by Anthony Freeman
Anthony Freeman is an honorary assistant priest at Crediton Parish Church and the managing editor of the Journal of Consciousness Studies. This article originally appeared in Modern Believing.
Three cheers for Cardinal Ratzinger! He’s quite wrong, of course, but being wrong with such glorious precision and clarity throws more light on a subject than any amount of fudging and fuddling and getting-it-half-right. I refer to his ‘Note on the Expression Sister Churches’. This short document sets out the very limited ‘proper’ sense in which the term may be used, ending with a reminder ‘that the expression sister churches . . . may only be used for those ecclesial communities that have preserved a valid episcopate and eucharist’ (section 12). That sting in the tail, felt sharply by the churches of the Anglican communion, was latched on to by the press and others in Britain. It implied a rebuke not only of certain catholic ecumenists today, but of Pope Paul VI, who appeared to use the term in 1966 at the time of his meeting with Archbishop Michael Ramsey in a way now at odds with official usage.
This paper will argue that the essential problem highlighted by Cardinal Ratzinger’s ‘note’ goes much deeper than has yet been appreciated. It concerns the whole way we think about the concept ‘church’, and requires of Anglicans no less than Roman Catholics a radically new perspective. If taken seriously, this fresh approach offers a way round many traditional stumbling blocks in theology concerning church and ministry. In particular, it will rescue Anglicans from what ought to be an acute ecumenical embarrassment to them; namely, that the Vatican’s attitude towards them (patronizing, historically untenable, and theologically unacceptable as it is) is exactly paralleled by the way Anglicans treat churches of the Reformed and other Protestant traditions. Further, a new understanding of the concept ‘church’ should also help to resolve the ambiguities surrounding women’s priestly and episcopal ministry. The first section of the paper considers recent work on ‘concepts’ in cognitive psychology. Subsequent sections will apply these insights to aspects of Anglican ecclesiology.
The classical theory of categories and concepts — which most people instinctively hold — assumes that every object or action possesses certain properties or features that define what it is. And for every category or concept there is a set of necessary and sufficient features that an item must possess if it is to fall under that concept or belong to that category. To give a slightly over-simplified example from physics: a substance is categorized as a solid if it has a fixed volume and a fixed shape; it is categorized as a liquid if it has a fixed volume but no fixed shape; it is categorized as a gas if it has neither a fixed volume nor a fixed shape. Yet even such a basic example as this raises uncertain cases: a quantity of fine dry sand, for instance, has no fixed shape, but is usually classified as a solid rather than a liquid, because each individual grain of sand has a fixed volume and a fixed shape.
Things get worse for the classical theory when we move to a more complex object, such as a bird. One obvious characteristic of birds is flying, yet some creatures classified as birds cannot fly — penguins, for instance, and ostriches — while other creatures that do fly — most insects, not to mention bats and ‘flying’ fish — do not fall into the category of bird. Flying is therefore neither a necessary nor a sufficient feature for the concept ‘bird’. So despite its being an overwhelming characteristic of most birds, the classical theory has no way of including flying in its definition.
In recent years this classical theory has been challenged, notably in work carried out by psychologist Eleanor Rosch and her colleagues during the 1970s. She began by investigating how people categorize colours. She found that her subjects, in addition to assigning each sample colour to a category, say ‘red’, would choose certain examples as ‘better’ or ‘more typical’ than others of that subject’s idea of the colour. For instance, the red of a fire engine was felt by most subjects to exemplify their concept of ‘red’ better than the red of red hair. Furthermore, with colour samples a long way from the ‘best’ example, subjects would prefer to assign a ‘degree of redness’ rather than designate the sample starkly as ‘red’ or ‘not red’. This may strike us as unsurprising, given the way that colours by their nature shade off into one another, red for instance moving through orange to yellow in one direction and through purple to blue in the other. Nonetheless it breaks one of the key conditions of classical theory, according to which categories have clear-cut boundaries: a colour either does qualify as red (in which case it is as fully red as any other example in the category) or it does not (in which case it is something else entirely, and is not red at all).
Rosch was struck by the possible wider significance of her results, and embarked on an extensive research programme to see whether ‘graded membership’ of categories applied in other cases as well. She found that it did. In all kinds of categories studied, subjects would happily rate items according to their ‘degree of membership’ of the category to which the items belonged (based on the subject’s own idea or image of that category).
The conclusion drawn was that in practice we do not normally define categories by a list of necessary and sufficient features. Rather, people characterize a given category (such as ‘bird’) by identifying certain members of the category as typical representatives, called prototypes by Rosch. In the case of birds, the robin or the sparrow might be taken as prototypes of the concept. Other members of the category are then judged more or less typical by comparison with these prototypes. A cuckoo (not building a nest) or a penguin (swimming but not flying) would in varying degrees be nonprototypical birds. In contrast to classical theory, the category or concept ‘bird’ is no longer to be thought of as a container, with each specimen either 100% in it or 100% out of it, but as an ideal model, to which any actual specimen will conform more or less fully.
