III

A survey of the provincial and private confessions which remain to us from the Western Church, belonging to the period from the fourth to the sixth (seventh) century,³³³³Hahn, §§ 20-45; Caspari, Bd. ii. and iii. The fullest appreciation, however, is in Kattenbusch, S. 59-215, and in the addenda, S. 392 ff. The number of symbols found is very great, and is still increasing. We know of six Italian (besides Rome, we have symbols from Milan, Turin, Ravenna, Aquileia, and possibly also Florence), African (but none from Sardinia, for the very important one of the date 340-360, which Caspari has discussed [ii. S. 128 f.], can scarcely be attributed to that country; see Kattenbusch, S. 202 f.). There are also Spanish, Gallican (South Gallican and Frankish, also one from Treves), and Irish. enables us to make six very important observations about them:—

1. In the choice and arrangement of 26the single parts they all exhibit the same fundamental type as the shorter Roman symbol.

2. The shorter a Western symbol is, the more closely it approaches the shorter Roman symbol. The shortest symbols of the provincial churches of the West are almost, if not altogether, identical with it.

3. The later a Western symbol is, the more it varies, as a rule in consequence of additions,³⁴³⁴Hardly ever by omissions; on the symbol of Venantius Fortunatus, printed in Hahn, § 27, see Kattenbusch, S. 130 ff. The question of alleged omissions in the Western symbols may be put aside in view of the uncertainty of the tradition. from the shorter Roman. With the exception of a few expressions, like the anti-modalistic “invisibili et impassibili,” 27added to the “omnipotente,” in the first article of the symbol of the church of Aquileia; the plerophoric “huius,” as an addition to “carnis,” in the third article of the same symbol; the position of “remissionem peccatorum, resurrectionem carnis et vitam aeternam per sanctam ecclesiam” in the Carthaginian Church (this arrangement, however, may be explained otherwise), none of these additions are of a directly polemical nature, but are to be regarded as completions and extensions held to be necessary in the interest of a clear understanding of the Creed. With these may be compared the manifold and various additions to the first article of the old symbol, for example:³⁵³⁵Hahn, § 42. the formula “natus de Spiritu Sancto ex Virgine Maria,” in the symbol of Aquileia 28and Ravenna; the formula “conceptus de Spiritu Sancto, natus ex Virgine M.” in the symbol of Faustus of Rietz; the differentiation of “crucifixus” into “passus . . . crucifixus” in the later symbols; the addition of “catholicam,” in the third article in the Spanish and Carthaginian symbols as well as in that of Nicetas; the addition of “vitam aeternam” for example, in Augustine’s symbol and in Faustus of Rietz; and so on. The fundamental character of the symbols is not altered by such additions, as they are not of a speculative or dogmatic nature.

4. The majority of the additions which the Western symbols exhibit are of such a character that they may be regarded as intermediate steps between the shorter and longer Roman symbol. This consideration, however, is not so important as 29the fact that the great provincial churches of the West in the third and fourth centuries, by the additions which they severally made, stamped the symbols with a definite character. Four such types can be readily distinguished, namely, the Italian, the African, the Gallican, which includes the Irish, and the Spanish.³⁶³⁶Kattenbusch, S. 189 ff., 194 ff. makes no distinction between the last two, and recognises only one type in Western Europe; but this view is not correct. As for the Gallican type which is seen in our Apostles’ Creed, one of its distinguishing features is that it is characterised by such historical additions as are to be found in the earlier Oriental Rules of Faith or symbols, as the case may be, such as “creator of heaven and earth,” “suffered,” “died,” “descended into hell”; and also the predicate “catholic.” The Gallican type in its 30final form is not in every respect the richest or the longest of the Western symbols; but in so far as its historical contents are concerned, it certainly is so. What gives it its peculiar character is the fact that. with the richest material contents it lacks all those finishing touches or elements of accurate definition which are present in other symbols of provincial churches, such as “invisibilem et impassibilem” in the first article; “omnium creaturarum visibilium et invisibilium conditorem” and “unum,” in the first and second; “Deum” in the second; “resurrexit vivus, omnium peccatorum, cum gloria venturus, per baptismum,” in the third; “huius carnis, etc.” In these important respects the final form of the Gallican type, that is, of our Apostles’ Creed, has completely preserved the distinguishing features of the old 31Roman symbol. It exhibits the same compact and severe style, and nevertheless also preserves all the significant historical features that became attached to the Symbolum Romanum in the course of its career. The Gallican Apostles’ Creed also exhibits the same classical elaboration as its Roman predecessor, and like it was regarded as possessing the same ecumenical authority.

5. The less any church was influenced by the church at Rome, the more significant become the progressive variations of its creed from the shorter Roman symbol. The symbols of the Gallican Church are relatively far removed from it.

6. If all the Western symbols be reduced to an archetype, and the differences be disregarded, we arrive without difficulty at the shorter Roman creed.

What conclusions are we to draw from these observations? The evidence justifies the assertions (1) that the shorter Roman symbol was the source of all the Western confessions of faith; and (2) that the longer Roman symbol was gradually developed from the other, and as a consequence also preserved the same attributes as originally characterised the shorter symbol. But the process did not take place in Rome.

