Description
HR Study on Competing Rationalizes in the Diversity Project of the UK Judiciary The Politics of Assessment Centers:- The Golden Triangle designates the confluence of the Ruak River and the Mekong River, since the term has been appropriated by the Thai tourist industry to describe the nearby border tripoint of Thailand, Laos and Myanmar.
HR Study on Competing Rationalities in the Diversity Project of the UK Judiciary: The Politics of Assessment Centers
Geraldine Healy
Queen Mary, University of London, UK
Gill Kirton
Queen Mary, University of London, UK
Mustafa Özbilgin
University of East Anglia, UK
Franklin Oikelome
University of Hull, UK
Abstract
Assessment centres (ACs) are increasingly used to recruit highly qualified staff, yet there are few attempts to appraise their introduction from a diversity perspective. This article addresses this gap by exploring the introduction of ACs as a means of increasing the diversity of the judiciary in the UK. It analyses the underpinning politics guiding the processes and subjective experiences of ACs using Weber's identification of types of rationality and argues that substantive rationalities drive the formal rationality of the AC.The study demonstrates the theoretical value of incorporating diversity into the rationality framework and shows that the design, management and implementation of ACs are shaped by the competing rationalities that surround and imbue judicial work.While ACs may be introduced to demonstrate fair selection practices and challenge the reproduction of unfair discriminatory practices, such formal interventions are inevitably partial and bound by the distinctive institutional context and the societal context of inequalities.
Corresponding author: Geraldine Healy, Centre for Research in Equality and Diversity School of Business and Management, Queen Mary, University of London, Mile End, London E1 4NS, UK. Email: [email protected]
2 Keywords
Human Relations XX(X)
assessment centres, careers, diversity, ethnicity, gender, judiciary, rationalities, selection,Weber
Introduction
In England and Wales, judges are selected from among the legal profession - barristers and solicitors. The unrepresentative demographic composition of this group of professionals, in particular the predominance of white, Oxbridge-educated males, has long been, and continues to be, of public concern (CJA, 2005). This concern has now permeated the political agenda; institutional change, in the form of proactive efforts to increase diversity, is sought against a backdrop of significant changes in the last 30 years in the legal profession overall. Available monitoring data indicates that there has been a marked increase in the proportion of women and black and minority ethnic (BME) lawyers (Malleson, 2000).1 Nevertheless, it is argued that there remains deeply entrenched gender and race inequalities within the legal profession that are attributable to approaches to recruitment and selection and to the androcentric organization of careers and life in legal institutions (Rolfe and Anderson, 2003; Sommerlad, 2002). Against this background, the Department of Constitutional Affairs2 (DCA), the government department with responsibilities for upholding justice, rights and democracy (with particular responsibilities for running the courts, and improving the justice system), undertook a programme of work aimed at 'modernising the process for appointing judges with the aims of making it more open, transparent, and increasing diversity in the judiciary without departing from the key principle of appointment on merit' (DCA, 2004). The perceived need to widen the diversity of the judiciary was highlighted in a DCA consultation paper, which stated that:
If the make-up of the judiciary as a whole is not reflective of the diversity of the nation, people may question whether judges are able fully to appreciate the circumstances in which people of different backgrounds find themselves. (DCA, 2004)
Part of the modernizing programme was to introduce assessment centres (ACs) for the selection of the judiciary (see Healy et al., 2006; Kirton and Healy, 2009). This article examines the politics and operation of ACs and their contribution to building diversity. We draw on Weber's types of rationality to explore how different actors shape and respond to the political imperatives of an institutional intervention with respect to diversity. Specifically, this study explores how different and competing rationalities shape the way that ACs for the selection of the judiciary in England and Wales impact on diversity.
Competing rationalities and the diversity project
We start from the point where there is an identified need for an intervention that is designed to lead to a particular change (increase in judicial diversity) in a complex and traditional institutional context and that this need is driven by a political imperative. We use Weber's four-fold rationality framework to illuminate our understanding of judicial ACs to allow us to grasp their multidimensional nature and to reveal how multiple actors
Healy et al.
3
working in a political context affect human resource management (HRM) practice. Moreover, such technical approaches distract our attention from the distinctiveness of the organization being studied. As Townley (2008) states with respect to bureaucratic emphases in organization studies, the substantive qualities of specific organizations become of less importance. While using rationality as an analytical concept to understand ACs and diversity meets these concerns, it is not unproblematic. We are cognizant of Kalberg's claims that Weber's discussion of rationality is 'scattered and fragmented' and 'more likely to mystify than illuminate' (1980: 1146). There is little wonder that there is often confusion about the meanings of rationality. Like Townley (2002) we draw on Kalberg (1980) who sought to reconstruct at the purely conceptual level, Weber's vision of a multiplicity of rationalization processes that variously conflict and coalesce with one another at all societal and civilizational levels with the discrete types of ratio- nality constituting the cornerstones for these rationalization processes. Kalberg considers Weber's four types of rationality (formal, substantive, theoretical and practical) in relation to his four types of social action. Moreover, rationalization processes take place each independently of the other and at their own rate (Kalberg, 1980). Formal rationality is characterized by an orientation of action to formal rules and laws and is tantamount to a rejection of all arbitrariness so that a means-end rational calculation is made by reference back to universally applied rules, laws or regulations (Kalberg, 1980). The action emerging from formal rationality will be goal oriented. According to Weber, 'there is a dominance of a spirit of formalistic impersonality: ''sine ira et studio'' without hatred or passion . . . everyone is subject to formal equality of treatment; that is everyone is in the same empirical situation' (1978: 225). The implied objectivity and impartiality of formal rationality for the diversity project is clear. Substantive rationality is a preference for certain ultimate values (Townley, 2002). For Weber (1978), the concept is full of ambiguities and there are an infinite number of possible value scales for this type of rationality. He states that some may involve elements of social justice and equality, whereas others are criteria of status distinctions or of a political unit. This will result in value rational action. Thus, the value spheres that will come into play with respect to judicial substantive rationality will emerge from, for example, political substantive rationalities, the professional substantive rationalities held by lawyers, the substantive rationalities associated with diversity projects and of course the dominant values of the society, rationalities that may be complementary or competing. Theoretical rationality involves 'a conscious mastery of reality through the construction of increasingly precise abstract concepts' (Kalberg, 1980: 1152). For Townley (2002), it is a cognitive template that informs understandings of how the world works and informs action on this basis. Thus, formal rationality means will be informed by insights from theoretical rationality. Practical rationality accepts given realities and calculates the most expedient means of dealing with the difficulties they present (Kalberg, 1980). Thus, it guides actions in daily routines, providing patterns of action for the expedient means of dealing with immediate practical difficulties (Townley, 2002). Townley (2002) argues that it is substantive and formal rationalities that inform the rational action familiar to organization theorists. In her study of museums, she found that that there is an underlying tendency for formal rationality to undermine substantive rationalities.
4
Human Relations XX(X)
Focusing on tension between substantive and formal rationalities, Brubaker postulates that 'the antagonism between formal and substantive rationality may be interpreted as a tension between conflicting values: between calculability, efficiency and impersonality on the one hand and fraternity, equality and caritas on the other' (1984: 41). Furthermore, he argues that the tension is not only a tension between conflicting values, but also a tension between competing interests and between the groups that are the bearers of these interests (Brubaker, 1984). Moreover, for Weber, rationality is a relational concept; 'a thing can be rational (or irrational) only from a particular point of view, never in and of itself' (Brubaker, 1984: 35). Selection methods are imbued with formal rationalities. It is assumed that if procedures are followed, the 'end' will result in the 'right' person being selected. Indeed formal rationalities (with the end of being seen to be fair) drive HRM techniques that seek to subdue substantive rationalities that may impede the diversity project. Such impeding substantive rationalities may not be of caritas but of exclusion. Moreover, different institutional actors may simultaneously promote competing substantive rationalities (of fairness and exclusion). As Weber states, 'many possible meanings of the concept of rationalization will often enter into the discussion' (1978: 30). Selection processes have long been saturated with competing rationalities. Research undertaken by Collinson et al. (1990) in the 1980s highlighted the high degree of informality in the setting and content of selection interviews that was prevalent at the time and that facilitated discriminatory recruiter behaviour. The growth of scientificity (see Townley, 2002) associated with HRM was in part a response to the kind of criticisms levelled by Collinson et al. and it also explains the rise in popularity of highly formalized ACs, at least for the recruitment of highly qualified workers (Dayan et al., 2002; Garavan and Morley, 1997; Iles, 1992). The rationalization of selection in advanced capitalist societies is sharply evident in ACs. An AC can be defined as 'a structured combination of assessment techniques that are used to provide a wide-ranging, holistic assessment of each participant' (Garavan and Morley, 1997: 160). The common features of ACs are: assessment in groups; assessment by groups; the use of multiple techniques (e.g. ability tests, peer evaluations; performance tests); the use of situational techniques (e.g. role-playing exercises); assessment along multiple dimensions (Garavan and Morley, 1997). The use within ACs of simulated experience and multiple methods of assessment is expected to provide a proxy for subsequent job performance and therefore a more objective and valid basis on which to make appointment decisions (Dayan et al., 2002; Iles, 1992), thereby overcoming some of the capacity for discrimination associated with less structured and less rounded approaches. Formal rationality shapes the dominant liberal equality perspective where 'equality of opportunity exists when all individuals are enabled freely and equally to compete for social rewards' (Jewson and Mason, 1986: 307), and where greater proceduralization is meant to lead automatically to greater fairness and equality. The problems identified with ACs include the malfunctioning of the system with respect to planning, job analysis, quality of exercises, the assessors, candidate preparation, poor documentation and scoring (Caldwell et al., 2003); issues of validity and measurement problems (Jackson et al., 2005); their overly complex and onerous nature (Dayan et al., 2002). Underpinning these critiques is a belief in formal rationality that assumes that with the correct means ACs will 'work'. A desire to understand the macro-politics of recruitment and selection (e.g. the wider social, political and labour market contexts, including discrimination and
Healy et al.
