Article Information

Annemaré Kotzé1

1Department of Ancient Studies, Stellenbosch University, South Africa

Contribution to ‘Augustine and Manichaean Christianity’, the First South African Symposium on Augustine of Hippo, University of Pretoria, 24−26 April 2012.

Correspondence to:
Annemaré Kotzé

Postal address:
PO Box 3350, Matieland 7602, South Africa

Received: 02 Nov.2012
Accepted: 10 Jan. 2013
Published: 10 Apr. 2013

How to cite this article:
Kotze, A., 2013, ‘Cui narro haec? Augustine and his Manichaean audience: A re-reading of the first three books of the Confessions’, HTS Teologiese Studies/Theological Studies 69(1), Art. #1357, 8 pages.

Copyright Notice:
© 2013. The Authors. Licensee: AOSIS OpenJournals.

This is an Open Access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Cui narro haec? Augustine and his Manichaean audience: A re-reading of the first three books of the Confessions
In This Original Research...
Open Access
A 'Manichaean reader'?
A protreptic or a paraenetic text?
The Confessions as a protreptic: An anomaly
Communicative purpose and audience in Books 1−3
   • Competing interests

The issue of intended audience in the first three books of Augustine’s Confessions is investigated in light of the presence of terms and phrases that may have had special connotations for potential Manichaean readers. This is done against the background of definitions of protreptic and paraenetic, which typically revolve around audience location and communicative purpose. Although it has become commonplace to refer to the Confessions as a protreptic the work displays a number of characteristics more in line with current mainstream definitions of paraenetic, amongst other things, by assuming the stance of addressing insiders in agreement with the author’s world view. It is argued that the type of reader most receptive to the insider stance and allusion to the Old Testament on the one hand and to the Manichaean material on the other, would be a Manichaean apostate recently converted to Catholic Christianity.


This article forms part of an effort to rethink the protreptic function and the Manichaean audience of the Confessions,1 specifically in light of the pervasive presence of Manichaean terms and concepts, echoes of Manichaean documents and the structuring role of Manichaean issues in the work as a whole, pointed out by experts on Manichaeism like Johannes van Oort (1997, 2002) and Jason BeDuhn (2010) before. In a previous publication (Kotzé 2004) I argued that such material contributed to one of the communicative purposes of the work, namely a protreptic purpose, which was directed specifically at a Manichaean audience. My quest here is to re-examine this hypothesis in light of a recent re-investigation of the nature of protreptic,2 the considerable insights gained by reading BeDuhn’s Augustine’s Manichaean Dilemma (2010) and a re-reading of the first three books of the Confessions. The article is also to an important extent a revisiting of and an elaboration on the points made by Johannes van Oort in his pioneering 1997 article, ‘Manichaeism and anti-Manichaeism in Augustine’s Confessiones’, where he investigates the presence of Manichaean material in the first three books in light of his reading of the anti-Manichaean passages in Confessions 3.6.10–3.10.18.

Before I continue the discussion of Books 1−3 it is essential to make a few points about the relationship between protreptic purpose and the presence of Manichaean material in the Confessions. Firstly, if the narrative is regarded as primarily anti-Manichaean, that is, designed to communicate to its audience – presumably mainly Catholic insiders – arguments against certain Manichaean positions is not necessarily protreptic (see the definitions and discussion below). Anti-Manichaean elements can form part of a protreptic to Manichaean readers, but they do not automatically constitute a protreptic purpose. Secondly, also a primarily apologetical purpose (which I understand here as a self-defence against accusations or suspicions of crypto-Manichaeism)3 is certainly not identical with a protreptic purpose. Thirdly, it is necessary to consider the possibility that the presence of the Manichaean material is simply a manifestation of the degree to which Augustine’s conceptual framework has become shaped by Manichaean categories of thought and that such terminology is not employed in pursuit of any conscious strategy. Of course these possibilities are not mutually exclusive; it is precisely my argument that in a multi-dimensional text like the Confessions various such communicative purposes or elements may be expected to coexist. The focus of the current article is, however, mostly on the presence or absence of protreptic purpose.

I have already mentioned in my earlier arguments that the Confessions is a protreptic addressed to Manichaean readers. Whilst it has become more or less commonplace to refer to the Confessions as a protreptic (Feldmann 1994; Mayer 1998; Lancel 2002:210; Hübner 2007:51; Lössl 2011:47–54), many scholars who use the term do not specify the kind of audience they think the protreptic is designed to target. BeDuhn (2010:129–130), however, in his remarks on a decidedly positive depiction of Faustus in the Confessions, sees this as ‘an essential part of the purpose of the Confessions as a protreptic appeal to the Manichaeans.’ Manichaean scholars in general would probably not have serious problems with the idea that much in the Confessions is designed to reach a Manichaean audience. Yet, I have encountered some resistance to the idea that the Confessions can be understood as addressed to Manichaean readers. Apart from the fact that Catholic insiders are regarded as the privileged addressees of the work, it is argued that Manichaeans would simply not have read the strong anti-Manichean rhetoric in Confessions 3, for example. Further, my survey of recent research on protreptic as well as my continued thinking about the intended audience of the Confessions have caused me to reconsider: the phrase ‘protreptic to the Manichees’ may for a number of different reasons not be successful in conveying the sense it is meant to communicate. These reasons concern both elements implicit in the phrase: ‘protreptic intent’ as well as ‘Manichaean reader’.

A ‘Manichaean reader’?

I start by formulating a few thoughts on the different shades of meaning the phrase Manichaean audience or Manichaean reader might imply. For example, one can assume that the Manichaean Elect – those carrying the responsibility for the Manichaean system of thought – would constitute a different kind of Manichaean audience for the Confessions than Manichaean Auditors. It is reasonable to envisage a member of the Elect reading the Confessions in the manner in which Augustine read Faustus’s Capitula: purely in order to write a systematic refutation.4 But it is also possible to conceive of an Auditor reading the Confessions simply out of curiosity5 about Augustine’s life but not so seriously concerned about his criticism of some of the more philosophical or theological aspects of Manichaeism. (The reaction of this reader to the very strong anti-Manichaean rhetoric in Confessions 3 may be a different matter, but I will say more about this later.)