From the first conclusion we may reasonably infer that the shorter Roman symbol must have originated considerably earlier than the middle of the third century. Otherwise how can we explain the fact that all the Western churches originally used the same symbol, and that the African Church, for example, had already developed its own special type, before the year 250, 33upon the foundation afforded by the old Roman symbol?³⁷³⁷Cyprian, Hahn, §§ 28, 29. Accordingly we must refer the Roman symbol to a date at least as early as the year 200, which admits of positive proof from the writings of Tertullian. Moreover, this conclusion is established by a comparison between the shorter Roman symbol and all the Western confessions of faith on the one side, and the provincial and private symbols of the East on the other; and, further, by a comparison of the shorter Roman symbol with the different editions of the Rule of Faith up to the middle of the third century.

The Eastern baptismal confessions are distinguished one and all by great flexibility, by freedom in form, and by richness of expression.³⁸³⁸See Hahn, op. cit. pp. 61 ff., pp. 183 ff.; Caspari, op. cit. ii. S. 112 ff., iii. S. 46 f.; Swainson, op. cit. p. 60; Hort, Two Dissertations, ii., On the Constantinopolitan Creed and other Eastern Creeds of the Fourth Century, 1876, p. 73; and, above all, Kattenbusch, S. 216 ff. As the Eastern 34Church never knew anything of any of the creeds having been composed by the Apostles, it always dealt with them in a much freer spirit, and in its baptismal confession gave expression at one and the same time to its interest in speculative theology and to its horror of every kind of heresy. It was mostly in the East that heresy originated. Thus the Eastern Church often puts dogmatic in the place of historical expressions, omits important passages, largely extends others by additional and preliminary matter, and interpolates anti-Gnostic, anti-Monarchian, anti-Modalistic, anti-Arian, anti-Semiarian, anti-Marcellian, anti-Photinian, anti-Pneumatomachian, anti-Apollinarian, and other 35observations. “The Oriental symbols frequently exhibit in their separate articles a greater or less freedom of form, whether by inserting dogmatic in place of simple historical expressions or by uniting the two, or by expressing the article in question in a somewhat fuller manner, or, finally, by making one or more additions not of a distinctly anti-heretical character. . . . Further, we often find that they contain whole articles wanting in the Western baptismal confessions. . . . As a general result the Eastern confessions exhibit, some in a higher and some in a less degree, a subjective, reflective and dogmatic character. They wear, moreover, a more or less parti-coloured appearance, and are more or less prolix, diffuse and verbose.” Lastly, catechetical instruction in doctrine, which, as is well known, was 36an accompaniment of the baptismal confession in the East, was much more strongly influenced by dogmatico-polemical theories than in the West. In the Eastern Church the symbol was accordingly in a constant condition of flux and movement. Not until the adoption of the Nicene-Constantinopolitan Creed was this state of things altered, and not even then was it completely altered. The Nicene Creed alone did not do it. From about the year 430 onwards this latter symbol supplanted the others in such parts of the territory of the orthodox Church as lay beyond the imperial jurisdiction. From that time the Byzantine Church became the home of severe conservatism in regard to the Creed, as up to the present day it has clung, persistently and exclusively, to the Nicene Creed. This state 37of things, which lasted in the East up to the middle of the fifth century, renders it difficult to describe the general characteristics of the Eastern symbols in their universality, and to reduce them to any fundamental type. Yet this much may be said: (1) That a considerable number of Eastern symbols—not all,³⁹³⁹See, e.g., the symbol of Gregory Thaumaturgus, Hahn, § 114. but certainly those of Syria and Palestine—are based on the same type;⁴⁰⁴⁰But, as Kattenbusch has proved, and as I previously maintained in my answer to Cremer’s polemic (Leipzig, 1892, S. 9 ff.), there is no universal, independent Eastern type of the baptismal symbol. (2) that in its range and the disposition of its articles this type exhibits an affinity with the shorter Roman symbol, but also the following variations from it:⁴¹⁴¹Caspari, ii. S. 44-88. 1. πιστεύομεν is almost always used, and in many symbols it is 38repeated with each article. 2. In the first and second article ἕνα is added to θέον and to κύριον. 3. In the first article, God is designated as the Creator of all things, that is, of Heaven and Earth. 4. The position of the words in the beginning of the second article is as follows: καὶ εἰς ἕνα (τὸν) κύριον Ἰησ. Χρ. τὸν υἱὸν αὐτοῦ τὸν μονογενῆ In the Western symbol the words Χρ. Ἰησ. stand first; τὸν υἱὸν αὐτοῦ τὸν μονογενῆ follow, and only after them comes τὸν κύριον. This order is almost everywhere preserved, and ἡμῶν is added to κύριον. 5. Frequently “ἐκ σπέρματος Δαβίδ,” or something similar, is added to the phrase γεννηθέντα κτλ. 6. In the East the separate clauses of the second article are run together, polysyndetically; in the West, asyndetically; there, the affirmations regarding 39Christ take the form of sentences placed in juxtaposition; here, of relative sentences. 7. The article τὸν ἐπὶ Ποντίου Πμλάτου σταυρωθέντα καὶ ταφέντα is almost entirely lacking; here and there it appears in a modified form. 8. The words τῇ τρίτῃ ἡμέρᾳ are placed after ἀναστάντα. 9. Instead of ἀναβάντα, ἀνελθόντα or ἀναληφθέντα is used. 10. The article concerning the “coming again,” is co-ordinated with the preceding. 11. Μετὰ δόξης or ἐνδόξως is added to πάλιν ἐρχόμενον. 12. In the third article the reading is τὸ πνεῦμα τὸ ἅγιον, or τ. ἅ. π. τὸ προφητικόν or something similar is often added. 13. ἐκκλησία has the predicate καθολική after the other predicate ἁγία. Where the former appears in the later Western symbols, it stands after “Ecclesiam.” 14. Baptism is frequently mentioned in the third article. 4015. The words ζωὴν αἰώνιον are found almost everywhere.

