In documenting examples of modern productions of Greek drama for the Research Project on The Reception of the Texts and Images of Ancient Greece in Late Twentieth-Century Drama and Poetry in English, the researchers are attempting to make real the conventional wisdom that performance is as important as text. A major complication is the problem of capturing ephemeral information about the details of the staging of Greek plays, whether in the original, in translation or as adaptations and versions. In addition to company material in theatrical archives (including acting scripts, prompt copies, stage managers' annotations, programmes, listings, photographs, videos) the theatrical Review is an important source of information, not only about archived performances but crucially about those which are not otherwise well documented, perhaps because the company lacks its own archive. In documenting performances it is particularly important to capture evidence about the work of smaller and 'ad hoc' companies and to research performances presented on tour or at festivals. A prime aim of this Research Project is to ensure that carefully assembled data is available for future cultural historians. If a representative range of productions is not documented, it will be impossible for cultural historians to assess the full impact of Greek drama on late twentieth-century theatre and its audiences. In the body of this article, references to modern productions which have been documented by the Research Project are accompanied by their database reference number.
As the Research Project develops, special studies will be made of the different types of primary source performances which are involved in documentation and critical analysis of modern performances. Possibly the most problematic of the primary sources in the documentation of Performance is the Review. Much work in Reception Studies makes use of the Review as a major source and critical evaluation of its strengths and weaknesses as a source is perhaps overdue. Some attempt at typological analysis and evaluation is necessary in order to assess the impact and reliability of the Review. It functions not only as one type of primary source which gives an account of the performance with details of staging, acting style, design, music, choreography, properties and audience response, but also as a secondary critique of the performance. In their critiques, reviewers argue from perspectives which may appear to be separate from the (selective) recording of detail or the narrative account of a particular performance but in fact these aspects are interwoven.
A basic typological analysis of Reviews relevant to this Research Project reveals a considerable variety of examples in relation to: -
In the discussion which follows, reference will be made to examples from local and regional newspapers, national 'broadsheet' newspapers, literary periodicals and academic journals.
A prominent example of a work heavily dependent on source material from Reviews is Karelisa Hartigan's influential study Greek Tragedy on the American Stage: Ancient Drama in the Commercial Theatre 1882-1994 (Contributions in Drama and Theatre Studies Number 60, Westport CT, Greenwood Press, 1995). This is a significant contribution to Reception Studies. It presents a chronological performance history of Greek tragedy on the late nineteenth and twentieth-century American stage and argues from this that the history of Greek tragedy production corresponds to the social history of the nation (and especially to periods of war and peace). In her discussion of her use of sources Hartigan states that :
'The information has been gleaned from the Reviews penned by drama critics of the major newspapers, from programs, press releases and magazine articles' (Preface page x).
Clearly, Reviews should not be placed in the same category of source as Programme Notes or Press Releases, the first of which aims to inform the live audience and the second to inform the media. Both aim to promote the play. Hartigan's discussion recognises that Reviews yield information and that at least some of it claims to be factual. It is surely an oversimplification, however, to group all Reviews together – and restrictive to confine notice to those written by the drama critics of the major newspapers. Far worse, however, would be the temptation to take the Review as an authoritative source of factual information about either the ancient play or the modern production. For instance, in discussing the Circle Repertory Company's staging of Aeschylus' Persians in May 1974 (Hartigan p103; DB ref. no. 1047), Hartigan quotes Joseph Mancini's description of the performance as an 'unimaginative staging of an anti-war play' (New York Post 20 May 1974) and yet fails to discuss the relationship between the two parts of this judgement, or indeed the evidence on which it was based. Similarly, a later reference (p 134) to Walter Goodman's Review (New York Times 27 December 1985) of Richard Schechner's The Prometheus Project (DB ref. no. 1048) concludes 'Goodman's Review pointed out what must have been the failing of this modern version (the attempt to link Hiroshima bombing and sexual abuse)'. In addition, reviewers' statements are sometimes quoted in a kind of oratio obliqua which has the effect of giving them factual status and which certainly neglects to examine the nexus between the narrative of the performance and the critique. This sometimes makes it difficult to distinguish between the judgements of the author and those of the reviewer and masks the evaluation of the Review which the researcher must undertake before accepting it as a reliable narrative or descriptive source.