5
inequalities) is largely absent. Whereas Young argues that unfair discrimination is not the only or primary wrong that groups suffer in so far as discrimination is an 'agentoriented and fault-oriented concept' (1990: 195). The logical formal rational action is to identify the person/organization responsible; remedy the procedural faults and the end will follow, that is, discrimination is eliminated and greater equality is achieved. This is the HR practitioner's or policy-maker's preferred problem-solving formal rationality oriented approach. Young (1990) contends that the problem is more deeply embedded and states that 'oppression, not discrimination, is the primary concept for naming grouprelated injustice'. Judges, the focus of this article, have historically been recruited primarily via secret 'soundings' (consultation) with the existing judiciary followed by a conventional interview. More emphasis was placed on 'who you know', rather than on 'what you know'. ACs are therefore, despite weaknesses, a significant innovation in the context of the judicial appointments process. The DCA anticipated that the introduction of ACs would serve the twin objectives of increasing diversity in the judiciary and increasing public and candidate confidence in the appointments process (DCA, 2004).
Research methods
Our research investigated each element of the AC selection process for two judicial 'competitions' - one for recorder and the other for deputy district judge (DDJ) appointments3 in 2005. This involved: exploratory investigations of i) application documents given to candidates, guidance to panels, training materials; ii) candidate documentation (application forms, assessors' mark sheets, written tests); iii) reviewing AC exercises, that is, role play, written and technical papers and the interview; iv) meetings with consultants who were engaged to aid with the design of the above competitions and DCA civil servants involved in the process. b) semi-structured interviews with 39 judicial candidates (11 BME women, eight BME men; 10 white women, 10 white men) and 10 assessors (four white women, six white men) in order to understand assessor and candidate perspectives and experiences of ACs.4 a) Both assessor and candidate interview schedules were informed by a combination of our interrogation of the documents and scoping interviews with DCA officials and consultants and our understanding of the HRM and diversity literature. The candidate interview schedule used open questions to explore perceptions, experiences and opinions of the equality and diversity aspects of the AC processes and activities in relation to predetermined competencies that the AC was designed to assess (see Figure 1). An indepth discussion of competencies may be found in Healy et al. (2006) and Kirton and Healy (2009). The assessor interview schedule included questions on the assessor role, assessor training, the sift process, the consultation process, the different elements of the AC and general points on enhancing the diversity of the judiciary.
6
Human Relations XX(X)
Each interview lasted approximately one hour. The interviews were digitally recorded and transcribed verbatim. Most interviews were face-to-face and took place in London, Bristol and Manchester. The interview data were analysed using a qualitative data software program (Nvivo).5 The initial analysis was guided by the research objectives and reported in Healy et al. (2006). The rationality framework emerged from our post hoc deeper engagement with the multiple forms of data and provided the means for an analysis of the politics and complexities of ACs. Quotations are used to illustrate these complexities alongside relevant quantitative data. We draw on quantitative data from a 2005 DDJ competition using ACs. This competition attracted 386 candidates, 43 percent women, 12 percent from BME groups (of whom women were a greater proportion).
Analysis of multiple forms of data
Five propositions are used to analyse our data in the context of four rationality types. In line with Yin's (2003) view, the propositions were helpful in relating the findings to our
Independence and professionalism: o Technical knowledge and expertise o Integrity and independencea o Showing authority o Developing knowledge o Managing workload o Managing selfb Judgement o Weighing and analysing o Resolving and deciding People skills o Communicating o Building relationships
Figure 1. The nine competencies
a b
Used only Recorder competition. Used only in DDJ competition.
Table 1. Judicial applicants to deputy district judge competition by sex and ethnicity Applicants Male Female White BM E Ethnic origin not disclosed N DDJ % 57 43 72 12 15 386
Healy et al. theoretical perspective and to focus the discussion within feasible limits. While we introduce propositions in the context of four rationality types, it should be remembered that in practice these interrelate and overlap.
7
Formal rationality
The implied objectivity and impartiality of ACs associated with formal rationality (and the liberal equality perspective) should lead to equality of treatment of applicants and, all things being equal, lead to the desired end of greater numerical diversity in the judiciary. Thus there is a political instrumental imperative behind the ACs' introduction. This rationality was shared by some of the assessors and applicants, but not unproblematically as will be seen in the next sections. Formal rationality because of its means-end focus partially lends itself to statistical analysis.
Proposition 1: Judicial ACs will lead to greater numerical diversity in the judiciary.
Substantive rationality
The judiciary as a body is underpinned by substantive rationalities of justice and fairness. Moreover, it is run by the body politic that in turn, has political substantive rationalities with respect to diversity. At the same time, members of the judiciary as people will hold substantive rationalities , to a greater or lesser extent reflect their professional and social backgrounds. These rationalities may influence their judicial decisions and have an uneven impact on different groups in society (see Healy et al., 2006). The drive to enhance the diversity of the judiciary was partly to introduce greater awareness and understanding of diversity to the judiciary. Moreover, public substantive rationalities may shape the perception of the judiciary with respect to diversity. Awareness of diversity in the AC is built into the Building Relationships competency and into the Communicating competency. Substantive rationalities also rest in the integrity and independence competency. These are assessed in the interview and the role play.
Proposition 2: The substantive rationality of assessors will shape the diversity of the judicial appointments.
Theoretical rationality
Kalberg (1980) states that theoretical rationalization processes may also be carried out by judges who interpret the incipient world view found in political constitution. The corpus of knowledge of the judiciary is laid down in legislation and case law. It has formed the basis of candidates' professional training and practice. While it may be the outcome of substantive political rationalities, it is a theoretical rationality to be acquired by the legal profession. Unlike the museums in Townley's (2002) study where tacit knowledge is central, judicial work is codified and outputs in the form of judgments are
8
Human Relations XX(X)
evident. Nevertheless, based on case law, judicial work also draws on tacit knowledge in the interpretation. Assessors seek to identify candidates' theoretical knowledge by assessing the competency of technical knowledge and expertise, weighing, analysing and resolving and deciding and developing knowledge.
Proposition 3: ACs will allow theoretical knowledge in written tests to be assessed regardless of sex and/or ethnicity.
Practical rationality
Expediency is a central aspect of practical rationality that always indicates a diffuse tendency to calculate and to solve routine problems by means-end rational patterns of action in reference to pragmatic self-interests. In judicial work, there is an important pragmatic element to its practice. The following quotation from a DDJ judge (also an assessor) illustrates this effectively:
District Judges and DDJs are dealing with people at a very vulnerable level in their life very often . . . there are two decisions which you can take that have the most effect upon them and one, is to take their home away and the other is to take their child away. It's a paradox; the administration allow no more time to deal with [such cases] . . . Now what do you do? Do you cut the thing short or do you take the time that it needs to make sure you are getting your decision right? And if you take the time then you are perceived as not being - how shall I put it - not being a company man. Well, I take that particular one very seriously. . . when children are involved. And if you think that children will grow up and will always remember being thrown out of their home. I mean, that's written indelibly on their minds. And these are [important] decisions that you have got to take quickly. And if you've got people that are indecisive and they dither around, I am afraid you've got a real problem.
This quotation illustrates that workload and time constraints are central to judicial work. Judges are also faced with many diversity based practical rationalities. For example, do they give more time to a person whose first language is not English; how do the demands to make reasonable adjustment for a person with disabilities compete with the time demands outlined in the above quotation? Thus there may be a potential tension between the practical rationalities in the daily work of a judge, and the different needs evident in a diverse society. The AC seeks to assess practical rationality by assessing the competencies of showing authority, managing workload and managing self.
Proposition 4: Candidates will be confident that diversity issues are treated fairly and consistently.
Proposition 5 links the Propositions 1-4 and reflects the second DCA objective to increase candidate confidence in judicial selection processes.
Proposition 5: ACs will not advantage one group of judicial applicants over others.
Before considering each of the propositions in turn, we outline the context of judicial ACs.
Healy et al.