All these characteristics, however, attach to a set of symbols dependent on the symbol of Nicaea, or, as the case may be, on that on which it was based (the symbol which Eusebius laid before the Council at Nicaea, usually called the Caesarean); also on Lucian’s. This symbol, therefore, is not older than the beginning of the fourth century. The assertion would, of course, be open to challenge if the symbol produced by Eusebius were the baptismal confession of the Church of Caesarea.⁴²⁴²As Hort and Loofs, S. 673, maintain. Both assume that the third article is abridged. But the connexion in which Eusebius communicates the symbol in his letter to his community makes it anything but probable that it is the symbol or baptismal confession 41of that place. It ought, rather, to be regarded as a symbol which Eusebius had constructed expressly for the existing situation,⁴³⁴³This may be inferred from the predicates applied to Christ: the series beginning with τὸν τοῦ θεοῦ λόγον is evidently made for the situation. not, of course, ab ovo but according to the formulas familiar at Antioch or, as the case may be, in the schools of Origen and Lucian.⁴⁴⁴⁴See his symbol. That the congregation at Caesarea in the course of its instruction heard the faith which Eusebius here formulated is certain; but whether, over and above the baptismal confession, it possessed a definite creed consisting of three divisions is very questionable. Any such contention is strongly rebutted by the fact that in Eusebius’ formula the third article simply consists of πιστεύομεν καὶ εἰς ἓν πνεῦμα ἅγιον. 42The assumption that Eusebius made any omission from the church confession is a very dangerous one to make. There is also the fact that a long-winded sentence follows,⁴⁵⁴⁵Cp. Lucian’s symbol. ending in the general order to baptize. Eusebius regards this as belonging to the confession of faith as much as what preceded it: τούτων ἕκαστον εἶναι καὶ ὑπάρχειν πιστεύοντες, πατέρα ἀληθῶς πατέρα καὶ υἱὸν ἀληθῶς υἱὸν καὶ πνεῦμα ἅγιον ἀληθῶς πνεῦμα ἅγιον, καθῶς ὁ κύριος ἡμῶν ἀποστέλλων εἰς τὸ κήρυγμα τοὺς ἑαυτοῦ μαθητὰς εἶπε· πορευθέντες μαθητεύσατε πάντα τὰ ἔθνη, βαπτίζοντεσ αὐτοὺς εἰς τὸ ὄνομα τοῦ πατρὸς κτλ. This is evidently the reason why Eusebius as well as Lucian went on to the baptismal confession and repeated it in extenso; he felt the necessity of presenting his new formula as a paraphrase of the formula 43known to the community. But if the Caesarean symbol is not one framed for a particular community, then we know absolutely nothing of any definite, detailed, ancient communal symbols in the East of any date preceding the Nicene Creed. This negative conclusion is confirmed by four considerations: (1) by the curious symbol of Gregory Thaumaturgus,⁴⁶⁴⁶Hahn, § 114. and the equally curious one of Aphraates.⁴⁷⁴⁷Kattenbusch, S. 249. The argument seems to me unassailable that, where such “symbols” as these can be constructed, there is as yet no communal symbol, such as the Roman, in existence; and Gregory knew the Eastern Church from Pontus to Egypt. (2) By the free and easy way in which the symbols were formed and also accepted in the East. 44With pain and astonishment we see this process going on in the fourth and down to the middle of the fifth century. If any old symbols had been in existence, which had come down from previous generations, how could this state of chaotic confusion and lack of reverence in the formation and acceptance of creeds in the East be explained? (3) By the above-mentioned typical similarity of structure exhibited in the Eastern symbols of the fourth century, where the type of the Lucian-Eusebean-Nicene creed is almost the only one which emerges. (4) By the uncertainty about the third article which prevailed in the East up to the middle of the fourth century. Even as late as the first Antiochian formula of the year 341 it runs as follows: εἰ δὲ δεῖ προσθεῖναι πιστεύομεν καὶ περὶ σαρκὸς ἀναστάσεως κ. ζωῆς αἰωνίου.

