Against instrumentalism

The Christian East has a long history of reflecting on the proper role or purpose of liturgical action. As early as the fourth century, figures such as Pseudo-Dionysius and Theodore of Mopsuestia produced elaborate interpretations of the liturgy whose point was to reveal the proper role or purpose of liturgical actions. Two features of their commentaries are especially notable given my purposes in this article.

First, while these commentaries concern themselves with the role or purpose of liturgical actions, they tend to pay little attention to scripted movement-touching sequences. And to the extent they do, they focus almost exclusively on the scripted movement-touching sequences performed by the clergy. When Theodore of Mopsuestia, for example, describes the Great Entrance, in which the unconsecrated gifts are brought to the altar, he does not describe how the assembled bodily engage with these gifts, but focuses exclusively on the typological meaning of the actions performed by the celebrant and the deacons:

In omitting mention of the ways in which the assembled engage in scripted movement-touching sequences and focusing entirely on the typological meaning of the liturgical actions performed by the clergy, Theodore's commentary is not unique. The commentaries produced later by figures such as Germanus (seventh century), Maximos the Confessor (eighth century), and Nicholas Cabasilas (fifteenth century) do the same. The primary differences between these commentaries reside only in the detail with which they describe liturgical action and what they take to be typologically represented by these actions.

Second, to the extent that these commentaries consider the scripted bodily liturgical actions of the assembled, they emphasize that the performance of these actions is for bringing about mental states and events such as the ‘contemplation of the divine things’. Explaining why even small children partake of the eucharist, Pseudo-Dionysius writes:

Bodily actions such as partaking of the eucharistic elements have the role of bringing about states such as contemplating the divine things, Pseudo-Dionysius continues, ‘for it is quite impossible that we humans should, in any immaterial way, rise up to imitate and to contemplate the heavenly hierarchies without the aid of those material means capable of guiding us as our nature requires’.Footnote ⁴

In his Commentary on the Divine Liturgy written nearly one thousand years later, Nicholas Cabasilas voices similar themes:

Here Cabasilas claims that many of those who engage in scripted moving-touching sequences, such as partaking of the holy gifts, ‘gained nothing’ in so acting. When such actions have worth, it is solely because they have certain effects on the soul, namely, engendering religious attitudes such as ‘love of God, faith’ and the ‘desire for mystery’.

What we seem to find in the so-called mystagogical tradition of liturgical commentary, I want to suggest, is an expression of what I earlier called the instrumentalist view of scripted bodily liturgical action.Footnote ⁶

Think of the instrumentalist view as having two commitments. The first concerns the proper role (or purpose) of scripted bodily liturgical action. To isolate this role, we begin by identifying a range of such actions, such as prostrating oneself before an icon, kissing a copy of the Gospels, or touching the hem of the celebrant's vestments. Having done this, we then identify a range of attitudes directed towards God (or propositions concerning God), such as belief, faith, contemplation, awe, reverence, wonder, or love. Finally, we identify a range of instrumental relations that the performance of the scripted bodily liturgical actions can bear to these religious attitudes, such as inculcating, evoking, reinforcing, or expressing them. The proper role or purpose of scripted bodily liturgical actions, the instrumental view tells us, is for an agent who performs them to stand in one or another of these instrumental relations to the religious attitudes. Theodore of Mopsuestia, for example, indicates that the purpose of movement-touching sequences is for their performance to evoke in liturgical participants mental states in which they associate the performance of these action-sequences with those described in the Gospels. Pseudo-Dionysius writes that the purpose of the bodily action of partaking of the eucharistic elements is to put a child who performs it in a position (via the ‘nourishment’ they provide) to contemplate the ‘divine things’.Footnote ⁷ Cabasilas claims that such actions are for producing religious attitudes such as faith and love of God.

The second commitment of the instrumentalist view articulates a thesis about what I earlier called religious worth. The instrumental view tells us that scripted bodily liturgical actions have religious worth but only in an instrumental sense. The religious worth of scripted bodily liturgical actions wholly resides in the fact that their performance is such as to instil, evoke, reinforce, or express those religious attitudes that constitute being fittingly related to God. In some cases, the instrumentalist view is the upshot of a commitment to broadly Platonic views of the person. When detailing the actions that constitute the funeral rite, for example, Pseudo-Dionysius writes that some of its participants anoint the body of the dead with oil. The reason he offers is that the body has participated with, or aided, the soul in the ‘holy struggles’.Footnote ⁸ Similar commitments seem to drive Cabasilas to say what he does (note his final observation about the souls of the dead). While these broadly Platonic commitments, which prioritize the activities of the soul in the spiritual life, can motivate a commitment to the instrumentalist view, I will not assume that they are constitutive of it.