9
Objectives and features of ACs for judicial appointments
The objectives of the judicial ACs were established to:
(i) introduce a 'one-day' selection system, that provides a range of different activities designed to test the criteria for appointment in a fairer and more comprehensive way than interview alone; (ii) collect more information about each applicant, particularly those who receive little or no comment from consultation [with serving judges], on which the decision to appoint or reject can be made; (iii) collect information about applicants using a range of different activities designed to test an applicant's abilities against the criteria for appointment . . . to make decisions on appointment in a fair, systematic and consistent way; (iv) to provide more helpful and specific feedback to the Lord Chancellor about an applicant's performance at the AC on which he can make decisions on appointment; to provide more helpful and specific feedback to unsuccessful applicants about their performance at the AC. (DCA internal report, cited in Healy et al., 2006: 38)
The ACs were tailor made for judicial appointments and consisted of a written technical legal test, written analysis of case studies, practical exercises (role plays) and an interview. For the role play exercises, each candidate, observed by assessors, takes on the role of a judge to handle cases and make decisions. These activities were d e s i g n e d t o s i m u l a t e t h e k e y e l e m e n t s o f j u d i c i a l w o r k a n d w e r e r e l a t e d t o t h e p r e d e- termined competencies. Assessment of all parts of the AC is recorded in writing on forms designed for each activity in order to ensure that decisions are justified. In addition, interviews and role play exercises are tape-recorded - this is meant to increase candidate perceptions of procedural fairness. Assessor panels consist of a serving or retired senior official from the DCA, a lay person (with a human resources background or significant experience of interviewing) and two judicial office holders. In addition, the policy is where possible to have a mixed panel to include male, female and BME members. The broad composition of the assessor panel arguably democratizes the selection decision-making processes and meets a fundamental condition of procedural justice (Young, 1990). It prevents a concentration of power over judicial appointments within the established judiciary (as in the 'secret soundings' approach). All assessors undergo compulsory training in 'sift', inter- view and competency policies and procedures. Mirroring Jewson and Mason's (1986) liberal perspective on equality, most assessors placed great faith in the theoretical ratio- nality of their training to ensure the fairness of the formal rationalities of the AC proce- dures and processes. The design of the process was undertaken by consultants and based on theoretical knowledge of ACs and formal rationalities of the means and ends of the process, whereas, the decision to introduce the ACs was imbued with substantive rationalities underpinned by a complex set of political and equality values.
Findings Proposition 1: Judicial ACs will lead to greater numerical diversity in the judiciary
If Proposition 1 is supported, it will mean that women and/or BME applicants are more likely to be recommended for appointment than white men. Appendix 1 shows that out
10
Human Relations XX(X)
of the 386 candidates, a total of 211 (54.3%) candidates attended the AC, and 78 (37%) were recommended for appointment. The DDJ data were examined to see whether there were any significant differences in performance at sift and at the AC for different groups according to sex and ethnicity. The findings are summarized in Appendix 1. Sex appears to influence overall success at both sift and the AC.6 Women are more likely to be successful at sift and also more likely to be recommended for appointment following the AC than men. Ethnic group did not appear to have any influence at sift7 nor at the AC.8 Further analysis was carried out on the DDJ Competition data to find out whether the influence of sex on the sift and recommendation is moderated by ethnicity. The results are shown in Appendices 2 and 3. Appendix 2 indicates that at sift, the influence of sex is significant both for white and BME applicants,9 although this does not appear to be the case at the AC for both groups.10 In other words women of both BME and white background are more likely to be successful at sift than their male counterparts but are not more or less likely to be successful at the AC. When sex and ethnicity are combined as categories, BME women may have greater success rates at sift than BME men. At the sift stage of selection against the competencies, the data show that women and BME applicants were more successful and that BME women are more successful than BME men. How do we interpret these data? One interpretation might be that the more open and transparent process not only attracted a wider pool of candidates, but also enabled those candidates to demonstrate their suitability using the predefined competencies. In other words, the AC would reflect its underpinning formal rationality. An alternative interpretation might be that the political agenda to increase the diversity of the judiciary had influenced sifters to look more seriously at candidates whom they might previously have rejected. However, of those appointed analysis reveals that women are more likely to be successful but that ethnicity is not significant. Why would this be? Theoretical rationality explanation would be that the system has identified the best candidates. However, it might point to the greater propensity to treat BME candidates as 'other' (see Young, 1990: 60) and an inability to demonstrate sensitivity to diversity that results in BME candidates having greater difficulty demonstrating their competencies, as we see to some extent in this article. Our results do not show unequivocal support for Proposition 1. Some caution is necessary in assuming we can generalize from these results. We have examined only one competition and the numbers for different categories are small. The social context of these competitions is outside the terms of this article. However, statistics show that BME candidates are likely to apply in proportionately greater numbers to lower level courts (e.g. DDJ) than to the higher courts, which have tended to retain their traditional gender and ethnic composition.11 Therefore for Proposition 1 to be properly tested, the exercise should involve a greater number and a diversity of competitions.
Proposition 2:The substantive rationality of the assessors will shape the diversity of judicial appointments
Assessors bring their substantive rationalities into all stages of the assessment process but this is intensified in the interview and the role plays. Our findings revealed a complex picture of contradictory substantive rationalities characterized by both caritas and exclusion.
Healy et al. The Assessor Guide on interviewing states:
11
Most discrimination occurs unwittingly. However unbiased interviewers believe themselves to be, it is natural to hold certain preconceived ideas as to how other people are likely to behave. In the interview open minded objectivity is vital in order to minimize the risk of bringing bias and prejudice into the process. In general this means that questions should not be asked of one group of candidates (e.g. women or members of ethnic minority groups) which are not asked of others. Questions about a candidate's domestic arrangements should be avoided altogether . . . Questions which may be misconstrued by candidates on equal opportunity grounds (or which may give offence) must be avoided.
This advice reflects the nature of formal rationality and of the liberal equality perspective, that is, that all candidates should be treated the same. In addition, assessors are provided with a set of possible interview questions, but are given discretion in the choice of questions posed to individual candidates and in the way that questions are asked. This gives assessors control over the perceived 'best way' to identify competencies. An exam- ple of how such control was used to shift the direction of an interview was given by one assessor who talked about his feeling that encouraging candidates to bring in examples from broader life experiences (as the competency approach allows) might actually give women an unfair advantage over men. He explained:
In [assessing the competence] 'building relationships' . . . sometimes you get a better response from women candidates particularly when you ask, 'members of the judiciary have sometimes been criticized for being too distant, do you agree with that?' Then the follow-up could be 'what steps do you take to ensure that you keep in touch with people?' Usually then women come out much better [than men] and that may well be through school, with children, social groupings, and interaction with people in the community. I am usually conscious of that and if that is the case, I try to get them off that and move them on somewhere else because that's advantaging them.
Such gendered value-based redirection was not found when candidates attempted to demonstrate their competency by way of more traditional work based examples. This illustrates how assessors can control the interview and therefore how they can actively shape the outcome; that is, if a candidate is prevented from demonstrating (gendered) strengths, then it is hardly surprising if their rating is low and in turn a low rating justifies non-appointment. Since judicial ACs are shaped by HRM approaches to diversity management, underpinned by principles of formal and theoretical rationality, there may be an expectation that there would be transparency about questions used to evaluate candidates' competencies. Yet some BME candidates recounted questions that seemed unrelated to any espoused competency; for instance, a question was asked about the candidates' view of racial disturbances in a particular area. Asking extraneous and even politically charged questions of one group of candidates enables interviewers to impose their substantive rationalities on the outcomes of interviews - 'inappropriate' answers clearly attracting low ratings. This clearly deviates from the formal rationality of the AC. Further, some assessors conveyed their own substantive values about the roles BME people should play in the community. Their ethnocentric assumption may be that candidates' ethnicity overruled their professional middle class status, for example:
12
Human Relations XX(X)
Well, again, if you ask them what they are doing in their community because, I think of an Asian candidate, telling me he was a member of Rotary wasn't actually what I was looking for.
The fact that the candidate did not play out the stereotype of the Asian person solely rooted in their ethnic community (rather than the middle-class England of their white professional contemporaries) meant that the assessor's value based substantive rational- ity might militate against awarding a high competency rating. Thus, paradoxically, the agenda to build diversity could work against some BME candidates who do not play up their ethnic difference. Whereas another assessor recognized the intersectional connection between class and ethnicity but viewed class as overriding ethnic disadvantage:
I mean most of them [candidates] weren't the sort of black people that I would meet in my Court. They were highly educated. Some public school, grammar school, graduates so that they weren't - they weren't ordinary and they'd never had to struggle with diversity because of their colour in society. It had not affected them.