In connexion with this last point I may observe that the construction of the old Roman symbol is perfectly clear. It is based on the baptismal formula with its three divisions. The first division is defined by the words, “God Almighty”; the second is characterised by the phrases “Only Begotten Son” and “Our Lord,” as well as by the historical account which it gives; the third is conceived of as a gift, and hence three further blessings are associated with it, which together express the content of the salvation which faith brings. Of the thirty Eastern confessions of faith from the fourth century which come into question more than two-thirds contain either no third article at all or else only a bare confession of belief in the Holy Ghost. Putting aside the symbols derived from the Nicene-Constantinopolitan,⁴⁸⁴⁸To which those mentioned by Hahn, §§ 68, 69, 70, belong. 46and also the obviously abridged symbols mentioned by Hahn, §§ 71, 72,⁴⁹⁴⁹As against Kattenbusch, i. S. 330. we find that the only symbol containing the third article in a complete form, or the more than complete form which mentions Baptism, is that in the seventh book of the Apostolic Constitutions, in the symbol handed by Arius to the Emperor, in that of Cyril of Jerusalem, in the symbol of Salamis (which developed into the Constantinopolitan), and in the longer symbol of Epiphanius.⁵⁰⁵⁰Hahn, § 68. These five symbols evidently go back to one common root, which is most visible in Cyril’s form, although it certainly does not easily lend itself to reconstruction. But in the close affinity which it exhibits with the old Roman 47symbol this very symbol takes precedence of all the rest. The relationship is so close that Cyril’s symbol can only be the daughter or the sister of the Roman one. That it can have been the mother is out of the question, as the Roman symbol undoubtedly reveals an older and simpler form. Hitherto there has been no reason for regarding it as even a sister, for the date of this set of Palestino-Syrian symbols is not earlier than the beginning of the fourth century, whilst we can certainly place the old Roman symbol a century earlier. Now, as regards the more than twenty Eastern symbols which possess only a rudimentary third article or none at all, it is clear from the way in which christological attributes are accumulated, even in the oldest of them, that we are dealing with symbols of late origin. Still, 48however, the formula θεὸς πατὴρ παντοκράτωρ, and the structure of the christological section, unmistakably exhibit a certain affinity with the Roman symbol. Moreover they almost all possess, in common with the former group, additions to the first article, as well as the ἕνα in the first and second. Finally, there is a certain grammatical and literary character common to them all. Hence the simplest solution of the problem presented by the relation between the Eastern confessions of faith of the fourth century and the old Roman symbol, is to say that, whilst there was no established baptismal confession of faith in the East in the third century, there was, however, an old, flexible “christological rule,” and also old, ceremonial or polemical formulas of belief in One God the Creator, and in His Only 49Son Christ. Apart from the singular confession of Gregory Thaumaturgus, the venturesome character of which is apparent in the very extravagances of the legend connected with it, we may say that it was towards the end of the third century, probably in the school of Lucian, at all events at some point in Syria-Palestine, that the formation of symbols began in the East, where men—first, it seems, in theological circles—had come to know and value the Roman symbol. At the period of the struggles with Paul of Samosata other features of the Roman Church also came to be appreciated. The direct and full acceptance of the Roman symbol was, however, hindered by (1) the circumstance that the christological section of the Roman symbol came into conflict with a christological type already established; 50(2) by the desire to give fuller expression to the “higher christology” in the creed. It was not until the time of the Arian controversy that fixed symbols in the East began to be formed. The type⁵¹⁵¹Lucian, Eusebius, Arius, § 117, the Nicene, the whole of the Antiochian and Sirmian symbols, etc. that was apparently, at least, the most frequent up to the year 381, was that with the short third article (in “the,” or, as the case might be, the “One,” “Holy Ghost”; or also, in some instances, with additions such as “Who spake by the Prophets”); whilst the type which, in the third article, is in essential agreement with the old Roman symbol came to the front in the Jerusalem-Salamis symbol, and in that contained in the seventh book of the Apostolic Constitutions,⁵²⁵²How old this symbol may be is a question. and 51then gradually gained the supremacy through the Nicene-Constantinopolitan Creed.

The question may be asked whether this conclusion is not upset by an examination of the Rules of Faith, and the fragments of those rules and formula-like sentences with which we are familiar as belonging to the Eastern half of the Church from the middle of the first to the middle of the third century. This is the opinion entertained by Caspari, Zahn, Loofs, and many others, and formerly I, too, shared it. The idea is that we must take an Eastern symbol or, to be more precise, a symbol from Asia Minor, and relate the old Roman symbol to it as daughter or sister. The assumption rests principally if not exclusively oft what we find in Clement of Alexandria, 52Irenaeus, Justin and Ignatius. The opponents of this view argue briefly as follows:—“The writings of Justin, who was baptized in Ephesus about the year 130, show us that he assumes the existence of a symbol which on the one hand much resembles the old Roman, and on the other is most characteristically distinguished from it. These distinguishing marks also appear in the majority of the later Eastern symbols (Ἰησοῦς Χριστός not Χ. Ἰ.; σταυρωθεὶς ἐπὶ Π. Π. not ἐπὶ Π. Π στ.; ἀποθανόντα; πάλιν μετὰ δόξης etc.); further, they are also to be found in the formulas of Irenaeus, who employs others as well as ἕνα, ποιητὴν οὐρανοῦ καὶ γῆς, and in certain peculiarities of style which may also be shown to exist in Eastern symbols of the fourth century. Some of these can be traced back as far 53as Ignatius, nay, even to the Epistles of St. Paul, or, in fact, to the New Testament in general. Finally, it follows from what Clement says that in his time there existed a formal and fixed baptismal confession in Alexandria. In the East, then, there existed in the second century a fixed symbol, or, rather, many symbols, related to the Roman symbol, but independent of it. The history of Eastern symbols may therefore be traced well into the second century, and this history, accordingly, though latent in the third century, was still existent. The Roman symbol at best is contemporaneous with the Asiatic or Syrian; more probably it is later; and this Asiatic or Syrian symbol leaves it free to the critic to assign it to the years 120-130, 100-120, or 70-100.” Such is the argument.

Against it four considerations may be urged:

(1) The fact that single sentences seem to be echoes of the symbol or tally with it offers no guarantee that they themselves derive from one symbol. Before any symbol existed God was παντοκράτωρ; Jesus Christ was called “the Only Begotten Son, our Lord”; he was proclaimed as “born of the Holy Ghost of the Virgin Mary,” as having suffered under Pontius Pilate, and as coming to judge the quick and the dead.