There is another category of Manichaean reader that I think one should consider: namely one already interested in Catholic Christianity as an alternative for Manichaean Christianity, someone, like one of the friends earlier converted to Manichaeism by Augustine or like Augustine himself some 15 years earlier, who – because of intellectual objections or the increasing clampdown on Manichaeans6 – seriously considers conversion to Catholic Christianity, or who has, in fact recently converted to Catholicism. In the past I also included in the spectrum of likely readers for the Confessions Catholics under pressure from Manichaean proselytising or those considering conversion to Manichaeism. In light of BeDuhn’s perspectives (2010:136–144, 196) on the growing persecution of Manichees from the 380s onwards it is debatable how significant a possibility this would have been at the time Augustine wrote the Confessions.

But the likelihood of the Confessions being designed as a protreptic to Manichees who have already showed interest in Catholic Christianity, or a paraenetic exhortation (see the discussion below) to those who may have recently converted to Catholicism, has to be considered. This would be in line with recent interpretations around the intended audience(s) of another protreptic text, Clement of Alexandria’s Protreptikos,7 where Von Stockhausen (2006:87–90) has argued that the intended readers are pagans already interested in or perhaps even already being instructed in Catholicism. Because this last type of Manichaean reader – or perhaps rather ex-Manichaean reader – features prominently in the rest of the article, I shall use the term ‘liminal Manichaean’ here as shorthand to refer to him or her.8

A protreptic or a paraenetic text?

The next section of the article is a concise exposition of some thoughts on protreptic and how it relates to paraenetic and apologetic, but still with the specific purpose of illuminating my arguments about potential Manichaean readers as part of the intended audience of the Confessions. For it is clear that arguments concerning the communicative purpose of a text are inextricably linked to arguments about the audience it aims to communicate with.

In recent research on protreptic, the term has come to be defined in counterpoint and as standing in a dichotomous relationship to a closely related term, parainesis. In such research the words protreptic and paraenetic have acquired dichotomous technical meanings and become the subject matter of separate research areas: biblical scholars focus on paraenetic (e.g. Perdue & Gammie 1990; Popkes 1996; Starr & Engberg-Pedersen 2004), whilst those working on ancient philosophy investigate protreptic (e.g. Jordan 1986; Slings 1995; Van der Meeren 2002). Mainstream definitions of protreptic and paraenetic revolve around communicative purpose and audience location: protreptic is characterised by the purpose to convert, parainesis by the purpose to confirm belief or strengthen the resolve of the audience; the intended audience of protreptic literature is the not-yet-converted (outsiders) who have to be persuaded of the validity of a certain way of life, whilst parainesis is aimed at an audience of the already converted (insiders) already sharing the author’s world view.9

My investigation of ancient protreptic over the last three years has persuaded me to follow Diana Swancutt (2004) in challenging these assumptions.10 I have analysed a number of ancient remarks (Kotzé 2011) usually adduced to argue for such a dichotomous view of protreptic and paraenetic, as well as a wide range of other examples, specifically in order to understand the relationship between the two terms (and not with a view to defining either one whilst neglecting the other as is more frequently the case in research of these terms). I have come to the conclusion that in ancient literature the two terms are almost consistently used interchangeably and synonymously.11 At the same time the texts categorised as being either protreptic or paraenetic more frequently display a mixture of characteristics ascribed to protreptic and paraenetic respectively than those of one type exclusively.

I also believe that there is another literary type that has to be studied in conjunction with protreptic and paraenetic for a nuanced understanding of these concepts to emerge, namely apologetic. In addition to a widespread conviction that a defence against specific (and explicitly named) accusations should be present to make a work formally part of the apologetic genre,12 the debate about the nature of apologetic literature, similar to that about the nature of protreptic and paraenetic, often revolves around communicative purpose and audience location. It is asked whether 2nd century Christian apologetic texts, for example, are in fact aimed at outsiders (the Roman rulers to whom they are formally addressed) or perhaps also – or rather – at insiders.13 In fact, in the case of both protreptic and apologetic modes of discourse that at least outwardly seem to target outsiders, more and more voices argue that these works are (also) directed at insiders and aimed at facilitating identity construction.14 There are, of course, also a number of texts where protreptic and apologetic purposes are commonly assumed to co-exist, like the Confessions, Clement of Alexandria’s Protreptikos or one of the most influential early examples, Plato’s Apology of Socrates.15

Categorisation in terms of communicative purposes and target audiences of each of these types proves to my mind unsatisfactory, mainly because works categorised as any one of these types frequently display a mixture of purposes and intended audiences: the same work may exhort outsiders towards conversion, confirm belief and aid identity construction in insiders, as well as defend against conceived accusations against the group it represents, with both outsiders and insiders in mind. There is one respect in which current distinctions between protreptic, paraenetic and apologetic do seem valid. It is possible to distinguish three types of rhetorical stance: the stance of trying to convince an outsider to convert (currently defined as protreptic), of confirming belief in insiders who share the speaker’s world view (paraenetic in current definitions), or the stance of defending the belief of the group against accusations from outside (apologetic in current definitions), whatever the ‘real’ intended audience or communicative purpose of the work may be.

Within such a paradigm the Confessions may be more aptly designated a parainesis, in the sense that it – at least on the surface – assumes a stance that presupposes a shared outlook between narrator and implied reader. It is formally addressed to the most insider of insiders, namely God himself, but also the frequent inclusion of the reader to share Augustine’s views and speak with his voice through the use of first person plural pronouns and verbs that create the impression of a reader sharing the author’s world view.16 Whilst the protreptic and apologetic characteristics of the Confessions are frequently referred to in scholarship, I believe that there are a number of its features that might be illuminated by an examination in light of some of the features ascribed to paraenetic in current scholarship but, apart from a few remarks here and there, this falls outside the scope of this article.