Thus the one assessor would see BME lawyers as 'other' in the sense of having entirely different experiences than white lawyers. The other saw them first and foremost as middleclass professionals who would have no connection with the black people who appear in his court, in this case denying the intersectional identities and disadvantage that BME professional workers experience in life and work in Britain (Heath et al., 2008; Oikelome and Healy, 2007; Skellington, 1996). Crucial to these different substantive values are the potential interpretations that result. In contrast, assessors also demonstrated the caritas associated with Weber's substantive rationality. Scrutiny of assessors' documentation exposed evaluations of candidates' insensitivity to diversity, for example in assessing 'building relationships' an assessor wrote '. . . his descriptions of some clients suggested either personal distaste or naïve surprise at other people's ways of life'. This was reflected in the low mark allocated for this competency. Turning to the role-play exercises, when assessors talked about their evaluation of different 'types' of candidates it was clear that some gender stereotyping occurred, but the outcome for candidates was not necessarily negative:
Often you get a complete surprise. For example a woman can be really interventionist, really assertive and you get a male candidate who is very sensitive, empathizes, caring. I found that with them all, you can never predict by the sex of the candidate that that's how they are going to play it.
Again, a complete mixture of response . . . many [women] are very, very good; quite surprising really. Some of them really empathize. But you would expect that, from their experiences and so on, you would expect them to be good.
Assessors' lack of understanding of stereotyping was evident; sex or ethnicity seemed to foreground applicants who were not white male. Nevertheless, assessor judgments could both advantage and disadvantage candidates from under-represented groups and
Healy et al.
13
yet ostensibly assessors were adhering to the procedures stipulated in the AC guidance documents. The AC is a political process where gendered and ethnicized dynamics may play out. Thus, the power relationship between the lay and the judicial assessor shape the way that the formal rationality requirements are interpreted. Moreover, the applicant too may enter these power dynamics, when it can also take a gendered turn. An assessor reported:
My experience is that they [the applicants], especially if there is a man and a woman on the panel . . . will address all their comments to the man.
Influenced by their substantive rationalities, applicants may unconsciously augment a gender power imbalance and potentially undermine the importance of the woman assessor thereby creating an additional and unnecessary hurdle for the woman assessor to assert her authority and perhaps influence on the process. Those who were lay assessors were more likely to uphold substantive values of e q u a l i t y a n d f a i r n e s s a r o u n d t h e d i v e r s i t y a s p e c t s o f t h e A C , w h e r e a s t h e j u d i ci a l a s s e s- sors were constantly balancing the formal rationalities with the substantive rationali- ties absorbed from their judicial work and professional socialization. How are these differences in substantive rationalities resolved in a practice? On the one hand, we learn that:
Obviously the judicial member will carry greater weight, if he [sic] says 'I think it's very good the way they dealt with that subject'.
In the case where there was disagreement:
. . . we did have a chat round the table about the three different candidates and there was a bit of give and take then because we'd all made very comprehensive notes.
Thus, the parties seek a compromise acceptable to them all, the outcome of which may of course reflect the power relations between the parties. Therefore, the variability of assessor interpretations was not only informed by the formal and theoretical rationalities and their requirement to undertake the practical rationalities of the process, but also and crucially by the different substantive rationalities that the assessors held as mediated by power relations. Proposition 2 holds.
Propositon 3:ACs will allow theoretical rationalities in written tests to be assessed regardless of sex and/or ethnicity
At the AC, theoretical rationalities assessing the required competencies are assessed by a) a written paper in which applicants analyse case papers and document their observations, and b) a technical paper that is a test of knowledge of relevant jurisdiction or potential to acquire it. Criticisms of the tests were articulated by candidates
14
Human Relations XX(X)
(e.g. insufficient time; extent of preparation required), although there were no apparent gendered nor ethnicized differences in these comments. Critical comments were more likely to emerge from those who had not taken examinations for a long time and/or who worked in legal specialisms that do not require day to day engagement with a breadth of theoretical knowledge. Generally, candidates felt that the tests were important, objective measures of their competency and suitability for judicial appointment, which reinforced the formal rationality of that part of the AC. Candidates were divided on whether these were appropriate means to assess the corpus of legal knowledge necessary for the judiciary often depending on how well they had done in the assessments. A common comment was that they were not fully prepared for the content of such tests. Nevertheless candidates accepted that the competencies assessed by the written tests (in technical knowledge and expertise and weighing and analysing) are a necessary requirement for judicial appointments. Furthermore, we found no difference in perceptions of the legitimacy of these processes regardless of sex and/or ethnicity. However, differences between the candidates emerged with respect to the lack of anonymity on the papers. Women and BME candidates (but not white men) did identify concerns that candidate names rather than identifying numbers were on the paper, which is the common practice of UK universities. Assessors were divided in the possible implications of this practice. While some thought that its introduction might have the benefit that candidates would perceive that their paper was treated fairly; in essence the anonymizing process would be tantamount to a rejection of any arbitrariness in the way that theoretical tests would be activated through formal rationalities associated with universally applied rules, laws and regulations (as cited earlier by Kalberg, 1980). Other assessors thought that anonymity was unnecessary, reflecting a belief in the formal rationality of the process. They argued that the pressure to mark the papers within the day meant that it was unlikely that they would notice who wrote which paper and therefore unlikely to bias the interpretation of answers. In addition, the technical paper was multiple choice where knowing the name would not alter the outcome. Whereas, women and BME candidates were more likely to identify this as a questionable practice; this suggests their greater awareness of potential discrimination. The following BME woman applicant responded:
I don't think [they] should know your identity. I mean for the same reason that examiners should not be aware of your identity when marking papers.
Proposition 3 is unproven. We find that the content of written tests is not permeated with a gendered or ethnicized bias. However, it is possible that such bias could creep in to the assessment owing to the lack of anonymity of the papers.
Proposition 4: Candidates will be confident that diversity issues are treated fairly and consistently
One of the ends of the formal rationality of ACs is that they should be seen to be fair; propriety of questions is one of the measures of fairness in recruitment and selection (Gilliland, 1993). Just as the assessors identified unevenness in the type of questions asked, particularly 'diversity questions', this was confirmed in applicant interviews. On
Healy et al.
15
the one hand we noted some applicants felt that the process was very formalistic yet on the other they noted that there were no questions on diversity, for example:
. . . it was a very mechanical because they had set questions. So they weren't diverting too much. And I had no diversity questions at all. Which I had been warned about, you know, expect really difficult questions on diversity. I didn't have one single question. (BME applicant)
For such candidates it would be difficult to demonstrate their knowledge and experience of diversity issues. In contrast, a number of BME and women candidates had the impression that their experiences of diversity issues were valued. One applicant reported that her application was strengthened by the difficult career route she had navigated and because:
. . . the fact that I was a woman and I think, finally and lastly, that I was a black woman. It wasn't the first thing that was said - it was like at the end that this was because I would bring something different again, as a result of my own cultural background. (BME applicant)
This example is important as it turns on its head the belief that being a woman or from a BME community will have a negative effect on an application; on the contrary, for this woman, these personal attributes were seen as definite advantages. This perspective emerged from a number of candidates and it can be directly related to the political substantive rationalities that were espoused by the Lord Chancellor at that time. These espoused values gave underrepresented applicants confidence that their applications would be treated fairly. However, others felt uncomfortable with the confessional nature of the diversityrelated questions, giving this area of questioning a different nuance for them as opposed to white male candidates. For example:
I find it strange that they would even have to ask the [BME] person about diversity issues. I live it, I live the life. So I don't criticize the question, it was part of the process, but I think it is unnecessary. It's like asking someone who is in a wheelchair how they can relate to other people in wheelchairs, why on earth would you need to do that? (BME candidate)
AC candidates in particular have to give away vast amounts of personal information, thoughts and feelings (Stansfield and Day, 1998). Women and BME applicants suggested that there were more subtle and indirect lines of questioning that were trying to uncover the proximity of candidates to diversity issues, which could potentially be disadvantageous. For example:
You did rather get the impression that if you lay claim to actually being close to them [diversity issues] that was going to be actually better than just knowing or being informed [about them]. (Female candidate)
Despite the claim made in the candidate guidance notes that broader life experiences would be valued, a small number of women indicated that they felt, from the reactions of the panel (body language, facial expressions, etc.), that competencies gained from
16
Human Relations XX(X)
experiences, for example, time out of career to care for children, were not properly valued or understood, for example:
I think it is an important thing to bring up children. I think the experience teaches you an awful lot. I get the impression that they don't think that.