(2) Formula-like sentences, if not obviously a part of the baptismal formula, need not necessarily have originated in a baptismal confession, even though they be identical with the sentences of that confession. The oldest tradition gave a fixed or, as the case may be, a more fixed 55shape to “The Faith,” not only in the form of a baptismal confession and for the purposes of baptism, but also in (a) liturgical sentences, (b) formulas of exorcism, (c) precepts concerning faith and morals, and (d) historical summaries, and that, too, with a view to the most diverse objects (instruction, apologetics, polemics, religious worship). As illustrating (a) we may take the prayers in the Didaché; (b) statements in Justin and others; (c) Hermas, Mand. 1 and Didaché 1-6; (d) 1 Cor. xv. 1 ff., Mark xvi. 9 ff. The words of John xvii. 3 ἵνα γινώσκωσι σὲ τὸν μόνον ἀληθινὸν θεὸν καὶ ὃν ἀπέστειλας Ἰ. Χρ. were in the middle of the second century as much a formula of faith as Hermas, Mand. 1 πρῶτον πάντων πίστευσον, ὅτι εἷς ἐστιν θεός κτλ., yet they have nothing to do with the baptismal formula. Such 56passages as Ephes. iv. 9 furnished themes for homiletical discourses; formulas were also set up which led from the confession of the One God to the chief practical commandments; of these some fine and powerful examples are found in Mand. 1 ff. and Didaché 1 ff. Finally, the preaching of Christ is not unfrequently attached, on the foundation of numerous Pauline passages, to a confession of belief in the One God, without any mention of the Holy Spirit, of the Church, or of Christian blessings.

(3) In particular, the preaching of Christ, apart from the detailed form which it received in the Gospels,⁵³⁵³Luke i. 4. also underwent various longer or shorter epitomisations,⁵⁴⁵⁴See the above-mentioned fragments 1 Cor. xv. and Mark xvi. 9. 57which took a fixed form without being placed in a Trinitarian framework. These epitomisations proceeded on various plans: (a) the mere chronicle, (b) the chronicle with proofs attached, (c) the plan of fulfilled prophecy, (d) the plan κατὰ σάρκα κατὰ πνεῦμα, (e) the plan of the first and second coming, (f) the plan καταβάς—ἀναβάς. All these plans, in part united with one another, issued in affirmations of a character relatively fixed, even if capable of being modified.

(4) Out of the great number of predicates attached to God, Christ, and the Spirit, some which were in general use very soon came to the front, apart from the detailed Trinitarian confession. Those chiefly used in connexion with God are, εἷς, παντοκράτωρ, πατήρ, δεσπότης and Creator, with additions; with Christ, ὁ υἱὸς τοῦ θεοῦ, ὁ 58κύριος, σωτήρ, διδάσκαλος, μονογενής, εἷς, λόγος; with the Holy Ghost ἄγιος, προφητικός. In the same way, out of the great number of blessings which the Christian faith affords, some are named with especial frequency, such as ἄφεσις ἁμαρτιῶν (with or without mention of baptism), ζωή (αἰώνιος), ἀνάστασις (with or without τῆς σαρκός), γνῶσις, ἀφθαρσία, etc. Everything thus variously produced was regarded as “the Faith,” “the Rule of Faith,” “Kerugma” (or “Proclamation”), “Truth,” “Rule of Truth,” μάθημα, παράδοσις, ταραδοθείς λόγος, διδαχή, etc.

A consideration of the facts contained in the foregoing, the truth of which no scholar will question, must make us very cautious in arguing from formula-like confessional sentences to a formulated baptismal confession in three parts. Caution of 59this kind seems to be everywhere wanting at the present time, as is seen, for example, in Zahn’s treatise on the Apostles’ Creed (1893) and in the way in which it has been received by the most distinguished students in this branch of learning. No one has a right to claim a particular proposition, which forms no part of any creed framed on the Trinitarian plan, as part of a fixed baptismal confession, unless he is in a position to offer very strong evidence for his contention.

What is the net result of the “testimony” of Ignatius, Justin, Irenaeus, and Clement of Alexandria?

(1) We find that Ignatius has freely reproduced a “kerugma” of Christ which seems, in essentials, to be of a fairly definite historical character and which contained, inter alia, the Virgin Birth, Pontius 60Pilate and the ἀπέθανεν. There is no trace of any evidence, however, that it was part of any creed based on the Trinitarian plan.

(2) As to Justin we find (a) that he knew of a definite christological “kerugma,” and used it again and again; this was closely related to the second article of the Roman symbol, although quite independent of it, and it even exhibits many of the characteristic peculiarities of the later Eastern symbols; (b) that with him this “kerugma” forms no part of any baptismal symbol, that is to say, is not a formal second article; (c) that with him the baptismal formula was not developed into a symbol at all, except that the three Persons were described as follows: ὁ πατὴρ τῶν ὅλων καὶ δεσπότης θεός, Ἰησοῦς Χριστὸς ὁ σταυρωθεὶς ἐπὶ Ποντίου Πιλάτου, τὸ πνεῦμα 61ἅγιον ὃ διὰ τῶν προφητῶν προεκήρυξε τὰ κατὰ τὸν Ἰησοῦν πάντα, or, simply, τὸ προφητικόν any such description, however, in the baptismal formula itself, is improbable; (d) that it is extremely likely that the christological “kerugma” above indicated was formally stated as fulfilled prophecy, that is to say, stood as part of a plan as follows: “The Holy Ghost prophesied etc.”; but we can go no farther in this direction than the assumption that Justin knew of a “kerugma”; that after the mention of πατὴρ τῶν ὅλων καὶ δεσπότης, and Jesus Christ, a “kerugma” of Christ, in the form of fulfilled prophecy or, as the case may be, in the form of a belief in the prophetic spirit, was added. But the contention that this μάθημα was a baptismal confession, or, as the case may be, claimed to be a developed baptismal formula, and that it existed in 62a crystallised form at all, is unsupported by any evidence.