The Confessions as a protreptic: An anomaly

I come back to the convention of calling the Confessions a protreptic, which in fact represents an anomaly. Apart from scholars who occupy themselves specifically with the Manichaean elements in the Confessions, the widespread consensus is that the work is written first and foremost for Catholic insiders, the servi dei or spiritales.17 This means that many scholars who call the Confessions a protreptic clearly do not have in mind the current mainstream definition of protreptic as aimed at converting outsiders. As indicated for example by the title of Mayer’s 1998 article, ‘Die Confessiones des Aurelius Augustinus: Eine philosophisch-theologische Werbeschrift (Protreptikos) für Christliche Spiritualität’, the tacit assumption seems to be that a protreptic is a general spiritual exhortation that may urge the most advanced members of the group to spiritual progress (probably not consciously excluding the possibility of and at the same time urging conversion of those who have not yet converted). Whilst this understanding of protreptic is in accordance with Swancutt’s and my own findings on protreptic and paraenetic, it is seriously at odds with current mainstream definitions. The view that a clear-cut dichotomous division between protreptic and paraenetic is not valid – and by implication that the way the term protreptic is currently used to refer to the Confessions is valid – is still a decidedly marginal view in scholarship. Yet, most scholars who call the Confessions a protreptic take for granted that the term is self-explanatory and make no effort to define it.18

Communicative purpose and audience in Books 1−3

Let me proceed to a re-investigation of the intended audience and purpose of the first 3 books of the Confessions in light of what I have said so far about the purpose and audience associated with protreptic in mainstream scholarship, as well as in light of my reservations about this. Here I investigate the narrative of Books 1–3 in order to determine (1) to what extent these books may be read as an exhortation to the still committed Manichee to convert to Catholic Christianity (protreptic in the restricted sense of mainstream definitions); (2) whether they are perhaps more suitable to exhort a recently converted Manichee or one on the point of conversion (the liminal Manichee) to persevere on the chosen course or to take the final step (this would be more akin to what is currently understood as paraenetic); or (3) whether these books rather seem designed to exhort committed Catholics, including the servi dei, to spiritual progress, or perhaps simply to inspire them through the sharing of experiences with like-minded readers (also closer to paraenetic purpose in terms of present definitions). I leave aside for now apologetic purpose, which is generally assumed to be part of the narrator’s intention in the Confessions.

In my previous work on the Confessions as a protreptic text, I pointed out those instances where conversion seems to be implied, those elements in the work that point to a protreptic purpose in the sense of mainstream definitions. Amongst other things, I read everything that could be construed as an argument against Manichaean positions as intended to convince a still committed Manichee of the error of his or her thinking, and to convert. The evident fact that the narrative is also written for insiders and the clear mention of the fratres or fellow Catholics as readers (in Books 10 and 11)19 I explained by saying that the work is a protreptic-paraenetic exhortation with protreptic dominating in the conversion story of Books 1–8 and Book 9, paraenetic dominating from Book 10 onwards and protreptic again towards the end of the work where Manichaean concerns seem to resurface more strongly (Kotzé 2004).

On the basis of my reading of a number of passages in the Confessions, and above all of Augustine’s meditation on Psalm 4 in Confessions 9, I would have liked to maintain that the Confessions as a whole aims – to a very important extent – to break through to Manichaean readers in general and to convince them of the fatal errors of Manichaean views about the nature of man, of God, or of evil. I am still convinced that Confessions 9.7–11 (the meditation on Ps 4) should be read as a highly effective protreptic to a broad spectrum of Manichaean readers.20 It also seems to me that the dramatic situation Augustine imagines and articulates in this section of Confessions 9 is to an important extent an apt description of the dramatic situation he, in fact, sets up for the Confessions as a whole. When Augustine wishes in Confessions 9 that the Manichees could have overheard him pouring out his deepest yearnings before God without him knowing that they were listening, this is very similar to what the Confessions formally represents: Augustine opening his heart before God to confess his sins as comprehensively as possible, sins of which the error of his thinking as a Manichaean just happens to form a substantial and integral part. This would imply that his stance is that of not consciously targeting a Manichaean audience and not in the first instance aiming at converting them or defeating them in a polemic fashion. Yet he is at the same time implicitly – in his expression of gratitude to God for being saved from such error – indicating his concern for their spiritual well-being.

But that protreptic purpose aimed at Manichaean readers is Augustine’s main concern is difficult to demonstrate on the grounds of a reading of Books 1–3. The problem is that the sincerity and the urgency of Augustine’s concern expressed in Confessions 9, the compassion for his erstwhile co-religionists that is palpable there, is roundly absent in the scathing account of his encounter with Manichaeism in Confessions 3 and also not at all clearly discernible in the narrative of Augustine’s early years in the opening books. In the next part of this article I echo the modus operandi of Johannes van Oort’s 1997 article in taking the anti-Manichaean passage in Book 3 as my point of departure before I proceed to the analysis of the preceding books of the Confessions.

The passage in Confessions 3 is a vivid representation of the narrator’s inner state of intense hunger and thirst for truth and of his experience of failure and utter emptiness whilst trying to satisfy this desire through what Manichaeism had to offer.21 I have to agree with critics who argue that it is hard to imagine committed Manichees calmly reading this, swallowing the insult of being called homines superbe delirantes carnales nimis et loquaces [men delirious with arrogance, overly focussed on the flesh and full of empty words] and continuing to read the rest of the Confessions, much less being persuaded to convert by such a harsh attack coupled with the blanket condemnation of Manichaean teaching in words and phrases like phantasmata, falsa or figmenta inania or the cutting sarcasm further on in this passage (in the description of the weeping fig tree in 3.10.18). The clear separation between the condemnation of erroneous Manichaean teaching and the expression of urgent concern for the Manichee as a person that is discernible in Confessions 9, is absent here.22

Yet, a large portion at the centre of this passage on Augustine’s sojourn as a Manichee is dedicated to a careful refutation of Manichaean objections to the Old Testament, as I argued in an earlier article (Kotzé 2008). This kind of rhetoric could conceivably have relevance for Catholic readers in general or the servi dei, as an example of how criticism of the Old Testament may be demolished or as a confirmation of their views in this regard. But to my mind it makes more sense to think of the type of content and the amount of space dedicated to it as directed to a Manichaean audience,23 and specifically the kind of liminal Manichee described before: one who may not have been repelled by Augustine’s impressionistic condemnation of Manichaeism in 3.6.10 but who may rather have identified with his urgent quest for truth, one who may still have been vulnerable to censure of the Old Testament or not yet persuaded that such criticism could be refuted.