Some women attempted to demonstrate various competencies by reference to family life, but their subsequent time poverty meant they also were unable to demonstrate other competencies by reference to voluntary work, from which they sensed some disapproval. In such cases candidates reported a sense that assessors were trying to trip up candidates who might be demonstrating strengths in one area, by making attempts to get them to reveal weaknesses in others. In contrast, we noted women who felt that diversity was 'in' at the moment and they should play to that, even though they would normally never mention family and children in an application. For some the political rationalities influenced their actions:
I was busy mentioning that I was a woman with children, which I would normally never put on an application. Or I wouldn't try and shoehorn it in, but I felt that it was, I did it with some misgivings. But in the round, I felt that I should try and get that in somehow. (Female applicant)
Moreover, this woman was confident in her experience and sensed that the female assessor empathized with her and understood that with two children and a busy career at the bar, it would be difficult for her to do voluntary work. Just as analysis of the assessor interviews demonstrated that assessors may not share a unitary substantive rationality (see Proposition 2), the applicants' interviews demonstrated a wide range of interview experiences with respect to diversity. With regard to the role play exercises where practical rationality is important, the overall perception of candidates was mixed. Candidates were required to take the part of judges in a simulated court room scene with actors taking the parts of barristers, the accused and witnesses. One case involved an asylum seeker; who in the words of an assessor 'doesn't speak very good English, wears a headscarf and looks rather poor'. The assessor questioned whether applicants would be comfortable with this woman. Yet the candidates reported that their efforts to integrate diversity concerns, such as adjustment to disability issues or to language deficits in the role play were judged negatively by assessors. Thus, where candidates demonstrated practical rationalities in, for example, giving more time to a person judged to require it because of poor English; this was considered poor time management of the managing workload competency. Some female and BME interviewees considered the role plays to be a useful vehicle for displaying their competencies in a practical way. They argued that the role plays allow the applicants to put themselves in the role of a judge and demonstrate competencies such as resolving and deciding and showing authority, which cannot be easily measured through the other methods of assessment. However, several interviewees considered that potential variations in style of authority and decisiveness, adopted by women, men, minority and majority ethnic groups as well as across other social divides, could be an
Healy et al.
17
issue that might give rise to disadvantage or assessor bias. They feared that such differences can be misconstrued as weaknesses. Yet white males reported that when discussing diversity issues (where asked), the interview seemed to become more interesting:
I think perhaps I found the interview seemed to flow better when we were perhaps touching on issues regarding diversity and that's not assuming too much about the entire process but . . . they seemed to be far more interested in what I was saying there.
The above quotation suggests that (white) assessors felt more comfortable discussing diversity with a white male because it was about the 'other' rather than with the 'other'. A comment from an assessor may go some way to explain the difficulty of discussing what is perceived to be 'otherness':
No, I think you stay pretty clear of race. Perhaps . . . so as we didn't put our feet in it. We had a couple of lesbian applicants and we decided we weren't going to go anywhere near that field in case we upset someone and so the black applicants were treated no differently. They were asked about diversity and the responses were very much the same as the white candidates because they'd never thought about it.
The above comment suggests that the substantive rationalities of the assessors are limited by narrow insights into diversity and that they may be uncomfortable discussing diversity issues, other than in the abstract with the 'other'. This tendency is compounded by lack of clarity in a common understanding and indeed confusion of diversity and what the formal system was really requiring. There seemed to be a different expectation from BME applicants than white applicants and as stated above seemed to be difficult for applicants to untangle. Proposition 4 is not supported. Candidates revealed inconsistencies in the way that diversity was addressed and interpreted.
Proposition 5:ACs will not advantage one group of judicial applicants over other groups
From an applicant perspective, many felt that the AC did not advantage one group over others, but there was a small number of women and BME interviewees who offered critical reflections. For example, responding to a question about whether the AC approach advantages or disadvantages any particular group:
I think it might in certain respects. The people conducting the interview are white and middle class, to that extent yes. (Female candidate)
While the dominant political substantive rationality may be about widening diversity, many applicants were yet to be convinced that the officials of the AC process would share these substantive values.
18
Human Relations XX(X)
The perception of relative advantage or disadvantage may be influenced by applicants' personal biographies that will shape their own substantive values and their perception of the AC. Women, for example, reported past experiences of overt sexism. A striking example was an applicant who stated that she was told that she was not taken on by a set of chambers because 'We wanted somebody with the looks of Diana and the brain of Athena'. The interlocking nature of gender and racism was overcome by some applicants in their early education, for example:
I was told at school that I could never be a lawyer . . . I think at the time the expectations of a black girl were factory or secretary but I thought that's rubbish. I can do whatever I want if I've got the determination to.
Moreover, many applicants, particularly those educated in post-1992 institutions thought they would not get to the AC stage and were pleasantly surprised that their educational background had not appeared to disadvantage them, a view held by both successful and unsuccessful applicants. It is not only their educational background that has the potential to advantage, or disadvantage, them; applicants were also con- scious of a hierarchy of legal experience. Indeed this hierarchy may be ethnicized. The Commission for Judicial Appointments (CJA) (2005) sought to explain the divisions:
. . . graduates who do not attend the elite universities often have a more pressing need for an immediate income, and as a result, gravitate towards Social Security or Immigration law. Consequently, a disproportionate number of ethnic minority practitioners work in these areas - which are as complex and important as any others, but not as high profile, as highly thought of or, in the long term, as highly remunerated as commercial or Chancery Law.
The legal hierarchy (including some AC assessors) is imbued with material substantive rationalities. Such rationalities shape the value given to particular kinds of legal work. For example, pro bono work is an important way of gaining experience and of making a contribution to society, and this may be used as evidence in judicial applications, as an assessor explained:
Pro bono work is another thing; ethnic minority candidates do a lot of law centre work for which they give up their time freely really.
When asked how pro bono work was valued, it was clear that the substantive values driving the assessment of such work reflected the hierarchical status within the profession, that is, that which tended to reflect the white male majority.
Well, it's fine if it's an add-on but if they see it as their life's work and central then I would say it is not thought of as being of the same level. It's not dismissed as being irrelevant . . . but it's not comparable with somebody who is a high-flying and adored barrister in a £500,000 a year fee earning chambers.
Healy et al.
19
The above response reveals how status is closely associated with material rewards and imbued in substantive rationalities. Lawyers from lower status (i.e. lower paid) legal work may find it harder to convince assessors of their professional competence than lawyers in high paid legal work. The consequence of this hierarchy and indeed of the legal fragmentation imposes additional burdens on those lawyers from less privileged specialties:
Because I think that it would generally be people who aren't from ethnic minority and from smaller practices where they would not have as much resources available to them or mentors and it would be difficult for them to prepare for the process than for someone who comes from a different background. (BME candidate)
Legal hierarchies and their associated substantive rationalities will also shape the AC context. A number of applicants reported how they used their networks to demys- tify the judicial selection process. The benefits that accrue to those in more privi- leged positions were spelt out by a white male barrister who is in court two or three times a week:
I am in court sometimes two, three times a week . . . to be seen by quite a few judicial styles and I know quite a few judges socially.
His experience exposes him to different judicial approaches and importantly puts him in the social network of a number of judges from whom he will undoubtedly gain a confidence not available to those in other sectors of the law profession. Moreover, he will gain insight into the practical and theoretical rationalities of judicial work and become aligned with the substantive rationalities of the judicial mentors. The evidence indicates that the fragmentation of the profession results in uneven access to professional, social and time resources to give to the AC preparation and presentation and therefore some judicial applicants would be better able to demonstrate the competencies in a more confident manner than those from less privileged backgrounds. From Proposition 2, it is evident that assessors will bring their substantive values to the AC. This raises the question as to what attempts are made to ensure these values are aligned with the espoused political rationality to build a more diverse judiciary to ensure that all groups are treated equally. Training of assessors is one means of seeking to ensure consistency of practice in the operation of the AC in order to achieve the end (associated with formal rationality) of greater judicial diversity. Yet, our interviews with both the consultants, who delivered the AC training, and with the assessors demonstrated an unstructured and superficial approach to the diversity content of the AC training. The consultants reported that:
We have a very limited amount of time for training; it took the form more of a discussion with the group so there probably wouldn't be that much formal material behind it because it was more thrown open to them. (Consultant)
20
Human Relations XX(X)
This view was confirmed in many of the quotations from the assessors, who commented for example, 'it [the training] was fairly unstructured in terms of talking about attendees experiences really' and 'if someone said to me what training did you get in diversity, I would have said nil'. That the training content is likely to be relative to the group receiving the training suggests inconsistent experience between different cohorts. In addition, this is an approach that is likely to reflect the values and experiences of the most dominant members of the group, potentially those who were part of the judiciary. Bearing in mind that it is well recognized that the judiciary is unrepresentative, it was somewhat surprising that more effort was not exerted to ensure that assessors are more aware of the way that their substantive rationalities might act to the detriment of a particular group. That little attempt was made to deconstruct the meanings associated with diversity and its theoretical rationality in the context of judicial ACs is revealing and suggests a limitation of means used to achieve the espoused end of a more diverse judiciary. We acknowledge that such deconstruction of diversity is rare in organizational studies (see Kirton, 2008). Notwithstanding the critical comments, on balance most candidates felt that the rounded approach of the AC, that is, not depending solely on an interview, provided an opportunity for a broader range of people to demonstrate their suitability for judicial appointment:
I think the AC approach is better because . . . somebody who, with a less middle class type background doesn't do so well in . . . interviews, But it doesn't matter who you are. If you're going to be good at doing your job, you get the opportunity to demonstrate it in the other parts of the assessment day. (Female candidate)
The data indicated that for many candidates the theoretical underpinning of the AC supported by the much publicized substantive political rationalities lent the selection process a certain social validity, which meant that they felt confident that regardless of outcome, they had been treated fairly. Proposition 5 is not supported nor refuted since the evidence is contradictory.