(3) As regards Irenaeus, (a) as I have shown in the first article against Zahn in the Zeitschrift für Theologie und Kirche, Bd. iv. S. 149 ff., we must be very cautious in drawing conclusions from his “testimonies on behalf of the baptismal confession”; a very small portion of the material which I collected from Irenaeus in the treatise on the old Roman symbol⁵⁵⁵⁵Patr. App. Opp. edit. 2, T. i. 2, pp. 123 ff. is sufficient to determine the “symbol” which he employed; (b) according to Irenaeus i. 9, 4 baptism bestows the κανὼν τῆς ἀληθείας; this canon he himself communicates in i. 10, 1. The form in which he here produces it, supplemented by the watchwords of his theology, and given in other places with fragmentary variations, 63shows that he is compiling it independently out of a large number of fixed confessional formulas of the Church. Among these may be distinguished:

i. The expanded formula of Hermas.⁵⁶⁵⁶Mand. 1.

2. The formula εἷς θεὸς παντοκράτωρ united with Johannine expressions or, as the case may be, with πεποιηκῶς τ. οὐρανὸν κ. τ. γῆν κ. τ. θαλάσσας καὶ πάντα τ. ἐν αὐτοῖς, or εἱς μονογενὴς Ἰησοῦς Χριστός.

3. A christological formula of confession (in an historical form), showing a close relation to the old Roman symbol, but a still closer one to Justin’s.

4. The θεὸς πατὴρ παντοκράτωρ of the Roman symbol.

5. A formula of confession which to the confession of belief in the One God and One Christ Jesus joined a confession 64of belief in the Holy Spirit, and incorporated with this confession the history of Christ as fulfilled prophecy. As we were enabled to make a similar conjecture in Justin’s case, so it is probable that not only in Irenaeus’ time but also in Justin’s a confessional formula existed in the East containing something like the following:—ἡ εἰς ἕνα θεὸν παντοκράτορα (or εἰς τὸν πατέρα τῶν ὅλων καὶ δεσπότην θεὸν) πίσστις καὶ εἰς ἕνα Ἰησοῦν Χριστὸν τὸν υἱὸν τοῦ θεοῦ, τὸν σαρκωθέντα ὑπὲρ ἡμῶν (or ὑπὲρ τῆς ἡμετέρας σωτηρίας) καὶ εἰς πνεῦμα ἅγιον, τὸ διὰ τῶν προφητῶν κεκηρυχὸς τὰς οἰκονομίας, τὴν ἐκ παρθένου γέννησιν κτλ. From this formula, which Irenaeus made the foundation of his κανῶν τῆς ἀληθείας, the historico-christological formula of confession containing the sentences about the birth, suffering under Pontius Pilate, burial, resurrection, 65and coming again in glory (in finite verb or, as the case may be, participle) is perhaps, or even probably, to be distinguished. Parallels are also to be found for this formula in Justin and Ignatius or, as the case may be, in 1 Cor. xv. This is as far as the material hitherto discovered will allow us to go on this subject. That Irenaeus assumed the existence of a symbol, or, in other words, that the formulas (plans) indicated above were in existence in their crystallised form, not only cannot be demonstrated but is entirely improbable. Irenaeus’ whole line of argument must have issued in a different conclusion had there existed in a fixed form, recognised in his community, what is necessary for his demonstration, that “Multa,” that is to say, many familiar formulas and short statements of faith, 66existed, but no “Multum,” that is to say, that there was no symbol. There is nothing in the objection that Tertullian proceeds in a similar way, and that he certainly assumes the Roman symbol to be already known. Tertullian’s references to a symbol are incomparably clearer.⁵⁷⁵⁷See the evidence adduced in my above-mentioned treatise, and Kattenbusch, i. pp. 141 ff. But that he had to serve up to his readers as Apostolic tradition the quid pro quo, that is to say, formulas constructed ad hoc, followed from the fact that the text of the Roman symbol was insufficient for the theological and anti-Gnostic objects which he had in view. We may, however, ask whether the Irenaeus of Asia Minor and Gaul had ever heard of the Roman symbol. In view of the distinct formula θεὸς πατὴρ παντοκράτωρ, and the way in which he uses 67the Roman community as evidence in his argument for tradition, I am disposed to assume that he had.

Lastly, as regards Clement of Alexandria, there is a still unsettled controversy as to whether he does not in one place assume the existence of a fixed symbol in that city. Even if this be so—it seems to me still extremely doubtful—there is no art which can discover how this symbol ran. It may have been something entirely different from what we call a symbol. Therefore we may leave it out of account.