The instrumentalist view has the virtue of directing our attention to important roles played by the performance of scripted liturgical bodily actions such as movement-touching sequences. The view is sensitive, for example, to how religious formation often occurs. When children are inculcated into a religious way of life such as Eastern Orthodoxy, it is typically by first learning how to perform scripted bodily liturgical actions, such as crossing oneself and bowing. If children form religious attitudes such as faith, this ordinarily comes later in their religious formation. Moreover, the instrumental view rightly emphasizes that the religious worth of scripted bodily liturgical actions resides, at least in part, in the instrumental relations that their performance bears to the religious attitudes. If such actions were to evoke attitudes such as faith and love, then they would thereby have religious worth. The question to raise about the view is whether the religious worth of scripted liturgical bodily action resides entirely (or even primarily) in their performance bearing these instrumental relations to the religious attitudes.

I think it does not. Let me indicate why.

Begin with some observations about the liturgical texts themselves. The text of The Divine Liturgy of St. John Chrysostom repeatedly calls our attention to the significance of the body. Speaking on behalf of the assembled, the celebrant says: ‘Forgive us every transgression, both voluntary and involuntary. Sanctify (ἁγίασον/hagiason) our souls and bodies and enable us to serve You in holiness all the days of our life. . . . For You are the illumination of our souls and bodies.’ And again later the people say: ‘enable me, even to my last breath, to receive the sanctification (ἁγιασμὸν/hagiasmon) of the most pure Mysteries, for the healing of soul and body . . . let them be for the healing of soul and body’. Under the plausible assumption that sanctification – literally, to be made holy – is a state that has religious worth, these texts seem to express views about the body that do not comport with what thinkers such as Cabasilas say. These texts seem to indicate that, in virtue of a liturgical participant's performing scripted bodily liturgical actions, such as partaking of the ‘mysteries’, that participant's body is a locus of divine activity: like a soul, the body is the sort of thing that can be sanctified or ‘illumined’.Footnote ⁹ Tellingly, there is no hint of the idea that bodies are a locus of such activity simply because their use instrumentally contributes to having the religious attitudes.

There are also cases that the instrumentalist view seems not to handle well. Consider, for example, some movement-touching sequence whose performance fails to bear an instrumental relation to the religious attitudes, since it does not inculcate, reinforce, evoke, or express any of these attitudes. A young child, say, approaches and kisses a copy of the Gospels or partakes of the eucharist. The child's actions are fitting insofar as the liturgical script directs all qualified participants in the liturgy – whether children or adults – to perform actions such as these. (In the case of partaking of the eucharist, the only qualification is that the child has been baptized.) But the child is too young to have anything that could be called the religious attitudes. So, given a straightforward sense of ‘express’ according to which to express an attitude by performing an action requires either having that attitude or committing oneself to having it, her actions do not express the religious attitudes.Footnote ¹⁰ She has no such attitudes and, arguably, cannot commit herself to having them since she lacks the conceptual sophistication to have them. And ex hypothesi since the child lacks the conceptual wherewithal to have the religious attitudes, the performance of these bodily liturgical actions does not evoke or reinforce these attitudes at the time at which she performs them.

Now suppose that the repeated performance of acts of these types also fails to inculcate or evoke these attitudes. Perhaps the child dies before religious piety has a chance to take hold. Or perhaps piety never ‘takes’ for this person even though she is raised in a thoroughly religious environment. In the first case, given the assumption that the only sort of religious worth that the performance of bodily liturgical actions has is instrumental, it follows that the child's acts have no religious worth; they do not relate her to God in ways that are fitting.Footnote ¹¹ In the second case, it may be that the agent expresses the religious attitudes in the sense of committing herself to them by the performance of scripted bodily liturgical actions. But since she lacks these attitudes, the instrumentalist view also implies that her acts lack religious worth.

I find these implications of the instrumental view jarring – not simply because it is hard for me to believe that there could be such an asymmetry between a child's performance of some movement-touching sequence and the religious worth of the same sequence performed by, say, members of the clergy. The conclusion also strikes me as jarring because it sits so uneasily with currents that run deep in the tradition.