The narrative contained within Reviews is complex because the author's voice is frequently coloured by underlying perceptions and emotions outside the apparently objective description. The authorial voice needs as much attention from the reader and researcher as is the case with canonical literary works. The patterning of the relationship between descriptive narrative and perspective is analogous to that of 'focalisation' as identified by literary theorists, except that in the case of the Review focalisation is not bound by the insights of a figure within the script but draws more widely on critical and cultural viewpoints, which are part of a wider discourse (one might even say, a larger stage). This broader 'focalisation' is often connected with the reviewer's perceptions of the relationship between Greek drama and modern theatrical issues or with means of translating the theatrical experience into the imagination and understanding of those who have not actually witnessed it. A related analogy may be with the Messenger Speech in Greek tragedy. As the Messenger in the play focusses the response of the actors and of the audience by reference to events which have taken place off stage, so does the reviewer become an important figure in seeking to focus the response of the reader who may either not have witnessed the performance at all or who may be seeking a re-enactment, interpretation and value judgement after the event. The Messenger's verbal communication involves a narrative which is presented through the filter of its relationship to the imagery and structure of the world of the play. In the Review this aspect of narrative is maintained but is also extended and located in the cultural framework which the reader and reviewer purport to share. It is this framework which shapes the language used by the Messenger/Reviewer.
An approach which contrasts with Hartigan's assumptions about the status of the Review as a source, is shown in a detailed case study of the response of reviewers to new work which has been made in a South African context by Yvonne Banning. Banning examined a selection of Reviews of the Mark Fleishman/Jenny Reznek production of Medea in order to examine a variety of perspectives on the relationship between the "new" and the "already known". She argues that since the Review is generated in the cultural context of the Reviewers themselves it is a mistake to construct a causal or sequential relationship between the production and reception of a performance or to suggest that the complex relationship between visual and auditory images and semiotic systems is subject to simple 'interpretation' or 'misinterpretation' by audiences. She demonstrates that the role of reviewers can play a crucial mediating function between theatrical intention and the cultural transformation which results not only from witnessing the play but from reading discussions of it. She argues that reviewers sit at the intersection of routes coming from theatre aesthetics, economics, popular culture and "personal preference". Banning's concern is with questions of cultural identity and of gender in South Africa and especially with ways in which new theatrical work has been measured against traditional Europe-orientated assumptions about the conventions and staging of Greek drama.
Research on the cultural stance and politicised perspectives of Greek drama in South Africa during the height of the apartheid regime is being undertaken by P.J. Conradie, who has suggested that traditional expectations concerning staging and dramatic conventions sometimes impeded openness to new interpretations which were culturally and politically challenging even when the political importance of such challenges was acknowledged. In such a context, the Review may be more valuable as a source for the values of the receiving individuals or society than as a record of the performance itself. Furthermore, I suspect that reviewers expressions of aesthetic distaste, whether in the South African context or elsewhere, may function as a mark for ideological resistance to the implications of the production.
Therefore I wish to argue that before it can be used as evidence, either simply in performance history or in the broader theoretical framework required by Reception Studies, the Review has to be subject to an evaluation which draws out its context and purpose, examines its language and style, and considers other types of evidence which can be used to check and complement it.
The range of cultural evidence provided by different kinds of Review can be demonstrated in examples from a variety of Reviews which the Research Project has used. Detailed comparison of Reviews reveals differences in critical standpoint, which also influence the selection of material in the Review. For all these reasons, in the Research Project we try where possible to reference a number of Reviews of different kinds for each production.
The Halifax Evening Courier, September 1996, published a Review by Margaret Woods of the Medea performed by the Actors of Dionysus at the Square Chapel, Halifax.
'It's got murder, madness, revenge and obsession – and, no, it's not the plot of the latest Hollywood blockbuster or new soap-opera…Its scaring themes of passion and vengeance speak down the centuries'.
This local paper Review operates on a number of levels and is actually quite an effective piece of writing in that it :
If the intention was to entertain and to inform a readership interested in local theatre but unaware of the conventions of Greek drama, then this Review probably achieved its object. However, it tells us virtually nothing about the translation, set, acting styles or effects of theatre crafts. Nor does it tell us about the audience's response.