That there existed in Asia Minor, or in Syria, or, in short, in the East before the beginning of the third century, symbols used as baptismal confessions which were based upon the baptismal formula, gave the second article in the form of an historical account, and summarised in the third the 68blessings which faith receives, cannot be shown. To prove the existence in the East at all, in the earliest period, of any fixed crystallised confession, and therefore of a primitive Eastern symbol closely related to the old Roman one, but still independent of it, is impossible. Not only can the existence of any such primitive symbol not be proved, but it is quite improbable, as the history of the Eastern Church shows in the third century by its silence, and in the fourth by what it says. Nevertheless the result of our investigations is not merely negative. On the contrary, we can agree that those who defend the existence of a primitive typical Eastern symbol are, up to a certain point, right. There did actually exist in the East (in Asia Minor or, as the case may be, Asia Minor and Syria), as early as 69the beginning of the second century, inter alia a christological μάθημα, which is most intimately related to the second article of the Roman creed, and which, as regards the formulas and details peculiar to it, made its way into the Eastern symbols of the fourth century. There existed, further, formulas referring to One God, the Creator of Heaven and Earth, and to His Incarnate Son, which also made their way, and exerted an influence on the whole process of forming symbols, including many modifications of the Roman symbols in the West. The exclusively theological tenor of the Eastern symbols in the second article may be traced to the primitive σαρκωθέντα. Finally, there existed a formula which asserted of the holy prophetic Spirit the facts which it proclaimed in regard to Christ. Apart from these leading formulas 70the words “descensus” and “catholica” point to the East. But nevertheless the great feat of having formed the symbol, and of therewith laying the foundation of all ecclesiastical symbols, remains the glory of the community at Rome.

When did this happen? We have traced the old Roman symbol to the time of Tertullian. It is this symbol that he means when he writes de praescr. haer. 36: “Si autem Italiae adjaces, habes Romam, unde nobis quoque auctoritas praesto est . . . videamus quid didicerit, quid docuerit, cum Africanis quoque ecclesiis contesserarit. Unum deum dominum novit, creatorem universitatis, et Christum Jesum ex Virgine Maria filium dei creatoris et carnis resurrectionem . . . et ita adversus hanc institutionem neminem recipit.” This symbol we unhesitatingly 71trace back to about the middle of the second century. Had a symbol been established in Rome at the time of the fierce struggle with Gnosticism and Marcionism (about 145-190), it would have taken a different form; on the other hand, to go back too far beyond the middle of the second century is unwise. There are a great many things in the Shepherd of Hermas, both as a whole and in its several parts, which would be difficult to explain if the Roman symbol had been familiar to the writer. Justin shows us that about the middle of the second century the distinction between ἐκ and διὰ Μαρίας had not yet been effected. The omission of Jesus’ baptism by John, and also of the Johannine expression υἱὸς μονογενής, the omission of the chiliastic hopes, and the sharp distinction between 72ἀναστάντα, ἀναβάντα and καθήμενον, are facts to be seriously weighed. In addition, the expression θεὸς πατὴρ παντοκράτωρ has no history behind it, and it gradually displaced an older expression εἷς θεὸς παντοκράτωρ. This I have already shown in the Zeitschrift für Theologie and Kirche,⁵⁸⁵⁸Bd. iv. S. 130 ff. in which I refuted Zahn’s hypothesis that the old Roman symbol originally began with the words πιστεύω εἷς ἕνα θεὸν παντοκράτορα. The old Roman symbol always ran as it now runs, but the text of its first article must have made its way in opposition to an older and very wide-spread form of the confession of God as the Creator. To Hermas the formula θεὸς πατὴρ παντοκράτωρ is as yet unknown. This also makes it probable that the symbol originated about the middle of 73the second century or shortly before. The text, too, of the Eastern christological μάθημα, which was presumably known to the author of the old Roman symbol, is, if it contains Jesus’ Baptism by John and does not mention the Ascension, older than the Roman symbol, just as the παθόντα, ἀποθανόντα, as well as πάλιν and ἐν δόξῃ, can be put a long way back. Finally, if we examine the Roman symbol clause by clause, the following facts are established:

(1) The symbol itself is the oldest witness for the formula θεὸς πατὴρ παντοκράτωρ, which gradually superseded an older form.

(2) υἱὸς ὁ μονογενής is Johannine.

(3) The oldest and frequently recurring formula for the Virgin birth always runs γεννηθέντα ἐκ Μαρίας τῆς παρθένου. The 74addition of ἐκ πνεύματος ἁγίου in the “kerugmatic” sentences is relatively late, and presumably comes from the Gospels.

(4) The ταφέντα there is in like manner late.

(5) The addition of τῇ τρίτῃ ἡμέρᾳ to ἀναστάντα. Both come from the First Epistle to the Corinthians.

(6) The special prominence given to ἀναβάντα between ἀναστάντα and καθήμενον is also relatively late, and shows a desire for completeness which is best explained by the high regard felt for the only existing account.⁵⁹⁵⁹Acts i.

(7) The enumeration of the blessings of salvation, as given in the third article, cannot be understood apart from the Pauline Epistles, but it lends a precision to what was taken from those Epistles by 75the particular prominence given to the Resurrection as a Resurrection of the body. The fact that in the Roman symbol older and shorter “kerugmatic” sentences were somewhat further developed under the influence of the New Testament writings, and particularly under that of John, the Synoptists, Paul, and probably the Acts of the Apostles, makes it unwise to trace the composition of the symbol backwards beyond the middle of the second century.