Conclusions
This article has investigated the multiple rationalities in judicial ACs and shown how different and competing rationalities interrelate with the judicial diversity project. The analysis reveals a range of theoretical, methodological and empirical contributions and implications of the study. Using a rationalities framework has enhanced our understanding of ACs but has also offered a theoretical purchase. It allows a multidimensional and multi-layered perspective to be employed that reduces the domination of formal rationalities as the focus of analysis at the expense of substantive, theoretical and practical rationalities and allows insights into the nature of rationalities in a political context. Moreover, it introduces the concept of diversity with respect to a particular change intervention (AC) into the rationalities debate. By locating Kalberg's (1980) four-fold rationalities framework in a diversity perspective, particularly with respect to gender and ethnicity, we reveal the contradictory nature of rationalities with respect to the diversity project in ACs and how the complex social
Healy et al.
21
and relational processes of the AC are imbued with competing rationalities. Moreover, our findings indicate that the use of a rationalities framework in organizational analysis will be partial unless a diversity perspective is incorporated. The importance of different levels of analysis is central to this approach. Thus, the gendered and racist nature of our society forms the backdrop to initially explain the unrepresentative nature of the UK judiciary and the political consequence that shapes the substantive rationality that drives reform at the high levels of government. This in turn led to the use of a formal rationality approach to change the nature of the judiciary to more closely reflect British society. Thus political substantive rationality has been key to understanding the complexities and contradictions in the operation of the AC. The rationalities framework reveals the different substantive rationalities that come into play in change initiatives. In AC judicial selection, the political substantive rationalities of diversity (promoted by a political imperative) were embraced, evidenced in the formal rationality of the official documentation and in our discussions with officials and assessors. Nevertheless, it was evident that the substantive rationalities of some assessors in different ways took precedence over the formal rule making. Whereas others, particularly officials who were accountable to the body politic, were more likely to align with political substantive rationalities thereby reflecting an instrumental formal rationality, that is, that such action is in the self-interest of the officials themselves. It was evident that the resolution of the resulting differences in assessor evaluations in the AC led to what Bolan (1999) (drawn from Habermas) called communicative action, oriented toward reaching understanding in the sense of reaching agreement on meaning, value, and the guidance of human association. Thus, there is some reconciliation of differing assessor views through communicative action, thereby denying the objectivity of the formal rational approach. Furthermore, the formal rationality of the AC was also undermined in the training process that avoided any systematic discussion of the theoretical rationality underpinning the diversity concept. Methodologically, we have found it helpful to identify five propositions drawing on multiple research methods to draw out the way that rationalities overlap, interrelate and compete in judicial ACs. If formal rationality is measured by numerical diversity, then we might tentatively suggest that ACs have the potential to 'work'. From this perspec- tive, recent figures provide some verification for this optimism. The proportion of women appointed to judicial office had increased to 41 percent of new appointments in year to March 2006 (in 1999 the figure was 24 percent). The proportion of BME appointments rose to 14 percent (in 1999 the figure was 5.4 percent) (DCA, 2007). T h e p o l i t i c a l s u b s t a n t i v e r a t i o n a l i t i e s a r e i m p o r t a n t f a c t o r s i n t h i s i n c r e a s e , a rg u ably more important than the mere existence of the ACs. ACs are not universally used and as our quantitative investigation of a DDJ study demonstrates, the results are inconclusive. The findings indicate that there is some consensus with respect to assessor and candidate perceptions of procedural justice and in this way, by affording candidates and the assessors multiple opportunities to demonstrate or evaluate a breadth of knowledge and skills. This would provide some verification for the instrumental means-end aspect of formal rationality.
22
Human Relations XX(X)
Engaging with Brubaker's (1984) argument that there are conflicting values between calculability, efficiency and impersonality on the one hand and fraternity, equality and caritas on the other' provided a link between rationalities and diversity. However, we suggest this conflict is not inevitable. Weber's (1978) ambiguities in substantive rationalities indicates a complexity in that some groups' substantive val- ues uphold calculability and may in fact reflect principles of exclusion, rather than caritas and inclusion, thus providing consensus between formal and substantive ratio- nalities as well as conflict. Accounting for the intricacies in our understanding of rationalities lies very much in the relational aspects of Weber's thesis. While all asses- sors shared a common understanding of ACs and understood the political imperative to increase the diversity of the judiciary, at the same time, they imposed their gen- dered and ethnicized substantive values on the means of the formal rationalities. In this way, their interpretations were able in subtle ways to influence the ratings of individuals and the appointment decisions made according. The limitations to the liberal equality approach (Jewson and Mason, 1986) underpinned by ideas of formal rationalities are evident. Thus by contrasting the caritas of ACs with the desired numerical formal end, we find a further contention between a diversity sensitive judi- ciary and simply greater numerical diversity in the judiciary. It is unsurprising that in some cases assessors' substantive rationalities conflicted with the political substantive rationalities. The existence of occupational segregation by gender, race and class within the legal profession and the associated asymmetrical nature of the wider relations of power limit the possibilities for a 'quick-fix' implied by such for- mal rationalities. The introduction of any particular selection method, such as ACs, could only ever have a partial impact on diversity, because essentially the method is seeking to eliminate discrimination not oppression by subjecting everyone to the same tests of ability and performance. Whereas Young (1990) would argue that we need not equal treatment but differential treatment to tackle oppression. ACs for judicial appointments do not and cannot tackle the institutional and structural framework of oppression that gives rise to occupational segregation by gender, race and class generally. In this case, in a highly systematic HRM approach, we did not find equal treatment, but uneven and unsystematic treatment that had the potential to lead to inequitable treatment. Our findings are contrary to those of Townley (2002) who found that formal r a t i o n a l i t i e s u n d e r m i n e d s u b s t a n t i v e r a t i o n a l i t i e s . T h e d i ff e r e n c e m a y b e e x p l a i n e d b y t h e d i ff e r e n t i n s t i t u t i o n a l c o n t e x t s a n d t h e p o w e r b a s e s i n f l u e n c i n g t h e i m p a c t o f t h e s u b s t a n t i v e p o l i t i c a l r a t i o n a l i t i e s . I n t h i s c a s e , i t i s t h e h i g h l e v e l p o l i t i c a l i m p e r- ative that drives the desire to increase the diversity of the judiciary. This leads to c l e a r s e l f i n t e r e s t a m o n g j u d g e s a n d o ff i c i a l s t h e m s e l v e s t o b e i m m u n e f r o m c r i t i - cisms resulting from the exclusion of certain groups. Nevertheless, the evidence indicated that competing substantive rationalities influenced assessors' actions and judgements at the micro-level of the ACs so that some assessors sought to give 'minority' applicants opportunities to display their capabilities, whereas others dis- played ethnocentric and masculine biased approaches that could disadvantage BME and female applicants' scores. Such action might be what Weber (1978) would
Healy et al.
23
characterize as traditional, that is determined by ingrained habituation but equally it could be the outcome of conscious value-rational action or a combination of both. The value of employing rationality concepts illuminates political processes and helps explain decisions and outcomes in particular institutional contexts that cannot be fully explained by formal rationality alone. Moreover, they reveal that politics are in the delivery at all levels of the AC. Indeed, this single study leads us to go some way to concurring with Bolan (1999) that practical rationality, theoretical rationality and formal rationality all presuppose and derive from substantive ratio- nality, however, we would argue the importance of seeing substantive rationalities in the plural and as relational. It is by revealing the multiple substantive rationali- t i e s t h a t e n a b l e s u s t o g r a s p t h e i n t e r r e l a t i o n s h i p b e t w e e n t h e d i ff e r e n t r a t i o n a l i t y forms. Understanding the complexities of the rationalities that come into play in a change intervention enriches our understanding, demonstrates their interrelation- s h i p a n d t h e d y n a m i c i n t e r a c t i o n a n d i m p a c t t h a t d i ff e r e n t s u b s t a n t i v e r a t i o n a l i t i e s have on organizational interventions. Moreover, the importance of the societal and i n s t i t u t i o n a l c o n t e x t p l a y s o u t t h r o u g h d i ff e r e n t a c t o r s ' s u b s t a n t i v e r a t i o n a l i t i e s a n d their dynamic interrelationship. It was evident that the candidates perceived irrationalities in the practical implementation of the AC, particularly with respect to the interviews and role plays. However, it was accepted, on balance, that overall the AC was a fairer system of selection than earlier methods. Meyer and Rowan's (1977) point that the assumption t h a t f o r m a l s t r u c t u r e s a r e r e a l l y w o r k i n g i s b u ff e r e d f r o m t h e i n c o n s i s t e n c i e s a n d anomalies involved in technical activities is apposite in this case. Moreover, it was evident that candidates brought different substantive rationalities to their understanding and experience of the AC and that ethnicity and gender shaped these perceptions. Organizations that attempt to reduce discrimination in recruitment and selection should be applauded, but a deeper investigation of what methods such as ACs can actually achieve, has to conclude that ACs should not be seen as a panacea to enhance diversity. While they may be introduced as an HRM initiative to challenge the reproduction of unfair discriminatory practices, such interventions are inevitably partial and context specific. It is suggested that ACs might be an 'elaborate charade' (Garavan and Morley, 1997: 170) because of the lack of evidence that they actually reduce selection errors. However, this could be missing the point. If the purpose of ACs is situated within the macro-political context as in this article, the key issue is the substantive political rationality to be seen to be doing something about the unrepresentative composition of the judiciary. Acknowledgement
The article is draws on original research undertaken for the DCA (Healy et al., 2006). We are grateful to those who made this work possible including the applicants, the assessors, Moira Calveley, Cynthia Forson and DCA officials. Thanks are also due to the anonymous referees and Paul Edwards for their helpful comments.