Writing of the same touring production in the West Sussex County Times (13 October 1996) Stephanie Polak was less appreciative –
'like watching an episode of East Enders…the play was pessimistic and full of gloom and doom, with much wailing and moaning..and after a while this can get a bit too much – especially as the characters aren't exactly sympathetic…the Actors of Dionysus did not go for shock value and show the murder of the two sons by Medea…maybe this would have 'livened' the play up a bit, but nowadays we are immune to such horror as nothing seems to shock people anymore'.
This tells us more about cultural attitudes in West Sussex in 1996, including the reviewer's ignorance of the conventions of Greek drama, than about either the play or the production. Interestingly, Charles Hutchinson in the Yorkshire Evening Press (6 September 1996) described the production as 'sulphurous – a smouldering horror show' while Time Out (Patrick Marmion) thought it was 'largely naturalistic and well mannered – costumes not out of place in a Laura Ashley Catalogue'. I shall not pursue the implication that Laura Ashley styles may be considered sulphurous in Yorkshire but merely suggest that comparison of Reviews is a necessary control on the status of apparently 'factual' information in narrative accounts. Not only does a Review have various strands (narrative, critique, self-advertisement) but local newspaper, national broadsheet and classical journal all have different aims and readership and a greater or lesser interest in discussing wider critical or social issues (although they all make assumptions about these).
Reviews also contain implicit assumptions about the audiences – audiences for the Review and for the performance. Fintan O'Toole has suggested that 'Greek plays are predicated on the idea that everyone in the audience already knows the story – we are moving from a theatre of conflict to one of linguistic evocation, one in which things are called up rather than simply acted out' ('How poetry joins dramatic action', Guardian 29 November 1990). This may become truer for modern audiences as the number of revivals of Greek plays grows ever larger, but I am not sure that it is true for readers of Reviews outside the literary and theatre journals.
A comparison between two Reviews of the 1997 production of Sophocles' Electra, directed by David Leveaux, reveals startling contrasts in critical judgement and in the status given to the Programme Notes.
In the Times Literary Supplement (21 November 1997), Jane Montgomery addressed in sequence the nature of Sophocles' source text, comparison with Warner's 1991 staging (this RSC production actually began in December 1988), analysis of performance style and the question of 'contemporary relevance'. Montgomery referred to 'a return to the bad old days of Greek tragedy productions: statuesque declaiming' and to the creation of an ill-conceived Bosnian context which 'cheapens both the tragedy of Sarajevo and the chorus' theatrical meaning'. She also criticised the steady drip of water, turning to blood as revenge is enacted, dismissing it as stage dressing which misses the theatrical point and 'does not invoke the interior private world of the dysfunctional family behind the palace doors'. This criticism is clearly based on her own interpretation of Sophocles' text. For instance, in the Programme Notes on her own production, Montgomery says that 'Sophocles creates a bell jar of suffering in his dramatic concentration on Electra'. (Montgomery's own response to creating Electra in the Compass Theatre Company's 1999 production of Sophocles' Electra is also documented in the Research Project, DB ref. no. 989.)
The strong opinions expressed in debates about contemporary relevance recall Patrick Rourke's distinction, in his Review of Auletta's production of Aeschylus' Agamemnon (1994, DB ref. no. 851) between 'creative and destructive anachronism'. Creative anachronism could, Rourke argued, facilitate understanding across time and space, whereas he thought that destructive anachronism laboured contemporary relevance, directing the audience's response in a way which could disrupt the relationship with the original play.
In contrast with Montgomery's response to the 1997 Electra, Peter Stothard in The Times (7 November 1997) praised the faithfulness with which Sophocles' 'subtle and balanced' re-interpretation of the myth had been realised by director and translator. Stothard stressed the affinities between the experiences of the ancient audience and those of people in Bosnia and linked Leveaux' programme notes about the impact on him of the sufferings of Bosnian children with universality of reference of the themes of family slaughter and revenge.
The difference in standpoint between the Reviews by Montgomery and Stothard, both in significantly different ways knowledgeable and thoughtful about the history of Greek tragedy in performance, was reflected in their comments about the translation of the conventions of Greek tragedy into modern theatre, their approach to psychological realism as a criterion of judgement and their verdicts on individual roles. Montgomery saw Clytemnestra as 'elegant but uninteresting'; Stothard perceived her as 'almost as magnificent as Electra herself'. Taken together and in conjunction with other sources, these Reviews are valuable and stimulating, yet each in isolation has limitations as a primary source for documenting the performance. (Truly, let the reader of Reviews beware.) Therefore, in quoting briefly from Reviews in the database we have tried to give sufficient indication of the tone and approach, to inform database users' decisions about the need to consult the complete Review. Where relevant we are providing a direct link to an electronic publication.