To sum up: the symbol originated in Rome about the middle of the second century. It was based upon the baptismal formula and on confessional formulas of a summarising character (such as we can identify from the New Testament and from Ignatius, Justin, and Irenaeus), which 76had been generally handed down, including Eastern formulas (Asia Minor, Syria), as also largely under the influence of the New Testament writings. Among these confessional formulas the most important was a christological μάθημα of fairly fixed form, yet capable of being added to and modified. Its main outlines, I presume, are recognisable. In Rome itself the Roman symbol was never altered. It made its way into the Western provinces from the end of the second century onwards, without raising any claim to have been, in the strictest sense, composed by the Apostles. That is why it underwent different modifications in those provinces. (In Rome it was not until some time between 250 and 350 onwards that it was designated as Apostolic in the strict sense of the term.) Amongst these 77modifications, those became historically the most important which derive from the primitive confessional formulas of the East or, as the case may be, the μάθημα, namely, “creator of heaven and earth,” “suffered,” “died,” “descended into hell,” “eternal life,” besides the “catholica”—these are just the modifications traceable in the Gallic symbols which issue in our Apostles’ Creed—in addition, the “conceptus,” which is obscure in its origin and otherwise of little importance, and, most perplexing of all, the “communio sanctorum.” In this connexion may rightly be borne in mind the particularly close relations existing between Southern Gaul and the East. But an historical circumstance of very special importance seems also to have played a part. Hitherto I have said nothing about 78the Symbol of Nicetas.⁶⁰⁶⁰Caspari, Anecdota, S. 341 ff.; Kattenbusch, i. pp. 108 ff.; Hahn, § 25. Morin⁶¹⁶¹Rev. bénédict. Tom. xi. Febr. has made it very probable that Nicetas means the Nicetas of Remesiana in Dacia, the friend of Paulinus of Nola.⁶²⁶²His date is the beginning of the fifth century. The symbol which he adduces can unhappily be no longer reconstructed in detail from his Explanatio; but so much is certain, that it is closely related to the old Roman symbol. What is much more interesting, however, is the fact that throughout (partly word by word) he explains it by the catechising activity of Cyril of Jerusalem, and in this connexion brings in the sentence “Ergo in hac una ecclesia crede to communionem consecuturum esse sanctorum.” Whether the catchwords belong to Nicetas’ symbol is very questionable 79(to me improbable); but in any case, so far as their origin is concerned, their presence there could be explained by reference to Cyril’s words. As there is a certain relationship between Nicetas’ symbol and the Gallican (we may ask whether his symbol was not even influenced by Cyril’s), and as connexions between Gaul and Pannonia are not lacking, the possibility presents itself—more than this I will not say at present—of conceiving the Gallican symbol, with the clause “communio sanctorum,” that is to say, our Apostles’ Creed, as having arisen about the year 500 under the indirect influence of Cyril’s catechising (carried on throughout the Remesiana in Pannonia and Aquileia). Loofs⁶³⁶³Loofs, S. 677. and I⁶⁴⁶⁴Theologische Literaturzeitung, 1894, Kol. 582. have 80indicated this possibility independently of each other. At all events a piece of ecclesiastical “ecumenicity” adheres to a part of the additions which distinguish our Apostles’ Creed from the old Roman symbol. If “communio sanctorum” is not to be traced to Cyril, but to be regarded, rather, as a product of chance, it must be understood in Augustine’s sense (i.e., the Church as the community of the Saints), or, with Faustus of Rietz, as a fellowship with the martyrs and specially holy men. Zahn⁶⁵⁶⁵Op. cit. pp. 82 ff. has recently suggested another derivation, namely, that “communio sanctorum” is equivalent to τὴν κοινωνίαν τῶν ἁγίων, the latter meaning “sacramenta.” Sub judice lis est.”

That the Roman Church after the beginning 81of the sixth century gradually let itself be separated from and finally robbed of the symbol which it had previously guarded so faithfully, is a striking phenomenon which has not yet had its causes clearly explained. Meanwhile, however, Caspari⁶⁶⁶⁶Op. cit. ii. S. 114 f., iii. S. 201 f., 230 f. has made some very important contributions towards a solution of the problem. The most critical fact that it was not in the first instance the longer (Gallican) daughter edition (our Apostles’ Creed) which displaced the mother symbol but the Nicene-Constantinopolitan, which from the beginning of the sixth century first took the place of the shorter one in Rome in the Traditio and Redditio symboli, while in the baptismal interrogation the old Roman still remained in use. The displacement of the old Roman symbol 82by the Constantinopolitan becomes very intelligible, as soon as we consider the conditions of the time. From the end of the fifth century, under the dominion of Odoacer and the Ostrogoths, Arianism had impinged upon the Roman Church, and had become a danger to it. By way of counteracting it the Roman Church will have resolved to give up its ancient practice, so as in its very baptismal formula to express its disavowal of Arianism. When three centuries later the church returned to a shorter symbol, the old Roman one had already retreated into the background, and the new Roman symbol, which was, in fact, the Gallican, the Apostles’ Creed, possessed the recommendation of having a series of elaborations which were wanting in the earlier one, and which now seemed indispensable. 83But we may probably also assume—direct information we have, of course, none—that the Roman Church would have difficulties about accepting the Frankish symbol as a baptismal one, had it not been recognised as an old acquaintance. It is, moreover, very probable that there was still enough historical tradition present in Rome to allow of the Frankish confession reminding people of one that was old and once highly honoured. The differences were overlooked or else not regarded as considerable. Thus the legend which had encircled the old symbol with a halo of glory awoke again around the new one, and again and for a long time became a power in the Church. Not until the age of the Renaissance and the Reformation was it exploded.