24 Appendix
Human Relations XX(X)
Appendix 1. Analysis of deputy district judge candidates by diversity and performance in AC Successful at sift Gender Male Female Total =15.819; Unsuccessful at sift Successful at assessment centre 32 (41.0%) 46 (59.0%) 78 (100%) Rejected after Statistics assessment centre 68 (53.5%) 59 (46.5%) 127 (100%) Sift: Difference significant (Chi-S p?? .000; Recommendation: Difference significant (Chi-S = 3.030; p?? .087) 157 (84.9%) 122 (85.9%) 28 (15.1%) 20 (14.1%) 185 (100) 142 (100%) 56 (83.6%) 11 (16.4%) 67 (100%) 96 (85.7%) 16 (14.3%) 112 (100%) Sift: Difference not significant (Chi-S = 0.71; p?? .875; Recommendation: Difference not significant (Chi-S = .149; p?? .829)
101 (47.9%) 119 (68.0%) 110 (52.1%) 56 (32.0%) 211 (100%) 175 (100%)
Ethnic group White BME Total
Appendix 2. Percentage of successful deputy district judge applicants at sift by gender and ethnicity Ethnic group Sift Gender Male BM E Successful at sift Not successful at sift Total Successful at sift Not successful at sift Total 8 (36.4%) 14 (63.6%) 22 (100%) 80 (49.7%) 81 (50.3%) 161 (100%) Female 20 (76.9%) 6 (23.1%) 26 (100%) 77 (65.3%) 41 (34.7%) 118 (100%)
White
Chi-square tests: BME: Chi-S = 8.066, p?? .008. White: Chi-S = 6.704, p?? .011.
Healy et al.
Appendix 3. Percentage of recommended deputy district judge applicants by gender and ethnicity Ethnic group Assessment Gender Male BM E Recommended Not recommended Total Recommended Not recommended Total 3 (37.5%) 5 (62.5%) 8 (100%) 25 (31.6%) 54 (68.4%) 79 (100%) Female 8 (42.1%) 11 (57.9%) 19 (100%) 31 (42.5%) 42 (57.5%) 73 (100%)
25
White
Chi-square tests: BME: Chi-S = .049, p?? 1.000. White: Chi-S = 1.909, p?? .182.
Notes
1. In 1975, 7 percent of solicitors on the Roll of the Law Society and 6 percent of practising certificate holders were women. By 2004, women made up 42 percent of solicitors on the Roll, 41 percent of practising certificate holders and 62 percent of trainees. By 2004 of those who identified their ethnicity, 8.3 percent were from BME groups (11 percent women, 6 percent men) (Law Society, 2005). Gender segregation has proved more resilient at the Bar, while BME people appear to be slightly better represented when compared with solicitors. Women comprise 26.5 percent and BME people 8.6 percent of self-employed barristers (the latter 12.6 percent women, and 7 percent men) and BME people 13.2 percent of barristers at the employed Bar (16 percent women, 11 percent men) (General Council of the Bar, 2004). 2. Seehttp://wwwdca.gov.uk/. In 1997 the responsibilities of the DCA were passed to the Ministry of Justice and to the Judicial Appointments Commission. 3. The Recorder is a more senior appointment than a District Judge. Recorders are appointed by the Queen on the recommendation of the Lord Chancellor. The statutory qualification for appointment as a Recorder is a 10-year Crown Court or 10-year County Court qualification. District Judges are appointed by the Lord Chancellor. The statutory qualification is a sevenyear general qualification. Both appointments are drawn from the legal profession, barristers and solicitors. Seehttp://www.dca.gov.uk/judicial/appointments/jappinfr.htm (accessed 14 March 2007). 4. The judicial candidates were selected to achieve as balanced a representation as possible according to sex and ethnicity. It is noteworthy that the candidate interviewees were from diverse backgrounds and did not generally fit the white male, Oxbridge stereotype. Of the 39, 14 were successful and 17 were rejected following the assessment centre, and eight were rejected following the sift (the shortlisting). Women and BME people are under-represented among assessors and a BME assessor was not available for interview. 5. The qualitative analysis software package NVivo, was used to sort and prepare the transcripts for thematic analysis. Initially, the transcripts were coded according to the responses to the interview questions (which included information on application preparation, opinions on the
26
Human Relations XX(X)
competency approach and experiences and perceptions of the various elements of the AC. A number of themes were then identified as emerging from these responses and the transcripts were then more finely coded to reflect the themes embedded in the questions and answers including issues of legitimacy, fairness, transparency and importance of diversity within the AC process. The use of NVivo also allowed us to conduct searches across the attributes of gender and race to enable comparison of BME/white and male/female responses. This approach facilitated the testing of the five propositions. However, it was not sufficient. In addition all transcripts were iteratively read in order to avoid the mechanistic dangers inherent in such analytical software packages. These processes were important making the links with other forms of documentation. Chi-S = 15.819; p< .000; Chi-S = 3.030; p
HR Study on Competing Rationalizes in the Diversity Project of the UK Judiciary The Politics of Assessment Centers:- The Golden Triangle designates the confluence of the Ruak River and the Mekong River, since the term has been appropriated by the Thai tourist industry to describe the nearby border tripoint of Thailand, Laos and Myanmar.
HR Study on Competing Rationalities in the Diversity Project of the UK Judiciary: The Politics of Assessment Centers
Geraldine Healy
Queen Mary, University of London, UK
Gill Kirton
Queen Mary, University of London, UK
Mustafa Özbilgin
University of East Anglia, UK
Franklin Oikelome
University of Hull, UK
Abstract
Assessment centres (ACs) are increasingly used to recruit highly qualified staff, yet there are few attempts to appraise their introduction from a diversity perspective. This article addresses this gap by exploring the introduction of ACs as a means of increasing the diversity of the judiciary in the UK. It analyses the underpinning politics guiding the processes and subjective experiences of ACs using Weber's identification of types of rationality and argues that substantive rationalities drive the formal rationality of the AC.The study demonstrates the theoretical value of incorporating diversity into the rationality framework and shows that the design, management and implementation of ACs are shaped by the competing rationalities that surround and imbue judicial work.While ACs may be introduced to demonstrate fair selection practices and challenge the reproduction of unfair discriminatory practices, such formal interventions are inevitably partial and bound by the distinctive institutional context and the societal context of inequalities.
Corresponding author: Geraldine Healy, Centre for Research in Equality and Diversity School of Business and Management, Queen Mary, University of London, Mile End, London E1 4NS, UK. Email: [email protected]
2 Keywords
Human Relations XX(X)
assessment centres, careers, diversity, ethnicity, gender, judiciary, rationalities, selection,Weber
Introduction
In England and Wales, judges are selected from among the legal profession - barristers and solicitors. The unrepresentative demographic composition of this group of professionals, in particular the predominance of white, Oxbridge-educated males, has long been, and continues to be, of public concern (CJA, 2005). This concern has now permeated the political agenda; institutional change, in the form of proactive efforts to increase diversity, is sought against a backdrop of significant changes in the last 30 years in the legal profession overall. Available monitoring data indicates that there has been a marked increase in the proportion of women and black and minority ethnic (BME) lawyers (Malleson, 2000).1 Nevertheless, it is argued that there remains deeply entrenched gender and race inequalities within the legal profession that are attributable to approaches to recruitment and selection and to the androcentric organization of careers and life in legal institutions (Rolfe and Anderson, 2003; Sommerlad, 2002). Against this background, the Department of Constitutional Affairs2 (DCA), the government department with responsibilities for upholding justice, rights and democracy (with particular responsibilities for running the courts, and improving the justice system), undertook a programme of work aimed at 'modernising the process for appointing judges with the aims of making it more open, transparent, and increasing diversity in the judiciary without departing from the key principle of appointment on merit' (DCA, 2004). The perceived need to widen the diversity of the judiciary was highlighted in a DCA consultation paper, which stated that:
If the make-up of the judiciary as a whole is not reflective of the diversity of the nation, people may question whether judges are able fully to appreciate the circumstances in which people of different backgrounds find themselves. (DCA, 2004)
Part of the modernizing programme was to introduce assessment centres (ACs) for the selection of the judiciary (see Healy et al., 2006; Kirton and Healy, 2009). This article examines the politics and operation of ACs and their contribution to building diversity. We draw on Weber's types of rationality to explore how different actors shape and respond to the political imperatives of an institutional intervention with respect to diversity. Specifically, this study explores how different and competing rationalities shape the way that ACs for the selection of the judiciary in England and Wales impact on diversity.