This change from classical theory to prototypes is so drastic that we are instinctively resistant to it: But a cuckoo is a bird, we want to insist, and bat isn’t. And a penguin either is a bird or it isn’t — it can’t half be a bird! But it is precisely the idea of there being ‘degrees of birdness’ that the theory of prototypes is putting forward.
It is important to be clear what is not being described here. We are not talking about the probability of a given item’s being assigned to a particular category. We are saying that items in a category are not all equivalent; there are gradations of membership, and people will assert directly that one member of a category is a better example than another. Nor is this true only of hazy classifications like colour, or of complex biological ones like birds. It is equally the case with a precisely defined concept such as ‘odd number’. People who judge 8421 as unequivocally an odd number also judge 7 to be a better example of their idea of the category ‘odd number’. It is also important to note that these ideas are not just philosophical speculation. They are based on empirical evidence of the way ordinary people actually do handle concepts and categorize items.
The Lambeth Conference of 1888 adopted four articles as the basis on which to approach discussions of church reunion. They were taken and slightly amended from a statement adopted two years earlier by the House of Bishops of the Protestant Episcopal Church of the United States of America (as the Anglican church in America was then known) at a meeting in Chicago. Hence the somewhat quaint title — the Chicago-Lambeth Quadrilateral — by which the articles have been called ever since. As enumerated in Resolution 11 of the 1888 Lambeth Conference they are:
Whatever the precise intentions of their framers may have been, these articles have operated in practice as a set of non-negotiable elements that Anglicans regard as essential features of any reunited church. They are in effect the necessary and sufficient conditions that any proposed religious body must meet in order to fulfill the Anglican concept of ‘church’. In classical terms, any religious grouping not exhibiting all four features does not meet the qualifications and so falls outside the category; it is not a church. The Church of England and those bodies in full communion with it meet the conditions already; but if they were to enter into reunion and merge their identity into some new body, their guaranteed status as ‘church’ would be put at risk. It is therefore essential, on this approach, that any proposed reunited body should also exhibit these four essential features.
The awkwardness of this position is immediately apparent. Put at its simplest, the articles look rigged. It is not just the case that the Church of England and those bodies in full communion with it are, on this reckoning, true churches; it also looks as if they are the only bodies to warrant the title. The ‘churches’ of the reformation all fail under article four (they lack the historic episcopate) and the Roman Catholic ‘Church’ fails under articles one and two (it requires as necessary to salvation things found neither in the Scriptures nor the Creeds). The Eastern Orthodox Churches might qualify, but they might also fail under article two, since they deny the double procession of the Spirit, which forms part of the (Western) Nicene Creed. This last point highlights a further problem. Not only does the word ‘church’ lack an unequivocal meaning across the Christian traditions, but so does almost every other term that appears in the four articles of the quadrilateral, including ‘the Nicene Creed’.
Consider for example the first article. It assumes that there exists a category of texts that qualify for inclusion in the concept ‘Holy Scriptures of the Old and New Testament’, and that it is possible to determine whether any given text is scriptural or not, in the required sense. The difficulties with such an assumption are many and familiar. Not only have different Christian bodies at different times (and still at the present time) disagreed as to which books should be included in the Old and New Testaments, there is also a question as to whether the original text only or certain authorized translations also should count as ‘containing all things necessary for salvation’. If the former, then does this apply to the autograph original or to one or more ‘received’ texts? If the latter, which translations qualify? What is the status of textual variants? And so on.
My purpose in highlighting these problems is not to answer them but to ask why in practice they cause so little difficulty. How is it that intelligent and serious Christians can be aware of these questions, which have serious implications for the central documents of the faith, and for the most part ignore them? Why does the Church of England as a body ignore them? The answer, I suggest, lies in a gap between theory and practice. In theory there is a sharp and essential distinction between ‘Holy Scripture’ and ‘not Holy Scripture’. In theory it matters a very great deal what is and what is not in the Bible, and if anyone were to suggest in General Synod that some book be added to or deleted from the canon, the opposition would be absolute. Yet in practice no-one seems to mind that the boundaries between the biblical and non-biblical grow ever fuzzier. Quite apart from the kinds of uncertainties raised in the last paragraph, large swathes of the Bible are no longer read at public worship, while the oft-repeated non-biblical words of hymns and songs and prayers quietly fashion and build up the theology of those who use them. Even among those who have a competent working knowledge of the whole Bible, there is rarely a special importance assigned to a passage or an idea or a command solely because it is in the sacred text. The practice of creating a ‘canon within the canon’ is not the preserve of any one doctrinal party.
