Masi's art could be called a form of displaced mimesis. It seeks to reproduce the world through a kind of literal substitution (the referents in Masi's tableaux having naturalistic status) but in order that this process be held up to sceptical inquiry. Thus everything in Masi's world may appear to be in place rationally and temporally, but on closer inspection begin to assume another less authoritative reality. His tableaux reproduce the amorality of bureaucratic 'rationality', the obliteration between manipulative and non-manipulative relationships in the pursuit of matching means to ends. The objects that are not-quite-real-objects, the rooms that are not-quite-real-rooms are the products of a dysfunction between meaning and use, between intention and performance; they are objects from a world in which claims to knowledge cannot be made good - claims staged by Masi as a carefully controlled fiction.

Thus when critics accuse him of 'playing to the audience' of fashioning a design-based art, they miss the point completely. The precision, the concern for finish, for synaesthetic effects, are elements co-ordinated to produce the maximum unease through a maximum familiarity with the material. It is no surprise that Masi has regularly cited Brecht and post-war European cinema, particularly the 'constructed' cinema of Syberberg, as influences. Just as conceptualism brought art face to face with its own epistemological status, Brecht's theory of alienation has been influential in securing a certain level of reflexiveness and experimentation for the visual arts. As such Masi's 'power-plays' make very little claim on any definable sculptural tradition. Although it is possible in general to talk about 'expanded' sculpture in relation to this work's site-specificity, its dramatic devices and psychological spaces are clearly theatrical in provenance and direction. As in the theatre, we are asked to construct intelligible and plausible causal links between phenomena, to connect objects and events to those which have preceeded and to those which could be anticipated. However, this dramatization of the tableau as a moment within an unfolding narrative (what Barthes has called the 'pregnant moment' in connection with Brecht's theatre) is less evident in the most recently completed work. In Arena the relationship between viewer and tableau is not so readily governed by the question: what has happened? The absence of a single point of focus loosens the work's narrative impulse; spectator and tableau form a more self-contained relationship; we are more concerned with what is internal to the work.

Arena consists of twelve 3ft square mirrored panels. On each are two stuffed rats in various aggressive or passive attitudes - two are fighting at each other's throats, three copulating, two 'boxing', whilst one pair are mummified, two committing acts of cannibalism and two pairs showing one of the pair retreating after killing the other. As in the other tableaux light and sound ("You are in the Arena" says Masi's growly electronically controlled voice) fill out the sense of apprehension and disturbance, but the viewer's sense of engagement (and estrangement) is extended by the viewers being able to move systematically through the piece. We are in a sense co-opted as viewers.

All Masi's work uses aesthetic distance to create a spectatorship which enforces the moral neutrality of the managerial mode. Given the medical and psychiatric models of the early tableaux, such a relationship might be held to be that between a doctor and a patient. We are there to inspect, to 'put in place' what we see. Thus Masi is not eliciting pity through his human analogues; we might identify with the animals but that identification is never made stable, or rather, never made the point of the work. In Arena this managerial form of spectatorship is brought into sharper focus. To walk into Arena and look down on the animals, each pair confined to its cold pen, is to push the relationship between dominator and dominated into an unambiguous light. If we are the administrator or expert on his rounds in the earlier tableaux then here we are also the taxonomist. Looking and evaluating become not simply a question of 'securing' or verifying the object of scrutiny but of accumulating it as data. The twelve encounters (domestic rituals and public conflicts alike) are surveyed under the same, all-encompassing gaze; we are immediately asked to cross-reference and compare.

However Arena is not simply about how the effects of modern managerial consciousness erodes the pretence of an authentic private world; it is also concerned with how relationships are conceived under bourgeois humanism in terms of those in nature -the constant recourse when making assumptions about human behaviour to use terms borrowed from natural history: the aggressive male, the passive female. Arena, in suitably Brechtian fashion, disturbs this order. Here laid out on twelve stands, like some natural history exhibit, is the life cycle of the rat. It copulates, it fights,and it dies. But as in the other tableaux such verisimilitude is not all that it seems. Some of the pairs are doing un-rat-like things. Two pairs are copulating in the missionary position - only humans do this. Two pairs are fighting at each other's necks - rats are never aggressive in this way. Thus another order - another set of behaviourial references - is inter-posed between what the rats are doing and what actually is permissible for rats to do. The effect of this is not to show how close rats and humans really are, but quite the opposite. The idea of nature being manipulated and made perverse dislodges our assumptions about natural patterns and hierarchies (just as it also reinforces the sinister side of such manipulations: the crudities of eugenics and behaviourial determinism).