I now turn to Books 1 and 2 in order to consider which types of readers the autobiographical narrative here may be best designed to reach. One of my arguments in the past for emphasising Augustine’s continuous concern with Manichaean issues in the Confessions was that material that we judge may have had special resonance with Manichaean readers occurs at pivotal points in the work: in the programmatic opening chapters, in the middle of the Confessions (the meditation on Ps 4 in Book 9), and right at the end of the work.24 Here I attempt to give a very concise overview of the narrative preceding the anti-Manichaean passage in Book 3 in order to determine whether protreptic purpose in the narrow or the broader sense can be discerned here.

Johannes van Oort (1997) has pointed out a number of features echoing Manichaean discourse in the opening paragraphs of the Confessions (1.1.1–1.5.6) but also the fact that from the start these terms and issues are imbedded in a context that illuminates the contrast between Manichaean thinking and Catholic ideas. Yet, in these opening books I do not detect the polemical or argumentative tone, or the sense of urgency to refute Manichaean positions or persuade Manichees that I do in Confessions 3.6.10–3.10.18 or in Book 9.4.7–11. The possibility has to be considered that in Books 1 and 2 the occurrence of Manichaean terminology is incidental and that it simply emerges from the subconscious where it has shaped Augustine’s thought on God and man for more than two decades.

At this point it is necessary to emphasise that, as a South African in the 21st century, growing up as a product of a (albeit protestant) Christian tradition and shaped by all that I have read about the Confessions (even including O’Donnell’s sceptical reading against the grain of the text [2005]) I find it difficult not to read the beautiful lyrical prose of the opening of the work as a smoothly flowing hymn, a prayer that lifts the emotions and the intellect of the believer towards God,25 not to join Augustine in his wonder at the greatness of God and God’s concern for insignificant human beings. To put it more scientifically: it can be argued that the prayer stance and the inclusion of the reader already in the opening paragraph, in the intimate fecisti nos ad te et inquietum est cor nostrum donec requiescat in te [You made us for Yourself and our heart is restless until it may rest in You], strongly indicates that the implied reader is regarded – at least on the surface – as an insider, a fellow-Catholic, one sharing the narrator’s world view. This is of course perfectly in tune with views that the Confessions is written mainly for like-minded Catholics.

Further, whilst the opening paragraph contains allusions to Romans and Matthew 7:7 that had special significance for Manichaean readers, the majority of the biblical quotations in the rest of the prologue are from the Old Testament, which points to an audience who would most likely recognise the authority of such quotes: Catholic readers. Of course, as in the case of the Manichaean contents in the Confessions, one could argue that the biblical phraseology simply surfaces from Augustine’s subconscious where his thought has been formed by amongst other things, his reading of Old Testament texts, or even that many such phrases may have become a standard part of everyday Christian language by then.

Yet I do not think that either the Manichaean or the Old Testament allusions are there by accident. Because allusion is such an integral part of ancient literary composition, and because Augustine quotes so systematically and purposefully for example, from Romans or from the Psalms throughout the Confessions,26 I think we can assume that these allusions are also deliberate and attuned to an audience who would be able to recognise both the source of the allusions and their implications in the new context.27

Committed Manicheans would probably not have found the opening paragraphs smooth contemplative reading, because of the constant – for them presumably jarring – juxtaposing of what must have seemed familiar with the radically different interpretation of man, his place in the universe and the nature of God and creation implicit in the prologue and subsequent narrative.28 This would quite probably have elicited in such a reader the hostile reaction Courcelle ([1950] 1968:236) speaks of or the irritation that Van Oort (1997:242) postulates. In light of my earlier interpretation of the meditation on Psalm 4 in Book 9, I find it hard to imagine that Augustine’s principal motivation for including the allusions to Manichaean discourse here would have been simply to spark a hostile reaction. In fact, for me the audience that forms the best match for the kind of contemplative prayer interspersed with words and phrases that would have been familiar to – as well as scriptural allusions that would have carried authority for – Manichaean readers in general, is one of Manichaean apostates, the liminal Manichees, who would not be repelled by the contra-Manichaean positions but perhaps reassured or touched in a special way by the familiar elements in the text.

In 1.6.7 we have the start of the narrative of Augustine’s life, perhaps better described as the effort to confess as comprehensively as possible his sins and God’s accompanying misericordia.29 This is also more aptly designated the beginning of Augustine’s conversion story, because it proceeds only up to the conversion, thence my arguments (Kotzé 2004) that the telling of the conversion story offers an example for the conversion of the reader, that is, that it has a protreptic purpose. Although there are a few places in Book 1 where Augustine’s awareness of the human reader and his fate surfaces,30 or where he prays for the salvation of those already calling to God as well as those not yet doing so,31 the existence of this reader is mostly ignored, primarily through the stance of addressing God alone. Yet, in many instances in the rest of Book 1 Augustine’s fascination with human beings and his ability to evoke universal human experience in vivid prose images that readers easily identify with (then as now) take over, a fact that seems to point towards a wide and mixed audience. The intensity of protreptic urging conversion or spiritual growth is absent. Whilst the descriptions of childhood are underpinned by an anthropology in agreement with the Catholic view of God, creation and man and in conflict with Manichaean views, there is no strong indication of an urgency to change the reader’s way of life or to induce conversion from Manichaeism to Catholicism. The narrative in Book 1 serves as vehicle for the criticism of classical education and is general and broad ranging, perhaps, amongst other things, an exhortation to Catholics to be less compromising about the subject matter used as the basis for education (something closer to paraenetic again).

