Xa\6vres eKelvwv of? or' eyiyveade crweSeiode. This passage became a
proof-text in the debate concerning the eternity of the cosmos,3 so that a
divine will is acknowledged even by Platonists whom one might expect
to ignore it, for instance Albinus {Epitome 10), whose theology is
largely Aristotelian. Ammonius denied the whole conception, however,
and Plotinus admits it only with some hesitation against the general run
of his system, as a counter to gnostics who could quote Stoic teaching
about divine necessity.
Creation by a sovereign act of divine will was taught by Philo, and
the Christian apologists also state that the Logos was generated by the
divine will for the work of creation.4 Origen admits the notion of will in
the generation of the Logos rather more guardedly, as a legitimate ana-
logy,5 and for much the same polemical reasons as Plotinus; it serves to
1
An interesting variant is that of Atticus, cited by Proclus, op. cit. iii. 37:
Xpovos lifv fy KaX vp& ovpayov yev^occos, rerayfi^vos Si %p6vos OVK TJV.
1
Cf. Origen, de Princ. iv. 4. 1; Calcidius, in Tim. §23 (Wrobel p. 89).
3
Thus it underlies Origen, de Princ. ii. 3. 6, Sane hoc . . . voluntas dei.
4
See WoUson, op. cit., p. 224. Creation of the world by the mere act of God's
will is taught by Clement, Protr. iv. 63. 3: p6vov avrov r6 povXrjfia Koo\urnoua.-
fi6vos yap o Beds €7TO(T]OCV i-nti KaX ji6vos ovrots Jerri 6e6s' tfsiXip T $fiovXeodax&rjfu-
ovpyfi Kal Tip pdvov idcXrjoai avrdv ^rrerai TO ycyevrjaBai. Cf. Methodius cited
below, p. 28.
5
De Princ. i. 2. 6, iv. 4. 1. For the concept of divine will as distinct from
passion, cf. Clement, Strom, ii. 72—73, esp. 72. 2, rd ftovXtjua TOV anaBovs Beov.
28 G. C. STEAD
exclude Valentinian notions of physical derivation and partition. It
seems, however, that Origen's somewhat reluctant admission did not
satisfy his critics, who thought he had reduced the traditional doctrine
of a creative act. Thus we find the doctrine of a divine act of will being
reasserted with the greatest possible emphasis in opposition to Origen
by Methodius, who indeed partly foreshadows the subordinationism of
Arius as well; see his de Creatis, ix:
Avo Se Swdfieis • . • €<f>afX€V elvoj. ironyriKas, TTJV i£ OVK OVTCOV yvfi.va> TOI
fiovX-qfuiTi xwpls fieXXrjanov dfui T<X> deXrjaai avrovpyovoav o fiovXerai
noieiv Tvyxdvei 8e 6 iraTrjp- darepav Se KaTOKoa/jLovoav xal TTOLKIXXOVOCLV

Downloaded from http://jts.oxfordjournals.org/ at Nipissing University on October 14, 2014

Kara (Ufirjaiv TTJS npoTepas r a TJSTJ yeyovora- ecm 8^ 6 vlos, T) iravraSvva-
/xos Kal Kparoua x^p TOV Trarpos, ev fj [itTa. TO iroiijaai TTJV VXTJV e£ OVK