In documenting productions for the Research Project I have been testing the possible use of a Review template to ensure that information recorded for the Project includes comment on conventions (Chorus, Masks); that judgements about staging are backed up by information on size and type of playing space, design and lighting and music (programmes and theatre archives are notoriously slack in giving details of the music used, especially if it is presented in a recorded version rather than live). However, I have learned from bitter experience that what one 'sees' and 'hears' and records as objective description is inevitably coloured by the approach one brings to the play and the kinds of questions one asks of it. As illustrations of how the semiotics of staging can turn into a minefield, here are some extracts from Reviews of the recent Royal Shakespeare Company's Stratford production of Troilus and Cressida (1998/9, DB ref. no. 965).
Did we all see the same play, you may well ask. And of course the answer is that we did not. What is significant, however, is that all the reviewers identified virtually the same issues of interpretation and staging for analysis and discussion. All were aware that the audience's response was being guided by key aspects of design and acting style. They disagreed about how to interpret these but responded to the dynamics of the performance and staging.
My final example is from an academic journal. This is Herbert Golder's iconoclastic Review article 'Geek tragedy? – or why I'd rather go to the movies', part of a substantial section of the journal devoted to essays and Reviews of Greek Tragedy in Performance. Golder's Review voiced a deep disenchantment with almost all contemporary stagings, partly because he thought they were flawed but mainly because they were contemporary in a way which, he felt, privileged modern resonances and acting styles over the Greek. Although he claimed that 'I applaud contemporizing productions' (and cited the Polish Solidarity Antigone, the Hecuba performed in Dubrovnik while the city was under fire and the Sarajevo Alcestis and Ajax), his main assertion was that ' the ephemeral must never be allowed to occlude the essential' (p 199). His reasons for holding this position are several. Reminding his readers that Aeschylus' contemporary Phrynichus was fined by the Athenians for his play The Siege of Miletus because it was considered too topical and reminded the Athenians only of their own troubles, Golder claims that a certain kind of staging and performance style, stylized and culturally familiar, was needed by the Greeks for monumental productions with a life of one performance only. This, he thinks, is the necessary condition for custody of the 'essential' [sic].
As a consequence, he emphatically rejects insights from Noh drama (which he characterized as 'nipponising' on the part of the French director, Mnouchkine), just as he castigates the 'wooden' production of Medea, with 'a dowdy chorus of three', played for laughs (p 181), which he thinks was designed to feed the construction of 'Englishness' thought to appeal to Broadway audiences (directed by Jonathan Kent, translated by Alistair Elliott with Diana Rigg as Medea, DB ref. no. 168). The Royal Shakespeare Company's Theban Plays (1992, DB ref. no. 103) is described as 'a pastiche of Anglicized colloquial classicism' (p 177) while the 'dead hand' of British Shakespearian acting conventions is said to lie over everything. The particular criticisms may have a point but the underlying thrust of Golder's polemic becomes clearer when it emerges that his sympathy in the preservation of what he calls 'the essential' is with the gospel score of the Breuer/Telson Gospel at Colonus (DB ref. no. 188 & 996), no doubt because it was culturally familiar to him. Here, he writes, 'was the sound of the sacred that still engages our common culture, derivative and traditional but transformed' (p 184). Whose culture? And in what sense is it common, even within the USA, let alone more broadly in the English speaking world? Perhaps what Golder is trying to pinpoint is the culture which is common to those holding a particular view of what is 'essential' in Greek drama? He puts it thus – 'This medley of music, miraculous and mundane, brings us nearer to the experience of Greek drama – with its curious mix of sacred and secular, aulic and vernacular, lyric and iambic, human and divine – than a melange of eastern mysteries and French neo-classical acting' (p 184).
Golder's deliberately provocative Review article is an example of the increasing involvement of scholars in performance criticism, whether in academic journals, in literary and theatre periodicals or in broadsheet Reviews. The scholarly perspective becomes yet another determinant of the content and judgement of a Review.