Of the 18 paragraphs that constitute Book 2 the first eight are dedicated to a continuance of the rather general and often more impressionistic than autobiographically specific confession of the sins of adolescence (of course with some exceptions where specific events in Augustine’s life are named).32 However, whilst Book 1 does contain regular references to sin and punishment, the opening of Book 2 seems to announce a more pronounced focus on sin right at the outset: recordari volo transactas foeditates meas [I want to recall my filthy behaviour]. The intensity of the confession of sin is also increased already in the opening paragraphs,33 making the first section of the book a good preparation for the microscopic analysis of the pear theft in the last section; both parts of Book 2 offer Augustine the opportunity to move from an autobiographical narration of sin to a ‘theoretical understanding of the nature of sin in general’ (McDonald 2006:46).

The question that interests me here remains: what kind of audience is presupposed by such an uncompromising reflection on sin and what does it aim to communicate to them? This is the context where the much commented upon first explicit mention of the Confessionss human audience is introduced (in 2.3.5) with cui narro haec? [whom do I tell this?].34 Somewhat unexpectedly the audience is here defined in very broad terms, as whichever part of the genus humanum might read the Confessions. This is in line with my perception of the audience implied by the narrative of Augustine’s early life in Book 1 but not with the kind of audience I would associate with the obsessive insistence on sin that characterises Book 2 (I return to this point below). Further, the mention of the audience is coupled with a formulation of communicative purpose; but this too does not throw much light on the kind of audience interested in this complicated examination of the nature of sin. The narrator’s question and answer concerning the addressees of his narrative is followed by: et ut quid hoc? ut ego et quisquis haec legit cogitemus de quam profundo clamandum sit ad te [And why do I do this? So that I and whoever reads this may consider the depths from which we have to call to you]. The reader is once again induced to speak with the narrator. Here protreptic purpose does seem to be implied, both in the narrow sense (an exhortation to the not yet converted to call upon God, irrespective of how sinful they may be) and in the broader sense (a reminder to the already converted of their own inherently sinful nature). However, the almost compulsive insistence on the nature of sin and human motivation also points to the possibility that Augustine may be arguing some very specific technical points with a layer of the audience who can appreciate the philosophical and theological implications at stake: the inner circle of advanced Catholics, but again, perhaps more specifically a reader with a Manichaean background. Scholars generally appreciate that the narration of the pear theft has special significance for the anti-Manichaean debate,35 but BeDuhn’s observation (2010:36–41) that the confession of the pear theft closely resembles a very specific Manichaean confession ritual at the occasion of initiation into the sect, has important implications for our reading of Book 2 and in fact, the work as a whole. It exponentially increases the importance we have to allot to the Confessions’ indebtedness to Manichaean categories of thought. Earlier meticulous studies of the different shades of the meaning of forms of confessio/confiteri as a basis for understanding what the work was about did not account for the fact that the word and its derivatives must also have accrued some very specific connotations from its use in a Manichaean environment and what the implications of this for the choice of title might be.36 In light of this another question has to be asked: to what extent can what is regarded as one of the most innovative aspects of the Confessions – the introspection that so markedly distinguishes it from its antecedents and that has earned it the epithet of ‘first modern autobiography’ – also be ascribed directly to the influence of Manichaean teaching and practice on Augustine’s thinking?

But more importantly, everything said above has implications for what is regarded as the intended audience of the Confessions. As in the case of the arguments in Book 3 concerning the defence of the Old Testament, it would make most sense to regard this meticulous unravelling of the nature of sin, within a framework recognisable to Manichaean readers (the confession ritual), as well designed to communicate with such readers, especially those who – like the liminal Manichee – have to be reassured that Augustine’s version of the Catholic concept of sin makes more sense than the Manichaean version.

The first nine paragraphs of Book 3 that precede the story of Augustine’s ‘falling in’ with the Manichees in 3.6.10–3.10.18 continue the narrative of Book 2 where it was interrupted to dwell on the pear theft. The first six paragraphs on the sins of his student years in Carthage (before the narration of the dramatic turning point occasioned by the reading of Cicero’s Hortensius) in their focus on sexual sin and central aspects of classical culture (the theatre and rhetorical education) are in many respects similar in narrative strategy and tone to the preceding narrative in Book 1 and the beginning of Book 2.37 Once again, the prayer stance is mostly kept alive only tenuously whilst the type of audience presupposed seems to be a broad (probably Christian) late ancient audience, that is, not excluding the spiritales, nor less advanced Catholic readers, Manichaean readers, or even pagan readers, but also not specifically targeting any groups above others.

A careful reading of Book 4 of the Confessions will surely throw more and different light on the perspectives proposed in this article, as far as the importance of Manichaean readers as part of the intended readership of the Confessions is concerned and most probably in terms of how the narrative aims to communicate with such readers. But this has to be left for a future investigation.


The reading of the first three books of the Confessions offered here leads to a number of observations about their communicative purpose and intended audience: (1) the presence in these books of terms and quotes or allusions that may have had special connotations for Manichaean readers points to the type of reader who would be receptive to such elements in the narrative; (2) the strong presence of Old Testament quotes in Book 1, and also to a lesser degree in the rest of the passages under consideration here, presupposes a Catholic readership able to recognise the allusion and willing to accept the authority of such material; (3) the way in which the work opens (through the prayer stance and the inclusion of the reader in the first person plural pronouns and verbs), makes the reader speak with the narrator and points to an audience in agreement with the narrator’s world view; (4) the understanding of God, man and creation underlying the narrative in the prologue and the autobiographical narration in the first books reflects a Catholic world view but is at odds with Manichaean thinking.

Thus it emerges that protreptic purpose in the sense of mainstream definitions, that is, the purpose to convert the not-yet converted, does not seem a strong concern in the first three books and the presence of terms that may have had special connotations for Manichaean readers or the more overtly anti-Manichaean section do not point to an effort to convert a committed Manichee. Protreptic purpose in the broader sense (including purposes ascribed to paraenetic in mainstream definitions), id est, a more general exhortation to insiders to praise God with Augustine, is present strongly in the prologue but sometimes recedes into the background in the rest of Books 1−3.