In rejecting the eternal generation of the Son Arius explicitly disowns

one argument commonly used to support it, the argument from rela-
tions. 1 ov8e dfia T(p IJarpl TO etvai excl» ^ Tives Xiyovai TO. Trpos TI. This
can hardly mean that it is illegitimate to speak of relations at all; the
point must be that they are irrelevant to theology.
Arius is here taking sides in a long-standing debate with which, since
he uses the Aristotelian technical term T<X vpos TI, we presume that he
was reasonably familiar. The thesis is that correlative terms must
coexist, and goes back to Aristotle's Categories, vii; note especially §19,
7618, SeaTTOTOV ovros, SovXos effTi /cat SovXov ovros SecrTTOTTjs iarlv.
In the Metaphysics (v. 15) the case of father and son is mentioned; these
two cases become the stock examples, and are brought together by
Origen in de Princ. i. 2. 10. Philo appeals to the principle of correlatives
in a cosmological discussion (Op. mund. xx. 104); while the argument
that Father implies Son is used as a theological commonplace; see
Tertullian, adv. Prax. 10, Novatian, de Trin. 31, Origen, loc. cit., and
in Jo. x. 37, Dionysius of Alexandria, ed. Feltoe p. 187, 13 = Athan. de
Sent. 15.
But an antithesis or counter-argument can be constructed on the
basis of Plato's Theaetetus, 155 a-c. Theaetetus becomes larger than
Socrates, although Socrates himself remains unchanged; a relationship,
therefore, may change, or begin or cease, without both its terms doing
so. Aristotle also qualifies his principle that correlatives must coexist; it
holds good enl [ih> TUIV vXelarutv . . . OVK hrl •navrwv hk . . . SOKCL {Categg.
1
For previous discussions of this topic see P. Arnou, 'Arius et la Doctrine
des Relations Trinitaires', Gregoriaman, xiv (1933), pp. 269-72; J. de Ghellink,
'Qui sont les QZ TINES AETOYZI de la lettre d'Arius ?', Miscellanea G. Mercati
I=*Studi e Tetti, cxxi, pp. 127-44. De Ghellink thinks of actual debates between
the contending parties.
 THE PLATONISM OF ARIUS 29
loc. cit.); and in the Metaphysics passage he makes it clear that the
father-son relationship depends upon a definite event in time. A con-
verse point is made by Plotinus (vi. 1. 7), who argues that a son remains
son even though his father has deceased.
It is, of course, possible to modify the thesis by saying that GW's
relationships must be eternal as a consequence of his being; and this can
be put in a more technical form by referring to the common opinion
that relations are classed among the 'accidents' (otJ/^Se/fy KOTO).1 Since
God is unchangeable he can have no ovufSeflrjKOTa (a point made by St.
Athanasius); and his relationships must therefore be substantial,

Downloaded from http://jts.oxfordjournals.org/ at Nipissing University on October 14, 2014

necessary, and permanent. This is presumably the type of argument
used by Hermogenes; God must always be dominus as well as pater, and
must therefore always have had a universe as his dominion. Origen
himself is sufficiently impressed by this argument to conclude that there
must always have been a world, even if in obedience to tradition this
world be admitted to have had a beginning (de Princ. i. 2. 10, i. 4. 3-4,
and iii. 5. 3, the two latter deferring to a rather similar argument about
God's 'inactivity' before creation took place).
This conclusion, however, seemed to stultify the Church's tradition;
and a reply was soon available, of which Methodius was perhaps the
ablest exponent. The first and obvious point is that it proves too much;
as the Arians quoted by Athanasius contend, the argument condemns
itself by proving the eternity of creatures (c. AT. i. 29); indeed human
history is impossible if God cannot enter into different relationships
with men according to their deeds.2 The second reply is more subtle.
In principle it is that correlatives must be symmetrical; if x is necessary
to y, then y is necessary to x. If God's nature demands that something
should exist, this fact cannot be stated in terms of correlatives without
making God dependent on the existence of some inferior being.3
Methodius deploys this argument against Origen in his de Creatis; it is
absurd to hold that the world makes God into a benefactor and creator:
TL 8e; KOX TOV Beov, erepov ovra. Koafiov, Sid K6O\UOV TOMTQ TTXOVOIOV xa.1
evepyirqv KOI S-qfuovpyov KAIJT^OV;—Unaye, ovSafjuiJs (iii. 4). At an early
stage in the Arian controversy we find Eusebius repeating this argument
1
See, for example, Hippolytus, Ref. vi. 24. 2.
1
Origen'8 doctrine of inlvoiai is intended to meet this point, but is qualified
by the doctrine that it is only the Son who can properly be credited with a
plurality of i-rrlvotai, the Father being a pure unity; see, for example, in Jo. i. 20.
119.
3
Cf. the Valentinian argument quoted by Hippolytus, Ref. vi. 29. 6: 1} Si
aydm) OVK lonv ayd-jn], iav fi-f/ f T6 ayairwpevov. Methodius himself {de Autex. xxii)
attempts a restatement, which, however, does not entirely succeed; God still
needs recipients of his goodness; he can no doubt enjoy his works, but hardly
love them, in prospect.
30 G. C. STEAD
in more technical language with reference to the Father and the Son;
the analogy of a source and ray of light breaks down, he says, since the
ray coexists with the source and is necessary to it: o 57 ira-rfjp npoimapxet
TOV vlov Kal TTJS yeveaews airrov Trpoij<f>iaTr)K€v, $ fi6vos ayewr/TOS TJV. KO!
6 [lev Ka.8' iavrov reAeio? /cat irpioros coy TTaTTjp, Kal Trjs TOV vlov avoTaoetos
aiTios, ovSev els ovfnrXrjpajoiv TTJS tavrovOeoTr/ros na.pa.T0V vlovXafjifidvajv-
6 84 . . . hevrepos, . . . napa TOV naTpos Kal TO etvai Kal TO TOioaSe eivai
elXrrfxLs (Dem. Ev. iv. 3. 5-6). And dependence on Methodius is sug-
gested by the fact that this passage brings together three points, the
priority of the Father, his sovereign act of will, and the rejection of the