Some scholars have taken a more open view of the similarities and differences between productions and reception of the same play by different companies and by the same company in different theatrical, cultural and national contexts. For example, in her analysis of the critical reception of two plays by Timberlake Wertenbaker, Susan Carlson has examined ways in which different senses of cultural, sexual and national identity shaped ever-changing attitudes to what was presumed to be the same play.
The different use of Reviews made by Hartigan, Banning and Carlson represent strategies which are divergent in theoretical approach as well as in handling of Reviews as sources. While Hartigan seeks factual information about performance and records value judgements, Banning emphasises the location of value judgements in the broad framework of cultural politics. Carlson turns the focus back onto the plays and compares details of responses to particular plays in the light of her analysis of the shaping frameworks given by differing senses of identity on the part of the receivers.
The debate about Golder's article continues (with a vigorous response from Oliver Taplin in a subsequent issue of Arion). In a related vein, in opera and theatre criticism in the UK it is currently possible to see the beginnings of impatience with 'contemporizing' productions. The preferred alternative often seems to be what some call 'naturalist' productions, a concept which, in claiming that staging of Greek drama is most authentic when not tied to any particular time or place, sometimes masks re-appropriation of Greek drama to the supposedly neutral and unexamined assumptions which lie behind claims that the plays are 'timeless' (presumably both in provenance and reference?).
It is clear, therefore, that the Review as a genre has a cultural voice of its own. This voice speaks to the interpretative influence of the world outside the text of the Review, and of course to the world outside the text/performance of the play. It speaks initially to the target audience/readership (for the Review, not the play) and its tone and emphasis varies according to whether the Review is directed at a charmed circle of the theatrically or classically knowledgeable, or perhaps to the even smaller circle of those who are knowledgeable in both spheres. It varies also if the reviewer is seeking to demystify the play and/or the staging for those who do not know much about it. It is also shaped by whether the reviewer does this from a standpoint of his or her own ignorance, and is further informed by the assumptions the reviewer makes about the cultural referents which will be understood (from the BBC soap-operaEast Enders to the work of Mnouchkine). This of course implies an element of awareness of reader-response (and especially in the areas which the reviewer seeks to trigger). The narrative form and texture of the Review is shaped by this.
The factor which is so far missing in all the types of Review discussed above is a record of response from the audience – by which I mean the audience of the performance, not the sometimes idealised or imaginary audience constructed by the reviewer or the audience to which the Review is addressed. I was struck recently by the gap between a somewhat damning Review in the Scotsman (Joyce McMillan, February 1999) of the Compass Theatre Company's Electra (Sophocles) and the response of a member of the audience and her group of students The students actually wrote to the theatre in Musselburgh asking it to disregard the Review and to continue to offer a stage to Greek drama. Their lecturer wrote to me: 'to judge by the deathly hush everyone in the theatre found it as gripping and gut-wrenching as I did. At the end, you didn't want to clap – not because it wasn't good but because applause somehow seemed so trite and inappropriate'. 
This raises another acute issue, that of the relationship between reviewer and audience of a live performance. Even though the individual audience member may have a different reaction from the collective response, he or she is likely to have been affected by the atmosphere, attentiveness and amount of applause. The Reviewer, as an individual member of the audience may or may not be so affected. However, it is comparatively rare for such effects to be examined, or even registered in a Review.
There are thus three factors which make the theatrical Review problematic as a primary source for research on particular performances:
Of course, these reservations do not detract from the value of Reviews as a primary source for the study of longer term cultural shift, or from their potential for shedding light (in conjunction with other sources) on the impact of theatre on cultural consciousness, or from their significance as indicators of openness or resistance to theatrical innovation.
The Review remains a valuable source for broad cultural issues in Reception, and would be even better if analysis of Reviews was to be accompanied by some serious research on theatre audience response, but it does not offer an unproblematic narrative or description of staging and performance.
 For the purposes of this discussion performance may be defined as the play in the theatre, as opposed to the play on the page. However, performance is a category with multiple fields. Its richness is summarised in S. Goldhill's 'Programme Notes' in (edd) S. Goldhill and R. Osborne (1999) Performance Culture and Athenian Democracy Cambridge, Cambridge University Press, pp1-32, especially pp10-17.