All of this argues for a relatively broad and mixed audience for the Confessions (or at least for this section of the work), but especially highlights the prominence of one type of potential reader. There is a strong possibility that the reader most receptive to both the insider stance and allusion to the Old Testament on the one hand and the Manichaean material on the other would be the liminal Manichaean reader postulated in this article.


Competing interests
The author declares that she has no financial or personal relationship(s) which may have inappropriately influenced her in writing this article.


Asher, L., 1998, ‘The dangerous fruit of Augustine’s Confessions’, Journal of the American Academy of Religion 66, 227–255.

Augustinus, 1981, ‘Retractationum libri II Cap.XXXII’, in M. Skutella (ed.), S. Aureli Augustini Confessionum Libri XIII, n.p., Teubner, Stuttgart.

Augustinus, 1992, ‘Confessionum libri XIII’, in J.J. O’Donnell (ed.), Augustine: Confessions: Introduction and Text, vol. 1, pp. 3–205, Clarendon Press, Oxford.

BeDuhn, J.D., 2010, Augustine’s Manichaean Dilemma, I, Conversion and Apostasy, 373–388 C.E., University of Pennsylvania Press, Philadelphia.

Brown, P., 2000, Augustine of Hippo: A biography, University of California Press, Berkeley/Los Angeles.

Courcelle, P., [1950] 1968, Recherches sur les Confessions de saint Augustin, 2nd edn., De Boccard, Paris.

Engberg, J., 2009a, “‘From among you are we. Made, not born are Christians’: Apologists’ accounts of conversion before 310 AD’, in J. Ulrich, A-C. Jacobsen & M. Kahlos (eds.), Continuity and discontinuity in early Christian apologetics, pp. 49–77, Peter Lang, Frankfurt am Main. (Early Christianity in the Context of Antiquity 5).

Engberg, J., 2009b, ‘Truth begs no favours – Martyr-literature and apologetics’, in A-C. Jacobsen, J. Ulrich & D. Brakke (eds.), Critique and apologetics: Jews, Christians and pagans in antiquity pp. 177–208, Peter Lang, Frankfurt am Main. (Early Christianity in the Context of Antiquity 4).

Feldmann, E., 1994, ‘Confessiones’, in C. Mayer (ed.), Augustinus-Lexikon, Schwabe, Basel, vol. 1, pp. 1134–1193.

Ferguson, E., 1993, Backgrounds of early Christianity, 2nd edn., Eerdmans, Grand Rapids.

Hübner, W., 2007, ‘Klassische lateinische Literatur und Rhetorik’, in V.H. Drecoll (ed.), Augustin Handbuch, pp. 49–60, Mohr Siebeck, Tübingen.

Jacobsen, A-C., 2009a, ‘Apologetics and apologies – Some definitions’, in J. Ulrich, A-C. Jacobsen & M. Kahlos (eds.), Continuity and discontinuity in early Christian apologetics, pp. 5–21, Peter Lang, Frankfurt am Main. (Early Christianity in the Context of Antiquity 5).

Jacobsen, A-C., 2009b, ‘Main topics in early Christian apologetics’, in A.-C. Jacobsen, J. Ulrich & D. Brakke (eds.), Critique and apologetics: Jews, Christians and pagans in antiquity, pp. 85–110, Peter Lang, Frankfurt am Main. (Early Christianity in the Context of Antiquity 4).

Jordan, M.D., 1986, ‘Ancient philosophic protreptic and the problem of persuasive genres’, Rhetorica 4, 309–333.

Knauer, G.N., 1955, Psalmenzitate in Augustins Konfessionen, Vandenhoeck & Ruprecht, Göttingen. PMCid:1532751

Knauer, G.N., 1957, ‘Peregrinatio animae (zur Frage der Einheit der augustinischen Konfessionen)’, Hermes 85, 216–248.

Kotzé, A., 2001, ‘Reading Psalm 4 to the Manichaeans’, Vigiliae Christianae 55, 119–136.

Kotzé, A., 2004, Augustine’s Confessions: Communicative Purpose and Audience, Brill, Leiden.

Kotzé, A., 2006, ‘Augustine, Paul, and the Manichaeans’, in C. Breytenbach, J.C. Thom & J. Punt (eds.), The New Testament interpreted: Essays in honour of Bernard Lategan, pp. 163–174, Brill, Leiden & Boston. (Supplements to Novum Testamentum 124).

Kotzé, A., 2008, ‘The “Anti-Manichaean” passage in Confessions 3 and its “Manichaean Audience”’, Vigiliae Christianae 62, 187–200.

Kotzé, A., 2011, ‘Protreptic, paraenetic and Augustine’s Confessions’, in J.A. van den Berg, A. Kotzé, T. Nicklas & M. Scopello (eds.), In search for truth: Augustine, Manichaeism, and other Gnosticism: Studies for Johannes van Oort at sixty, pp. 3–23, Brill, Leiden.

Lancel, S., 2002, St Augustine, transl. A. Nevill, SCM Press, London.

Lössl, J., 2011, ‘Augustine’s Confessions as a Consolation of Philosophy’, in J.A. van den Berg, A. Kotzé, T. Nicklas & M. Scopello (eds.), In search for truth: Augustine, Manichaeism, and other Gnosticism: Studies for Johannes van Oort at sixty, pp. 47–73, Brill, Leiden.

Mayer, C.P., 1998, ‘Die Confessiones des Aurelius Augustinus: Eine philosophisch-theologische Werbeschrift (Protreptikos) für Christliche Spiritualität’, Theologie und Glaube 88, 285–303.

McDonald, S., 2006, ‘Petit Larceny, the beginning of all sin: Augustine’s theft of the pears’, in W.E. Mann (ed.), Augustine’s Confessions: Critical essays, pp. 45–69, Rowman & Littlefield, Lanham.

O’Donnell, J.J., 2005, Augustine: A new biography, Harper Collins, New York.

Perdue, L.G. & Gammie, J.G. (eds.), 1990, ‘Paraenesis: Act and from’, Scholars Press, Atlanta. (Semeia 50).