Some of these currents are at work in what Paul says about illicit sexual activity in his first letter to the Corinthians (I Cor 6: 15–20). When Paul admonishes his (male) readers not to engage in such activity, he does not say that such activity is demeaning because it uses another simply as a means, or contributes to oppressive social arrangements, or is contrary to a divine command (although the passage should probably be read as taking the latter for granted). In fact, he makes no mention at all about the intentions or mental states of the people who engage in these activities. Rather, what he emphasizes is that our bodies bear certain relations to God: they are part of the ‘body of Christ’ or are the ‘the temple of the Holy Spirit who is in you’. It is in virtue of the fact that our bodies bear these relations to God, Paul writes, that bodily actions of certain kinds have normative properties. In fact, Paul seems to suggest that the disvalue that sexual activity of certain types can have is religious disvalue; the person who performs these acts is not fittingly related to God.

Paul does not spell out for his readers why this is. That is, he does not tell us what it is for a person's body to be part of the body of Christ or a dwelling place of the Holy Spirit. And he does not explain why it is that if our bodies bear these relations to God, then bodily actions of various kinds have religious value or disvalue. He simply points to the phenomenon itself. But if Paul's thinking on this topic does in fact articulate commitments about the religious worth of bodily action, then it sits uneasily with the instrumentalist view. For what he says readily generalizes to the case of liturgy. If bodily actions can have religious value or disvalue in virtue of the fact that our bodies bear certain relations to God – and not simply because they bear one or another instrumental relation to the religious attitudes – then scripted bodily liturgical actions can too. They can have religious value or disvalue simply in virtue of the fact that the bodies of those who perform them bear certain types of relations to God.

The authorization-appropriation model

The Pauline-inspired approach mentioned above is an instance of what I earlier called the non-instrumentalist view. Non-instrumentalism holds that scripted liturgical movement-touching sequences have religious value but denies that they have it solely in virtue of the instrumental relations that their performance bears to the religious attitudes.

The primary attractions of non-instrumentalism are that it not only appears well-suited to account for why scripted liturgical bodily actions can have religious worth in a sufficiently wide range of cases – such as cases in which their performance fails to bear instrumental relations to the religious attitudes – but also that it is sensitive to the various ways that we can be fittingly or unfittingly related to God, including the ways in which we use our bodies. Still, the position is puzzling. For suppose our bodies could bear relations to God that are fitting or unfitting in virtue of how we use or move them. We are familiar with many physical relations that our bodies can bear to things in virtue of how we use or move them, such as being to the left of and having higher velocity than. Presumably, though, relations such as these do not have religious worth. What, then, is it about the performance of scripted movement-touching sequences, other than their instrumental relation to the religious attitudes, that could give them religious worth?

In the passages to which I alluded above, Paul seems to suggest that bodily actions can have religious worth in virtue of the fact that our bodies bear certain kinds of relations to God. In what follows, I want to explore a position that develops this Pauline approach. In doing so, I am going to largely neglect one image that Paul uses (being a temple of the Holy Spirit) but draw loosely upon another (being part of the body of Christ). I should forthrightly admit that I take this Pauline-inspired view to be only one way to develop non-instrumentalism. If non-instrumentalism is true, I suspect that there are a variety of ways in which (and explanations why) scripted liturgical bodily actions have non-instrumental religious worth. The Pauline-inspired view identifies only one such relation.

The task of the Pauline-inspired approach, then, is to:

identify a relation that God bears to something such that the performance of scripted movement-touching sequences has religious worth (at least in part) in virtue of God's bearing this relation to that thing.

I propose that the ‘something’ in question are the liturgical scripts and that a promising candidate for the relation in question is:

having authorized and appropriated the liturgical scripts that prescribe these actions to liturgical participants.

Defending this proposal will require developing what I'll call the authorization-appropriation model of liturgical composition.

The authorization-appropriation model is an extension of one that Nicholas Wolterstorff has employed to explain why it is that the scriptural texts count as God's speech.Footnote ¹² The model hinges on the claim that a crucial element in Jesus of Nazareth's announcement and inauguration of the kingdom of God was his authorization of a group of people to speak and act in his name and, hence, to speak and act in the name of God. Among the actions that this group, the church, took itself to be authorized to perform is that of identifying and composing its canonical texts and determining the character of its worship – it being understood that worship lies at the very core of the church's mission.