Downloaded from http://jts.oxfordjournals.org/ at Nipissing University on October 14, 2014

argument from relations, with which Methodius explains the Father's
relationship to the world, but which Eusebius and Arius, as we have
seen, apply to the Son.
The force of these observations will be variously estimated. But I
will state my opinion that, whatever may be true of his adherents, Arius
draws upon a Platonic tradition evolving within the Church, rather than
representing a violent incursion of alien philosophy. His theology fits
naturally into place among the disputes which arose over the disposal of
Origen'8 effects. His main debt to Origen is a subordinationist doctrine
of the Son, which he greatly intensifies and divests of its qualifications.
But he is indebted to Origen's critics, especially Methodius, for the
doctrine of the Father's priority, which he considers necessary for
monotheism: God the Father existed before all things and created all
things out of nothing by his unprompted act of will.
Arius no doubt conceived himself to be reasserting traditional
Christian positions which Origen and his followers had obscured: the
absolute primacy of the Father, and the importance of sacred history.
But as an educated theologian he no doubt drew upon philosophical
literature in defence of his Christian themes. There are two main points
at which pagan philosophy could help him; first, in his defence of the
Father's primacy, and secondly in his violent reaction against any con-
ceptions which would involve the Father in natural or physical processes,
of which the Valentinian and Manichean theologies were the extreme
examples. We should therefore suspect the influence of a form of
Platonism which gained its distinctive character by rejecting those
elements in the Platonic tradition which could prove useful to those
theologies; that is, by rejecting the Stoic and Aristotelian elements in its
physics and theology which are present in Plutarch, Numenius, and
Plotinus and in the Christian Platonism of Clement and Origen.1 Among
1
I do not suggest that the two latter are strictly comparable. Clement's
Platonism is highly eclectic, and incorporates both Stoic elements and reactions
against them of a kind which I suppose were attractive to Arius.
 THE PLATONISM OF ARIUS 31
Christian Platonists, Anus probably owes most to Methodius; among
non-Christians, despite the similarity of his trinitarian scheme with that
of Numenius, I am inclined to think that his closest affinities are with
Atticus. Like Arius, Atticus accumulates theological predicates upon a
single Creator and Father,1 reducing all other powers to subordinate
rank, rather than teaching a transcendent Absolute who can only be
known and described in terms of a second principle. Atticus again
stresses the divine will, and teaches (even if not in the later Christian
sense) that the world had a beginning in time and was made 4£ OVK OVTWV.
It is even possible that Arian exegesis, which I have so far ignored, has

Downloaded from http://jts.oxfordjournals.org/ at Nipissing University on October 14, 2014

some connexion with Atticus' rather literal exposition of his Platonic
texts. The evidence does not, I think, permit us to say positively that
Arius' theology was shaped by a dialogue with non-Christian Platonist
contemporaries; but it is compatible with the existence of a Platonist
tradition at Alexandria which was marked by a similar reaction against
Valentinian and Manichean theories; which is attested, perhaps, by
Alexander of Lycopolis, if he is not the Christian bishop of tradition;
and which escaped the attention of pagan historians because it was not
involved in the pagan revivalism of Porphyry and Iamblichus, but
remained acceptable to Christians, including heretical Christians such
as Arius.
G. C. STEAD
1
From Eusebius, P.E. xv. 6 and 13 (pp. 802 a-8o3b and 815 a) one can
collect the following list: KOXKLOTOS TCSV h-qmovpyajv, TOU -navrds iroujrrfs, 7ra/i-
paoiXevs, dpiOTOiVxvTjr, wo-njp, S-7]^uovpy6;, SfcrmJn/r, ia]5*fiuiv.