Peterson, A.G., 2009, ‘The diversity of apologetics: From genre to a mode of thinking’, in A-C. Jacobsen, J. Ulrich & D. Brakke (eds.), Critique and apologetics: Jews, Christians and pagans in antiquity, pp. 15–41, Peter Lang, Frankfurt am Main. (Early Christianity in the Context of Antiquity 4).

Popkes, W., 1996, Paränese und Neues Testament, Katholisches Bibelwerk, Stuttgart.

Rankin, D., 2006, From Clement to Origen: The social and historical context of the Church Fathers, Ashgate, Aldershot.

Slings, S.R., 1994, Plato’s Apology of Socrates: A literary and philosophical study with a running commentary, Brill, Leiden/New York.

Slings, S.R., 1995, ‘Protreptic in ancient theories of philosophical literature’, in J.G.J. Abbenes, S.R. Slings & I. Sluiter (eds.), Greek literary theory after Aristotle: A collection of papers in honour of D.M. Schenkeveld, pp. 173–192, VU University Press, Amsterdam.

Starr, J.M., 2004, ‘Was paraenesis for beginners?’ in J. Starr & T. Engberg-Pedersen, Early Christian paraenesis in context, pp. 73–112, De Gruyter, Berlin.

Starr, J. & Engberg-Pedersen, T. (eds.), 2004, Early Christian paraenesis in context, De Gruyter, Berlin/ New York.

Stowers, S.K., 1986, Letter writing in Greco-Roman antiquity, Westminster Press, Philadelphia.

Swancutt, D.M., 2004, ‘Parainesis in light of protrepsis: Troubling the typical dichotomy’, in J. Starr & T. Engberg-Pedersen, Early Christian paraenesis in context, pp. 113–153, De Gruyter, Berlin/New York.

Van der Meeren, S., 2002, ‘Le protreptique en philosophie: essai de définition d’un genre’, Revue des Etudes Grecques 115, 591–621.

Van Oort, J., 1997, ‘Manichaeism and anti-Manichaeism in Augustine’s Confessiones’, in L. Cirillo & A. van Tongerloo (eds.), Manichaean Studies III, Atti del terzo congresso internazionale di studi ‘Manichaeismo e Oriente Cristiano Antico’, pp. 235–247, Brepols, Turnhout.

Van Oort, J., 2002, Augustinus’ Confessiones: Gnostische en christelijke spiritualiteit in een diepzinnig dokument, Brepols, Turnhout.

Von Stockhausen, A., 2006, ‘Ein “neues Lied”? Der Protreptikos des Klemens von Alexandrien’, in C. Schubert & A. von Stockhausen (eds.), Ad veram religionem reformare. Frühchristliche Apologetik zwischen Anspruch und Wirklichkeit, pp. 75–96, Universitätsbund Erlangen-Nürnberg, Erlangen.


1.See Kotzé (2004) for earlier arguments about the Confessions as a protreptic to the Manichees.

2.Part of the findings of this investigation can be found in Kotzé (2011).

3.Many scholars use the term apologetic as roughly synonymous with the term protreptic, but in specialised research on apologetic literature (where the issue of definition is still debated) there seems to be a strong argument for seeing defence against real or imagined accusations as a core characteristic of apology. For the latter, see Jacobsen (2009a:5–13).

4.A refutation not only of the explicit criticisms of Manichaeism in the Confessions but also of Catholic positions that diametrically oppose Manichaean views on issues like creation, the nature of God or the origin of evil.

5.Scholars often remark on the fact that throughout the Confessions Augustine consciously associates the sin of (misdirected) curiositas with the Manichaeans.

6.See the incisive interpretation of the historical evidence by BeDuhn (2010:136–144, 196).

7.Interestingly, this work is frequently discussed as an example of apology without any accompanying attempt to explain the title Protreptikos or the distinction between protreptic and apologetic. See for example in the list of Apologetic works in Engberg (2009a:52).

8.Seeing the Confessions as directed at recently converted Manichees would also be in line with my arguments that the work displays a number of characteristics associated with paraenetic in mainstream definitions and the convictions of many scholars that parainesis was meant mainly for beginners. See Starr (2004:73–111) for a discussion of such views.

9.I quote a few definitions by some of the influential exponents of these views: Stowers (1986:92): ‘I will use protreptic in reference to hortatory literature that calls the audience to a new and different way of life, and paraenesis for advice and exhortation to continue in a certain way of life’; Ferguson (1993:302): ‘Paraenesis is a broader term [than protrepsis] for moral exhortation to follow a given course of action or to abstain from a contrary behavior. It thus consisted of encouragement and dissuasion. Rules of conduct are prominent. Paraenesis presupposed some positive relationship between the parties’; the definition arrived at by a group of scholars at a conference in Oslo (Starr & Engberg-Pedersen 2004:4): ‘Paraenesis is [a] concise, benevolent injunction that reminds of moral practices to be pursued or avoided, expresses or implies a shared worldview, and does not anticipate disagreement.’ That these are still widely regarded as the standard definitions, is illustrated, for example by Rankin’s confident assumptions (2006:8): ‘If we identify apologetic as that form of discourse which seeks to explicate or to articulate a given position – to either believers or non-believers – without any intention, explicit or implicit, to seek the conversion of its addressees, and protreptic as that which seeks explicitly to convert by exhortation non-believers (paraenetics [sic] is exhortation to believers), then we will not be far from the truth’ (author’s own emphasis).

10.Diana Swancutt argued (already in 2004) that the terms protreptic and paraenetic in fact refer to similar types of exhortation whilst only the affiliation of the philosopher writing it determined whether it was called protreptic or paraenetic, but her views do not seem to have impacted on scholarship yet.

11.Though scholars almost invariably refer to the interchangeable use of the terms they persist in using the words protreptic and paraenetic as dichotomous technical terms denoting two clearly distinguishable types of literature, I argue that the interchangeable use of the terms reflect a practice where the functions and audiences ascribed to protreptic or paraenetic respectively are so frequently intertwined that the dichotomous use of the terms may hinder rather than aid insight into exhortative literature.