At this point, it will be helpful to regiment some terminology. Let's call any sequence of act-types that composes the church's worship a liturgy and the performance of these act-types an enactment of a liturgy. While the church's liturgies could have taken radically different forms, they didn't. The church endeavoured to formulate guidelines for its worship, wherever and whenever it might be assembled. Call a more or less complete set of guidelines for the enactment of a liturgy a liturgical script. A liturgical script addresses such questions as: which actions constitute a liturgy, when a liturgy is to be performed, how it is to be performed, by whom, and in what manner. What I have been calling scripted liturgical actions are simply those that are prescribed by – that is, permitted, recommended, or required by – a liturgical script.

I'll make three assumptions about the relation between liturgies and liturgical scripts. First, I assume that for every liturgy, there is a corresponding liturgical script. I do not assume, however, that a liturgical script must take a written form (any more than the rules of chess must take a written form). Second, I assume that the performance of a sequence of actions counts as the performance of a liturgy only if it sufficiently conforms to its corresponding liturgical script. And, third, I assume that the church composes important parts of its liturgies by composing sections of its corresponding script. Indeed, I'll assume that the composition of a liturgy consists in the composition of sections of its corresponding script: to compose is to script actions.Footnote ¹³ I acknowledge that it is a mistake to identify the church's activity of determining the character of its worship with the activity of composing liturgies. In some cases, the church's liturgies seem not to have been composed but to have simply organically developed. When it did, the church ratified these existing patterns.Footnote ¹⁴ That noted, I am going to stretch ordinary usage and call the distinct activities of composing and ratification ‘composing’. I'll speak, then, of the church having composed its liturgies, but this should be understood to mean that the church composed or ratified its liturgies. These clarifications having been made, we can identify the first implication of the authorization model, which will prove important in what follows: the composition of the church's liturgies is divinely authorized.

Authorized action comes in a variety of forms. In some cases, when A authorizes B to act, B acts on behalf of A. B's acts count as A's acts. Authorization of this sort is deputization. In other cases, when A authorizes B to act, B does not act on A's behalf. B's acts do not count as A's acts. Authorization of this sort is delegation. I think the authorization to compose the church's liturgies is best understood to be a blend of both deputization and delegation. An example may help to illustrate the idea.

We know that during different periods of history, participants in the social practice of producing artworks have operated with different arrangements for the composition and attribution of artworks. According to one such arrangement, an artist A authorizes an apprentice B to produce a painting – say, a still life – which is to have a certain look. B paints the still life by performing a series of actions. A grants B considerable freedom in the production of the still life. While always keeping in mind the look that the painting is supposed to have, B is often free to choose from different colours, textures, shading techniques, and so forth. The actions performed on the basis of these choices are B's. However, at various points, A carefully instructs B to perform certain actions, such as using a certain brush technique or painting over a section of the work that A doesn't like. In these cases, since A has deputized B to perform these actions, they count as A's. At a certain point, B finishes painting the still life. A approves of it, since it has the look he wants, signing his name to it. Given the conventions in effect, the acts that constitute composing the painting are a blend of B's and A's. When the composition is finished, A signs the painting, appropriating it as his own. It counts as A's work.

According to the authorization-appropriation model, something similar is true of the church's liturgies. When God authorizes the church to determine the character of its worship, it does so by composing (often over a significant stretch of time) a variety of liturgies by way of composing their corresponding liturgical scripts. The authorization in question, I think, is best viewed as a species of delegation. God delegates the authority to the church to compose these liturgies, albeit subject to constraints of various sorts, such as that these liturgies satisfy certain intentions that God might have with regard to worship. Three types of decisions face the church as it composes the scripts that correspond to these liturgies.

The first concerns the scope of a script's application or those to whom the script applies. The church's answer to this question has been that the guidelines that constitute a script apply to whoever assembles for worship whatever their social group, sex, age, or background might be. In rendering this compositional decision, the church presumably took itself to apply to the case of worship something like the Pauline declaration that there is no partiality or favouritism in Christ (cf. Gal. 3:28; Rom. 15:17). All are invited to the banquet.