To access database examples given in this paper, choose 'search the database' from the side menu. Proceed to the Database and then choose 'modern titles' and type in the title of the play you are interested in. One or more plays of that title will then be listed. Click on the DB reference number (listed beside the title of the play) that matches the one in this paper in which you are interested. The details of that play will then appear. You may prefer to complete your reading of the paper before entering the database.
 For discussion of theoretical relationship between narratology and the Messenger Speech see Irene J.F. de Jong, Narrative in Drama : the art of the Euripidean messenger speech, Leiden, Brill, 1991.
 Y. Banning, 1997, '(Re)viewing Medea : cultural perceptions and gendered consciousness in reviewers' responses to new South African Theatre', South African Theatre Journal II, 1 and 2, May/September pp 54-87.
 Banning p. 56.
 For an overview of the relationship between modern staging of Greek drama and the struggle for freedom in South Africa see Margaret Mezzabotta 'Ancient Greek Drama in the new South Africa', in Selected Proceedings of the 1999 January Conference on Theatre: Ancient and Modern (forthcoming).
 I thank Professor Conradie for generously allowing me to consult his work in progress and for providing English translations of Reviews which were originally published in Afrikaans.
 'Timelessness and Timeliness: Anachronism in the Performance of Greek Tragedy', Didaskalia vol. 3 no 2.
 Stothard, the editor of the Times took a Classics degree at Oxford University and lists his recreations as ancient and modern literature: Montgomery is both a practitioner and an academic, having held a research fellowship in the performing arts at the University of Cambridge.
 For the importance of music as a constituent of the relationship between the text and the staging, see John Chioles (1993) 'The Oresteia and the Avant-Garde : Three Decades of Discourse', Performing Arts Journal 45, 1-28, especially pp 4-5 and 16-17.
 Ruth Hazel has suggested to me that because the visual is so important to us, as an audience we try to 'locate' the play by the messages its design seems to send out. When the design is eclectic (whether by tradition in the RSC, or as a result of an intention to suggest multiple resonances and obscure boundaries) then 'readings' of the performance are bound to vary.
 Arion, Third Series 4.1 Spring 1996 pp 174-209.
 As in Les Atrides, DB ref. no. 152.
 Almeida Theatre London, 1992 and Longacre, Broadway 1994, DB ref. no. 168.
 Susan Carlson, 1993, "Issues of Identity, Nationality and Performance : the Reception of two plays by Timberlake Wertenbaker", New Theatre Quarterly no. 35, August, pp267-289. The two plays discussed are Our Country's Good, and The Love of the Nightingale, which is derived from the myth of Procne, Tereus and Philomele.
 Arion, Third Series, 5.3, Winter,1998.
 An example is Tom Rosenthal's polemical Review article on recent opera productions in Prospect, April 1999, pp 62-64. Rosenthal accepts the validity of some 'modern dress' productions but laments the loss of fidelity to the original which has resulted from some attempts to affirm modern relevance or to transplant the action into another time or place - "it is particularly annoying when a superb musical performance is let down by an arrogant production. Katie Mitchell's Welsh National Opera Jenufa autumn 1998 is a case in point. This tale of 19th century Moravian peasants was placed in 1930s suburbia with Parker-Knoll type armchairs, electric bar fire and cloche hats. The two young men in the opera were heroically sung by two burly, middle-aged, balding tenors who were, absurdly, denied wigs and proper make-up. Worst of all Mitchell grafts a scene of her own on to the end. This is an outrage to Janacek, who provided a far more moving ending of his own" (p 63).
 By the term 'cultural voice' I refer to the fifth of the voices analysed by Roland Barthes in his discussion of the five voices which define the contours of literary texts. The four within the text are: hermeneutic, sign system, proairetic (e.g. structural links in the plot) and symbolic. The cultural voice links the text with the cultural universe which is external to it, but of which it represents a part.
 By reader-response I mean the kind of analysis of the relationship between the reader and the text developed by W. Iser, The Act of Reading: A Theory of Aesthetic Response, Baltimore, Johns Hopkins University, 1978.
 Letter, Margaret H. Williams, March 1999.
 An earlier version of this discussion was presented at the April 1999 meeting of the Classical Association. I am extremely grateful for the insights and suggestions of all those who participated in the ensuing discussion. I also thank the referees for this publication for their criticisms and advice and have also especially profited from the comments of my colleague Ruth Hazel.