12.The genre does derive originally from real forensic speeches (Petersen 2009:36–37). For the prominence of accusations in apologetic texts, see also Engberg (2009b) (a comparison of the accusations treated in apologetics and in martyr-narratives) or Jacobsen (2009b:85).

13.See for example Engberg (2009b:177, including the long footnote 3; 2009a:62–65).

14.See for example Von Stockhausen (2006:87–90) on Clement of Alexandria’s Protreptikos and Jacobsen (2009a:14–17) on Early Christian Apology.

15.See Slings’s arguments (1994:13–16) for the protreptic nature of the Apology.

16.For example Augustinus, Confessionum libri XIII 1.1.1, fecisti nos ad te et inquietum est cor nostrum [you made us for yourself and our heart is restless]; 1.3.3 et cum effunderis super nos, non tu iaces, sed erigis nos, nec tu dissiparis, sed colligis nos [and when you are poured out over us you do not flow away, but you lift us up and you are not spilt but you gather us]; 1.10.16 et libera nos iam invocantes te, libera etiam eos qui nondum te invocant [set us free who already call on you, set free also those who do not yet call on you]; 1.11.17 vita aeterna promissa nobis per humilitatem domini dei nostri descendentis ad superbiam nostrum [eternal life promised to us through the humility of our Lord God who descends to our pride].

17.Brown’s view (2000:153) has not, to my knowledge, been challenged and is still valid: the Confessions is to a very important extent addressed to Augustine’s fellow Catholics.

18.See for example Feldmann (1994) or Lancel (2002:210).

19.See for example Augustinus, Confessionum libri XIII 10.4.5 (fraternus animus, fraternus ille, fraternis cordibus), 10.4.6 (servi tui, fratres mei) or 11.2.3 (non mihi soli aestuat, sed usui vult esse fraternae caritati).

20.See my arguments in an earlier article, Kotzé (2001).

21.In light of BeDuhn’s arguments (2010:36–37) about Augustine’s impulsive joining of the Manichees without knowing too much about their teachings initially (in fact quite well expressed by incidi) and his arguments about the relatively subordinate place that Augustine’s religious identity may have occupied in his life at that stage, I think that we can assume that Augustine is telescoping events in order to create the dramatic narrative here. This is quite probably a re-interpretation that reflects more of his later insights than of his initial experience in joining the Manichees.

22.This narrative would, of course, conceivably be a highly effective apology against accusations of crypto-Manichaeism, accusations never voiced in the Confessions but probably at the back of Augustine’s mind and generally accepted by scholars as part of the motivation for writing the Confessions.

23.It constitutes three of the five pages dedicated to the Manichaean section in Book 3.

24.All these Manichaean echoes and the presence of such material in the opening and closing sections of the Confessions have been pointed out by Johannes van Oort (2002). See also my article (Kotzé 2006) on the Manichaean subtext for Confession 13.25.38–13.27.42.

25.Augustinus, Retractationum libri II 2.32: Confessionum mearum libri tredecim et de malis et de bonis meis deum laudant iustum et bonum atque in eum excitant humanum intellectum et affectum [the thirteen books of my Confessions praise the just and good God both for the bad and for the good in me and it they impel the human intellect and emotions to move towards Him].

26.See for example Knauer (1955, 1957).

27.The reaction of a liminal Manichaean to Old Testament quotes is more difficult to gauge: whilst they may have come to accept the authority of the Old Testament it is not certain that they would have recognised the source of the quotes or their meaning in the new context.

28.For example the references to homo portio creaturae tuae, or the implied insistence that God cannot be confined to physical space in paragraphs 2 and 3.

29.Nothing in the opening lines or the title of the work, in fact, prepares the reader to expect a comprehensive life story. The only, implicit, motivation for giving details about Augustine’s early life seems to be the desire to confess both his own sins and God’s grace.

30.In 1.6.7 misericordia tua est, non homo, inrisor meus, cui loquor [it is your mercy that I address, not man who would mock me]; or in 1.6.10 quid ad me, si quis non intellegat? [what does it matter to me if someone does not understand?]. libera nos iam invocantes te, libera etiam eos qui nondum te invocant, ut invocent te et liberes eos [set us free who already call on you, set free also those who do not yet call on you so that that they may call on you and you may set them free].

32.For example the reference to his father’s efforts to find money for his future education in 2.3.5.

33.This is affected by an accumulation of words pertaining to dirt and darkness (umbrosis amoribus, contabuit, computrui, nebulae de limosa concupiscentia, obnubilabant, obfuscabant, caligine libidinis, gurgite flagitiorum).

34.See Asher’s interesting arguments (1998:231–232, 235) about the unexpectedness and the implications of the first mention of the human audience at this point in the narrative. Yet, it does seem logical to me that Augustine feels he has to motivate the negative picture of humankind he draws here by explaining that this illustrates the depths from which humans have to call to God.

35.See for example MacDonald (2006:55): ‘That account of sin provides Augustine the conceptual leverage he needs to overturn his Manichaean convictions.’

36.For an overview of studies investigating the connotations associated with title Confessiones see Feldmann (1994:1144–1146).

37.Book 3 is, however, substantially different from Book 2 in the sense that it creates the impression of faster forward movement: the change of location to Carthage in 3.1.1 (veni Carthaginem), the initial impetus for the quest for truth resulting from the reading of the Hortensius (in 3.4.7) and the move to the Manichees (in 3.6.10–3.10.18). It also contains the first formulation of the need to increase the tempo of the narrative (nam et multa praetereo, propter quod propero ad ea quae me magis urguent confiteri tibi in 3.12.21) as well as a foreshadowing of Augustine’s eventual conversion to Catholicism in the narration of Monnica’s dream (in 3.11.19–3.12.21).


Crossref Citations

1. Church, mission and ethics. Being church with integrity
Wim Dreyer
HTS Teologiese Studies / Theological Studies  vol: 72  issue: 1  year: 2016  
doi: 10.4102/hts.v72i1.3163