The second decision concerns which actions the script is to prescribe and, hence, which actions to select (or ratify) for legitimate inclusion in a liturgy. These decisions are made by employing two sorts of criteria for selection (or ratification). Some actions are selected (or ratified) on the basis of whether their performance is divinely required. For example, the church interpreted Jesus to command his followers to celebrate the eucharistic meal and baptize those wishing to join the church. But other actions are selected (or ratified) on the basis of whether their performance is fitting. Singing the Psalms, for example, has been selected (or ratified) not on the basis of whether singing them was divinely mandated but because it is fitting to sing them in the enactment of a liturgy. As I indicate below, I take most instances of movement-touching sequences to be selected (or ratified) on these latter grounds.

The third decision concerns the scope and normative force of the prescriptions that constitute the guidelines of a liturgical script. I have just noted that the church selected (or ratified) actions for legitimate inclusion in its liturgies on the grounds that their performance is either divinely mandated or fitting in the context of worship. But it is one thing to select (or ratify) an action on the basis of whether its performance is divinely mandated; it is another to decide whether the performance of that action is required in the context of a liturgy. Similarly, it is one thing to determine whether an action is fitting to perform in the context of worship; it is another to decide whether a liturgical script should prescribe the performance of such an action.

For sometimes the prescriptions concerning an action go beyond (and even come apart from) the normative basis on which that action is selected (or ratified). For example, the normative basis on which the church selected (or ratified) the eucharistic meal for legitimate inclusion in the liturgy is that the Gospels present Jesus as having commanded his followers to celebrate a meal of a particular kind in his remembrance. When Jesus issued this command, however, neither the content of the command nor the context in which it was issued appears to determine that the celebration of this meal is to occur only in the context of the church's worship. Moreover, once it is decided that the celebration of this meal is to occur only within the church's worship, that decision does not determine that it should be celebrated every Sunday – or at most once a day by a parish on a given Sunday. Yet this is what the script of the Divine Liturgy prescribes. By including these directives in its liturgical scripts, the church went beyond the normative basis on which it selected (or ratified) the eucharistic meal for inclusion in its worship, fixing the normative scope of Jesus' directive, and issuing higher-order directives regarding how to conform to it.

A second example: in the composition of its liturgical scripts, the church eventually determined that it was fitting for the assembled to venerate copies of the Gospels, replicas of the cross, relics, and icons; it selected (or ratified) these actions for legitimate inclusion in the liturgy on this basis. But it took a further step. Rather than leave it up to the individual parishes to determine whether to disallow such behaviour on, say, prudential grounds – perhaps because engaging in it could easily lead to impious behaviour – it also required that every parish permit the veneration of icons within the context of the enactment of its liturgies. Moreover, given both the language and the church's understanding of the character of the ecumenical councils, it appears that the church did not understand itself to be delegated to make these decisions. It appears as if the church understood itself to be deputized to make them. In requiring every parish to permit the veneration of icons, the church took itself to be acting on behalf of God, expressing God's intentions as to how human beings are to interact with material reality.

A final case: for centuries, the eucharist had been celebrated by the celebrant placing the eucharistic bread in the hands of a communicant. Around the eighth century, however, this practice changed in the Eastern churches.Footnote ¹⁵ The church decided that, when celebrating the eucharist, the eucharistic bread should be mixed with wine and fed to the communicants by spoon. While the exact rationale for this change is unclear, it seems not to have been because the action of taking the bread in hand was deemed to be unfitting; rather, it is because the action at times was considered to have undesirable consequences, such as that communicants drop the bread to the floor when communing.Footnote ¹⁶ In issuing this prohibition, the normative force of the liturgical script's prescriptions regarding an action (namely, that its performance was prohibited) failed to coincide with the normative basis on which that action was selected (or ratified) for legitimate inclusion in the liturgy (namely, that it is fitting). In virtue of the church's decision, the action fell into the category of fitting but not permitted.

We have been exploring what it would be for the composition of the church's liturgies to be divinely authorized and which activities constitute that composition, paying attention to the decisions that face the church when it composes its liturgical scripts. Let us now turn to the second part of the authorization-appropriation model, which holds that the church's liturgies are divinely appropriated: God does not simply authorize their composition but appropriates them as God's own work.