Another way of framing these observations is to say that in theory we accept a classical approach to the concept of Holy Scripture, but in practice we adopt a prototype one. To start with there is a graded structure to our concept of a biblical book. Genesis or Isaiah, for instance, stand a good chance of being named ‘best examples’ of the Old Testament, with maybe Luke or Romans for the New. Zephaniah and Jude, on the other hand, although unequivocally in the canon, would not come close to being seen as prototypical books. Similar considerations apply when we turn to the different biblical editions and versions, but it is less easy here to predict which of the cluster of items belonging, say, to the concept of the Old Testament would emerge as the prototype. Would an Old Testament scholar choose the unpointed Hebrew or the Massoretic text? Would a New Testament scholar go for the Septuagint? Would Roman Catholic scholars give priority to the Vulgate, paying heed to the special place assigned in their tradition to that version? Would an average Anglican who had done the Alpha course instinctively adopt the NIV as the norm? We don’t know and it’s not actually important. What is important, for the purposes of this paper, is that we recognize here an accurate description of how we do relate to the plethora of texts, versions and editions that together make up our experience of Holy Scripture. Perhaps the most important thing of all is that we acknowledge an element of contingency in the whole business. It is not the case that one text is objectively the true one and all the others are more or less defective. Holy Scripture has come together in various forms from diverse sources, developed in different ways, been corrupted and amended, translated and mistranslated, treasured and abused, and all these elements taken together contribute to the concept of it held by each of us. Each particular form of the text will have its own value and validity, and which is taken as the prototype will depend to some extent upon the circumstances of the selection and the use to which the text is going to be put. There simply is no such thing as a definitive text.
When we turn to what Cardinal Ratzinger coyly refers to as ‘ecclesial communities’, we find at the informal practical level a divergence between theory and practice prallel to that found in the case of scripture. It is now many years since my wife, as a young Englishwoman visiting a country that was 95% Roman Catholic, was asked in all innocence whether she was ‘Christian or Church of England’. The ecumenical movement, for all that its progress has been halting and often back-sliding, has led most of those who ‘profess and call themselves Christians’ to accept as brothers and sisters in Christ all others who profess likewise. As a consequence we implicitly accept the organizations they belong to as churches, even though they fail to meet the criteria of the Chicago-Lambeth articles. In line with this, many — let’s be optimistic and say most — Anglican clergy regard free-church ministers and Roman Catholic priests as holding a ministerial responsibility and authority parallel and equivalent to their own. We may accept as a matter of discipline that ministries are not exercised willy-nilly across ecclesiastical boundaries, but we do not — for most practical purposes — question that all concerned exercise a Christian ministry in a Christian congregation. And it is certainly true that all Church of England clergy have a closer theological kinship with some of their free-church or Roman Catholic neighbours than they do with some of their fellow-Anglicans.
Here, with the concepts of church and ministry, we again see the fuzzy-edged prototype approach working in practice, despite a stronger and more overt official commitment to the classical model than was the case with scripture. This suggests a way forward. When considering attitudes to the Bible, we saw that the classical approach has not been officially abandoned, but the major consequences of a shift to the prototype view have been accepted. That is to say, the central importance of scripture has not been compromised by acknowledging that (1) there exists a graduated status of texts around the core concept of scripture, rather than a sharp in/out classification; and (2) different people in different situations take different texts/versions/editions as their prototypical ‘best example’, with no particular text/version/edition having an intrinsically privileged status. There is no reason why open acceptance of the similar practical shift in the concepts of church and ministry should compromise their centrality either. We have become accustomed to think that any blurring of the edges, any compromise in the purity of the Anglican church and ministry, would be fatal. But if the Holy Scriptures themselves can survive such ambiguity — ambiguity that in some cases goes back two thousand years — there is no reason why church and ministry should not prove equally robust.
The potential benefits of such a shift are enormous. Ecumenically, the prototype approach to the concept of a church retains all the advantages of the traditional Anglican ‘branch’ paradigm, while removing some significant problems with that classical model. In particular, (1) it concentrates on present realities rather than pinning everything on dubious or unverifiable historical claims; (2) by not privileging any one prototype, it removes the need for any tradition to give up what is most precious to it, except perhaps a belief in its own absolute rightness. That is a cue for Cardinal Ratzinger. It is crystal clear from his ‘Note on the Expression Sister Churches’ and even more so in the document Dominus Jesus, issued over his signature shortly afterwards, that the Vatican will only seriously contemplate reunion on the basis of other Christian communities re-establishing communion with Rome. It is hard — to the point of being impossible — to see that ever ceasing to be the case, no matter how liberal a pope the future might bring. What could change — and at some stage is bound to change — are the conditions on which such a re-establishment of communion might be achieved. At that point the ball will be in the court of the other churches, not least the Anglicans, whose attitude to other churches so closely mimics Rome’s attitude to them.