When presenting its understanding of scripture, the Christian East has typically committed itself to two claims explicitly. The first is to a particular understanding of what it is for scripture to be inspired (2 Tim. 2: 36). According to the main lines of thought in the East, the inspiration consists in there being a synergy, ‘a cooperative effort between the human author and the Holy Spirit’ in the production of the text.Footnote ¹⁷ The second claim concerns the relation between scripture and liturgy. For scripture to be understood and for it to fulfil its aim of enabling ‘the people of God to hear . . . [it] . . . and to receive it for their salvation’, it must be heard and celebrated within the ‘liturgical “synaxis” ’, the gathering of ‘seekers, catechumens, and faithful’.Footnote ¹⁸ Unpacked a little, the claim is that the liturgy provides a context in which the meaning(s) and significance of scripture can be properly understood and appropriated. It is, in part, the liturgy's hymnody and creedal declarations that make manifest its meaning(s) and significance. In making these claims, the tradition implicitly commits itself to the further claim that the liturgy is also inspired in the sense of being the product of a cooperative effort between its human authors and the Holy Spirit. For only if the liturgy has this character could there be sufficient reason to hold that it provides the context in which the meaning(s) and significance of scripture are made manifest in such a way that the assembled can ‘receive them for their salvation’.

To appreciate the point, suppose the liturgy were not a cooperative effort of this sort. In that case, it would be entirely the product of divine action, or entirely the product of human action, or the product of both but not a cooperative venture. The problem with the first option is that it's incompatible with what we know about the liturgy: the liturgy did not fall from heaven. The problem with the second and third options is that, if the liturgy were to provide the sort of context and content in which the meaning(s) and significance of scripture can be properly understood and appropriated, this would be a happy accident. But, unless we had independent reason to believe that this sort of accident had occurred, we would have no good reason to trust that the liturgy provides the sort of context and content in which the meaning(s) and significance of scripture can be properly understood and appropriated. I take it that we have no such independent reason.

That noted, reflection on our earlier illustration of how a work of art could be authorized and appropriated reveals that it's possible for something to be the product of a cooperative effort between a pair of agents and yet for neither of them to own the work. You and I, for example, might work on a painting together but upon completion, you might thoroughly dislike it. In such a case, there is a straightforward sense in which the work belongs to you. But there is also a perfectly straightforward sense in which you do not own it. For you to own the work – for it to be yours or represent you – you must endorse it. However, in its understanding of scripture, the Christian tradition has insisted not simply that scripture belongs to God in the sense that God has been active in its production, but that God owns the scriptures in the sense of endorsing its contents. If this is right, appealing to the notion of synergy is not enough to capture what the tradition has wanted to say about scripture. Nor is it enough to capture what the Christian East is committed to saying about the liturgy. There is a gap between a work's belonging to an agent and an agent's owning a work.

The authorization-appropriation model bridges this gap. For by appropriating both the liturgies and their corresponding scripts, God thereby owns these works. This observation allows us to identify a second implication of the authorization-appropriation model, which concerns religious worth. Suppose that God were not simply to cooperatively produce but also to appropriate the church's liturgies and their corresponding liturgical scripts. If so, then the performance of the actions prescribed by the script would have the property of meeting divine approval or being the subject of divine endorsement. Given the further assumption that, when the performance of an action meets divine approval or is subject to divine endorsement, then (all else being equal) it fittingly orients that person to God, it follows that the performance of that action has religious worth and not simply because it bears one or another instrumental relation to any attitudes of the person who performs them.

Applying the model

In the last section, I presented the rudiments of the authorization-appropriation model of liturgical composition. I want now to apply the model to scripted liturgical movement-touching sequences.

A point that emerged in our discussion is that the church employs selection criteria of two kinds when composing its liturgies: one is whether the performance of an action has been divinely mandated; the other is whether its performance would be fitting in the context of worship. We also saw that movement-touching sequences of various sorts can be selected (or ratified) because they satisfy either of these criteria. Some movement-touching sequences, such as eating in remembrance, fall into the category of being divinely mandated; other movement-touching sequences, such as venerating icons, fall into the category of being fitting. While there is this difference, the vast majority of movement-touching sequences prescribed by the liturgical scripts appear to be of the latter sort: they were selected (or ratified) because they are cultural expressions of reverence, love, awe, wonder, and longing that were deemed fitting for legitimate inclusion in the liturgies.

In making this last observation, I am working with a distinction between an action's expressing a religious attitude and its being expressive of such an attitude. Your performing an action expresses a religious attitude A at t, I said earlier, only if you are in A at t or you commit yourself to being in A at t. In contrast, your performing an action is expressive of an attitude A at t just in case and because it would be fitting to express A by performing that action at t.Footnote ¹⁹ This distinction opens up the possibility of an act's being expressive of an attitude without expressing it. An agent's performing a piece of music, for example, might be expressive of joy because it would be a fitting way to express joy. But the performance itself might not express joy. For this agent might not be in a state of joy or commit herself to being in such a state when performing the piece. Neither need the composer of the work have been in such a state when composing it (assuming, of course, that the composer of the work is distinct from the agent who is performing it).Footnote ²⁰

Suppose, then, that the church's liturgical scripts often prescribe the performance of actions on the basis of their performance being fitting in a liturgical setting. We now have a clearer sense of why their performance is fitting in such a setting: it is because these acts are expressive of the religious attitudes. But we now face this question: If an act is expressive of the religious attitudes, does it thereby have religious worth simply in virtue of its being so expressive? This question matters for our purposes. For if the answer is yes, then there would be no need to appeal to the authorization-appropriation model to account for the religious worth of the performance of scripted movement-touching sequences in the enactment of a liturgy. Their being expressive of fitting religious attitudes would do all the requisite explanatory work.

This question raises some subtle issues about the character of worth that we need to address. To orient ourselves, distinguish prima facie worth or defeasible worth, on the one hand, from worth on the whole or undefeated worth, on the other.Footnote ²¹ The first point to see is that, although an action is sometimes expressive of some fitting attitude and, thus, has prima facie interpersonal worth, its worth is easily defeated.

Consider the following two cases. Your neighbour's child has behaved very badly. You punish the child, thereby expressing attitudes that are expressive of deep disapproval. These attitudes are apt in one sense because they orient you towards the child in ways that are appropriate given her behaviour. Still, they lack interpersonal worth on the whole because they do not fittingly relate you to the child. For you ought not to have punished the child. You lack the standing to do so: you are not the child's parent or guardian.

Or imagine that it is your twenty-fifth wedding anniversary. You perform actions that are expressive of affection, such as giving gifts and exchanging cards. But you feel no affection towards your spouse. The actions you perform are in one sense apt because they are expressive of attitudes that are fitting. But these acts do not have interpersonal worth on the whole because they are deficient in affection. What is called for in this situation is your having or feeling affection for your spouse. Only then will your actions have interpersonal worth on the whole, fittingly relating you to your spouse.

What such cases illustrate is that acts can be expressive of attitudes that are apt in one sense and yet lack interpersonal worth on the whole because this prima facie worth is defeated. Arguably, the iconoclasts pressed a point in this neighbourhood against the iconodules. Venerating icons might be apt in one respect insofar as it is expressive of genuine piety directed towards God. But, the iconoclasts claimed, such veneration is misguided because it is idolatrous. No one has the standing to do it. As such, veneration of icons is an act-type whose performance is forbidden by God and, so, its performance lacks religious worth on the whole because whatever prima facie religious worth it has is defeated. Under a natural reading, the instrumentalist view is driven by the conviction that, insofar as scripted bodily actions are expressive of attitudes of reverence and awe that are fittingly directed towards God, they might be apt in one sense. But Cabasilas and others indicate that more is called for. For these actions to have genuine religious worth on the whole, their performance would have to do more than be expressive of the religious attitudes. According to the instrumentalist view, their performance would have to express these attitudes or be such as to instil, evoke, or reinforce them.

The authorization-appropriation model is sensitive to these points, explaining why movement-touching sequences can have stable, non-easily defeasible religious worth. Suppose, first, the model correctly claims that the church's composition of its liturgies and corresponding scripts is divinely authorized. And suppose that these liturgies and their corresponding scripts satisfy certain criteria, such as being in keeping with whatever intentions God might have regarding the character of worship. It follows that the actions that they prescribe, such as moving-touching sequences in which agents venerate icons, are divinely authorized. If so, then these scripted movement-touching sequences do not fail to have religious worth on the whole because no one has the standing to perform them.

Suppose, second, that the authorization-appropriation model correctly holds that the church's liturgies and corresponding scripts are divinely appropriated. If they are, then God thereby owns these works in the sense of endorsing or approving what the church has been authorized to produce. (This is, incidentally, consistent with God's not endorsing every element of these works. But it does mean that these works on the whole meet divine approval.Footnote ²²) Liturgical scripts, we've seen, prescribe the performance of actions of various sorts, including movement-touching sequences. All else being equal, by appropriating a liturgical script, God thereby endorses the performance of those actions: an agent could not coherently endorse the prescription to act in a certain way and not approve of someone's acting in that way in virtue of conforming to that prescription. But if the performance of an action meets A's approval in virtue of A's having appropriated the work that prescribes it, then (all else being equal) an agent's performing the prescribed action fittingly relates that agent to A: that agent's performing that action has interpersonal worth on the whole, provided of course that the action prescribed does not violate some other prescription that has genuine normative force, such as a moral prescription.