Our best strategy in the meantime is surely to create a theological climate among the non-Roman churches (including our own) where at least they are not perpetuating their own barriers. The ecumenical history of the last century shows that church order — rather than differences over doctrine, biblical interpretation, ethics, or even liturgy — is consistently the ultimate stumbling block between churches. The other things are still controversial, but the controversies cut right across denominational lines and are not the major issue in ecumenical dialogue. The adoption of the prototype approach to the concept of a church could achieve that necessary climate where, for example, Anglicans did not feel bound to force episcopacy on a church whose tradition emphasized an egalitarian form of decision-making and leadership. This would be possible because episcopacy would no longer be a necessary condition of being a ‘real’ church. Instead it would be a characteristic of the prototype example as understood by many (but not all!) Anglicans. Anglicans would thus regard a church without the historic episcopate as non-prototypical, but not as a non-church.
We look finally at the relevance of this discussion to problems arising from the ordination of women as priests and bishops. Interest in priests and bishops may be somewhat limited outside church circles, but the concept of a person is of central importance, relating to issues in law, ethics, politics, etc., and that makes it a good point of entry to our topic. Philosopher Mark Johnson, criticizing the classical view that ‘person’ is an unambiguous term with a self-evident meaning, suggests instead that: ‘[T]he concept person is a radial category consisting of certain prototypical instances (e.g. sane adult white heterosexual males) surrounded by nonprototypical instances (e.g., females, nonwhites, children, senile elderly, mentally handicapped) and fading off into borderline cases (e.g. higher primates)’. In a note he adds: ‘The extent to which females are regarded as prototypical persons varies from culture to culture and from one historical period to another. But it is clear that in American and European culture women have not yet attained prototypical status, insofar as they have not been accorded rights and privileges on an equal basis with men.’
This observation helps to explain one aspect of the psychologically driven opposition to the ordination of women. The Catholic argument against women priests tends to be put in the form: A priest must be male in order to represent the male Christ (especially at the eucharist); the maleness of Christ was not just a chance occurrence (‘humans have to be either one gender or the other and Christ just happened to be a man’) but God’s deliberate choice because a man is a more complete representative of humanity (and therefore also of God in whose image humanity is made) than a woman. This argument is normally decried by supporters of women’s ordination as a throw-back to discredited Aristotelian/Thomist attitudes to human nature and gender difference. Perhaps this reaction is too glib. It is probably true that using a classical approach to the concept of person, where an all-or-nothing choice has to be made, very few people today would consciously class a woman as ‘not a person’. But there is surely truth in Johnson’s claim that for many Europeans and Americans a woman is still less of a prototype person than a ‘sane adult white heterosexual’ man. This is to state the situation, not to approve it; but if (as Rosch shows) human minds in fact operate the prototype approach, even if unaware of it, that explains how someone could intellectually reject the Aristotelian/Thomist view and still psychologically feel that a male made a ‘better’ (i.e. more prototypical) representative human being. This creates a vicious circle, because the only way that women will come to be judged prototypical persons is when they are accorded equal status with men in society — and that includes the church.
We end with a more optimistic application of the ‘graded membership of categories’ to women’s ministry. One of the motives for Anglo-catholic opposition to women priests has been a fear that if the ‘purity’ of the apostolic ministry were tarnished, then the Church of England would no longer fulfill article four of the Chicago-Lambeth Quadrilateral (i.e. it would fail to satisfy its own conditions for being classed as a true church). Applying a rigorist doctrine of ‘safety-first’, drawn from moral theology, the argument goes:
Men are able to be priests/bishops (they have been for 2000 years); women may or may not be able to be priests/bishops (it is an open question); if women are ‘ordained’ and it turns out that they are not able to be priests/bishops, major harm will have been done to the church (destroying its apostolic ministry by contaminating it with non-ministers); if women are not ordained and it turns out that they are able to be priests/bishops, only minor harm will have been done to the church (by depriving it of the ministry of certain individuals); thus the safer course — the morally required course — is not to ordain women.
As with other arguments we have considered, this one depends for its force on the classical all-or-nothing concept of the church and the ministry. A woman who has been ‘ordained’ either is a priest/bishop (100%) or she is a total impostor. Either the Church of England has a pure apostolic ministry and is a true church (100%) or it is no church at all. With the prototype or graded-structure approach, the alternatives look much less catastrophic. A woman priest may be slightly further from the prototypical priest than a man would be, but that does not make her ‘not a priest’. A church with women bishops may be a less prototypical example of a church than one with an all-male episcopacy, but that does not make it ‘not a church’. Indeed, for those in an episcopalian tradition, it would still be closer to the prototype than a church without bishops at all. And at some future date, perhaps not very far in the future, a church without both genders represented in its ministry will no doubt be regarded as less prototypical than one that does. In human thinking no concept is static. They are all dynamic and developing, and ‘church’ is no exception.