The application of these last points to the performance of scripted movement-touching sequences is straightforward. The church's liturgical scripts prescribe the performance of these actions. All else being equal, by appropriating and thus endorsing these scripts, God thereby endorses the performance of these movement-touching sequences. But if the performance of these actions meets divine approval or endorsement, then (all else being equal) an agent's performing these actions fittingly relates those who perform them to God: all else being equal, their performance does not lack religious worth on the whole because more is required of those who perform them (such as their being in certain attitudinal states while performing those actions).

Now let me add several clarifications. The first concerns the basis on which God might appropriate the liturgical scripts that prescribe the performance of scripted movement-touching sequences. The authorization-appropriation model does not leave us entirely in the dark as to why the performance of these actions would meet divine endorsement: these are authorized actions whose performance is expressive of religious attitudes such as love, reverence, awe, and longing, which fittingly orient us to God. It is true that sometimes we can do better than perform authorized actions that are simply expressive of some attitude that is apt. Sometimes we can also have (or commit ourselves to) the very attitude of which an act is expressive. When venerating an icon of Christ, for example, an agent might perform an action that is not only expressive of reverence for Christ but also expresses reverence in the sense that she is reverential (or feels reverence) while venerating.

Nonetheless, it may that the best way to understand God's appropriation of the products of human effort, such as the church's liturgical scripts, is not as an acknowledgement of their excellence, or the excellence of the actions they prescribe, but as an exercise of grace. If that were so, God's appropriating and endorsing these scripts – and, hence, endorsing the performance of the actions that they prescribe – would not be explained completely by their excellence or the excellence of the actions that they prescribe. And, hence, the fact that these actions have religious worth would not be explained completely by the excellence of their performance. Rather, these actions would fittingly relate us to God and, hence, have religious worth, because God endorses their performance. In this way, small and imperfect gestures could thereby be transmuted into something considerably more significant. In fact, the authorization-appropriation model can even take a page from the instrumental view, holding that God endorses the performance of scripted moving-touching sequences (in part) because they often tend to bear instrumental roles to the religious attitudes, which have religious worth. If that were so, there might be multiple reasons why God appropriates the church's liturgical scripts and endorses the actions that they prescribe.

The second clarification concerns the type of religious worth that movement-touching sequences have if the authorization-appropriation model is correct. What I claimed is that the authorization-appropriation model explains why these action sequences have stable, not easily defeasible religious worth. Still, the worth in question remains prima facie. For the worth of these acts can be affected by a variety of factors.

An assumption that has guided my discussion is that for scripted movement-touching sequences to have religious worth, those who perform them needn't be in any particular mental state while performing them. For like many activities that are aptly performed – whether musical, artistic, or athletic – liturgical action that is aptly performed often involves its performers entering into what psychologists call a flow-state. They are entirely absorbed in the activity and do not form explicit beliefs about how to do what they are doing – or even beliefs about what they are doing – while they are doing it. But it is also true that there are mental states such that it would be bad to be in them when participating in those rites. In fact, the liturgical text itself calls this to our attention when, in the prayer of confession that is recited immediately prior to communion, it states: ‘Of Your Mystical Supper, O Son of God, accept me today as a communicant; for I will not speak of your Mystery to your enemies; neither like Judas will I give you a kiss.’ Kissing is an action that is expressive of reverence and affection. But it can also (at once) be an act of betrayal, depending on the intentions with (and perhaps the context in) which it is performed.

So qualification is called for. If the authorization-appropriation model is correct, the religious worth enjoyed by scripted movement-touching sequences is prima facie, since it is defeasible. Their prima facie value is defeated when an agent performs them in ways that are bad. In principle, their performance can be sufficiently bad in numerous ways such that this badness defeats their prima facie value. Being carelessness, indifferent, or hostile are among those ways. (A person's lacking the conceptual wherewithal to form the religious attitudes is not.) To which I'll add that in this regard the worth of such actions is similar to that enjoyed by the religious attitudes. The religious worth of the religious attitudes is also only prima facie. The religious worth of love, reverence, piety, and the like can also be defeated if their expression is deeply defective.