Competing rationalities and the diversity project
We start from the point where there is an identified need for an intervention that is designed to lead to a particular change (increase in judicial diversity) in a complex and traditional institutional context and that this need is driven by a political imperative. We use Weber's four-fold rationality framework to illuminate our understanding of judicial ACs to allow us to grasp their multidimensional nature and to reveal how multiple actors
Healy et al.
3
working in a political context affect human resource management (HRM) practice. Moreover, such technical approaches distract our attention from the distinctiveness of the organization being studied. As Townley (2008) states with respect to bureaucratic emphases in organization studies, the substantive qualities of specific organizations become of less importance. While using rationality as an analytical concept to understand ACs and diversity meets these concerns, it is not unproblematic. We are cognizant of Kalberg's claims that Weber's discussion of rationality is 'scattered and fragmented' and 'more likely to mystify than illuminate' (1980: 1146). There is little wonder that there is often confusion about the meanings of rationality. Like Townley (2002) we draw on Kalberg (1980) who sought to reconstruct at the purely conceptual level, Weber's vision of a multiplicity of rationalization processes that variously conflict and coalesce with one another at all societal and civilizational levels with the discrete types of ratio- nality constituting the cornerstones for these rationalization processes. Kalberg considers Weber's four types of rationality (formal, substantive, theoretical and practical) in relation to his four types of social action. Moreover, rationalization processes take place each independently of the other and at their own rate (Kalberg, 1980). Formal rationality is characterized by an orientation of action to formal rules and laws and is tantamount to a rejection of all arbitrariness so that a means-end rational calculation is made by reference back to universally applied rules, laws or regulations (Kalberg, 1980). The action emerging from formal rationality will be goal oriented. According to Weber, 'there is a dominance of a spirit of formalistic impersonality: ''sine ira et studio'' without hatred or passion . . . everyone is subject to formal equality of treatment; that is everyone is in the same empirical situation' (1978: 225). The implied objectivity and impartiality of formal rationality for the diversity project is clear. Substantive rationality is a preference for certain ultimate values (Townley, 2002). For Weber (1978), the concept is full of ambiguities and there are an infinite number of possible value scales for this type of rationality. He states that some may involve elements of social justice and equality, whereas others are criteria of status distinctions or of a political unit. This will result in value rational action. Thus, the value spheres that will come into play with respect to judicial substantive rationality will emerge from, for example, political substantive rationalities, the professional substantive rationalities held by lawyers, the substantive rationalities associated with diversity projects and of course the dominant values of the society, rationalities that may be complementary or competing. Theoretical rationality involves 'a conscious mastery of reality through the construction of increasingly precise abstract concepts' (Kalberg, 1980: 1152). For Townley (2002), it is a cognitive template that informs understandings of how the world works and informs action on this basis. Thus, formal rationality means will be informed by insights from theoretical rationality. Practical rationality accepts given realities and calculates the most expedient means of dealing with the difficulties they present (Kalberg, 1980). Thus, it guides actions in daily routines, providing patterns of action for the expedient means of dealing with immediate practical difficulties (Townley, 2002). Townley (2002) argues that it is substantive and formal rationalities that inform the rational action familiar to organization theorists. In her study of museums, she found that that there is an underlying tendency for formal rationality to undermine substantive rationalities.
4
Human Relations XX(X)
Focusing on tension between substantive and formal rationalities, Brubaker postulates that 'the antagonism between formal and substantive rationality may be interpreted as a tension between conflicting values: between calculability, efficiency and impersonality on the one hand and fraternity, equality and caritas on the other' (1984: 41). Furthermore, he argues that the tension is not only a tension between conflicting values, but also a tension between competing interests and between the groups that are the bearers of these interests (Brubaker, 1984). Moreover, for Weber, rationality is a relational concept; 'a thing can be rational (or irrational) only from a particular point of view, never in and of itself' (Brubaker, 1984: 35). Selection methods are imbued with formal rationalities. It is assumed that if procedures are followed, the 'end' will result in the 'right' person being selected. Indeed formal rationalities (with the end of being seen to be fair) drive HRM techniques that seek to subdue substantive rationalities that may impede the diversity project. Such impeding substantive rationalities may not be of caritas but of exclusion. Moreover, different institutional actors may simultaneously promote competing substantive rationalities (of fairness and exclusion). As Weber states, 'many possible meanings of the concept of rationalization will often enter into the discussion' (1978: 30). Selection processes have long been saturated with competing rationalities. Research undertaken by Collinson et al. (1990) in the 1980s highlighted the high degree of informality in the setting and content of selection interviews that was prevalent at the time and that facilitated discriminatory recruiter behaviour. The growth of scientificity (see Townley, 2002) associated with HRM was in part a response to the kind of criticisms levelled by Collinson et al. and it also explains the rise in popularity of highly formalized ACs, at least for the recruitment of highly qualified workers (Dayan et al., 2002; Garavan and Morley, 1997; Iles, 1992). The rationalization of selection in advanced capitalist societies is sharply evident in ACs. An AC can be defined as 'a structured combination of assessment techniques that are used to provide a wide-ranging, holistic assessment of each participant' (Garavan and Morley, 1997: 160). The common features of ACs are: assessment in groups; assessment by groups; the use of multiple techniques (e.g. ability tests, peer evaluations; performance tests); the use of situational techniques (e.g. role-playing exercises); assessment along multiple dimensions (Garavan and Morley, 1997). The use within ACs of simulated experience and multiple methods of assessment is expected to provide a proxy for subsequent job performance and therefore a more objective and valid basis on which to make appointment decisions (Dayan et al., 2002; Iles, 1992), thereby overcoming some of the capacity for discrimination associated with less structured and less rounded approaches. Formal rationality shapes the dominant liberal equality perspective where 'equality of opportunity exists when all individuals are enabled freely and equally to compete for social rewards' (Jewson and Mason, 1986: 307), and where greater proceduralization is meant to lead automatically to greater fairness and equality. The problems identified with ACs include the malfunctioning of the system with respect to planning, job analysis, quality of exercises, the assessors, candidate preparation, poor documentation and scoring (Caldwell et al., 2003); issues of validity and measurement problems (Jackson et al., 2005); their overly complex and onerous nature (Dayan et al., 2002). Underpinning these critiques is a belief in formal rationality that assumes that with the correct means ACs will 'work'. A desire to understand the macro-politics of recruitment and selection (e.g. the wider social, political and labour market contexts, including discrimination and
Healy et al.
5
inequalities) is largely absent. Whereas Young argues that unfair discrimination is not the only or primary wrong that groups suffer in so far as discrimination is an 'agentoriented and fault-oriented concept' (1990: 195). The logical formal rational action is to identify the person/organization responsible; remedy the procedural faults and the end will follow, that is, discrimination is eliminated and greater equality is achieved. This is the HR practitioner's or policy-maker's preferred problem-solving formal rationality oriented approach. Young (1990) contends that the problem is more deeply embedded and states that 'oppression, not discrimination, is the primary concept for naming grouprelated injustice'. Judges, the focus of this article, have historically been recruited primarily via secret 'soundings' (consultation) with the existing judiciary followed by a conventional interview. More emphasis was placed on 'who you know', rather than on 'what you know'. ACs are therefore, despite weaknesses, a significant innovation in the context of the judicial appointments process. The DCA anticipated that the introduction of ACs would serve the twin objectives of increasing diversity in the judiciary and increasing public and candidate confidence in the appointments process (DCA, 2004).
Research methods
Our research investigated each element of the AC selection process for two judicial 'competitions' - one for recorder and the other for deputy district judge (DDJ) appointments3 in 2005. This involved: exploratory investigations of i) application documents given to candidates, guidance to panels, training materials; ii) candidate documentation (application forms, assessors' mark sheets, written tests); iii) reviewing AC exercises, that is, role play, written and technical papers and the interview; iv) meetings with consultants who were engaged to aid with the design of the above competitions and DCA civil servants involved in the process. b) semi-structured interviews with 39 judicial candidates (11 BME women, eight BME men; 10 white women, 10 white men) and 10 assessors (four white women, six white men) in order to understand assessor and candidate perspectives and experiences of ACs.4 a) Both assessor and candidate interview schedules were informed by a combination of our interrogation of the documents and scoping interviews with DCA officials and consultants and our understanding of the HRM and diversity literature. The candidate interview schedule used open questions to explore perceptions, experiences and opinions of the equality and diversity aspects of the AC processes and activities in relation to predetermined competencies that the AC was designed to assess (see Figure 1). An indepth discussion of competencies may be found in Healy et al. (2006) and Kirton and Healy (2009). The assessor interview schedule included questions on the assessor role, assessor training, the sift process, the consultation process, the different elements of the AC and general points on enhancing the diversity of the judiciary.
6
Human Relations XX(X)
Each interview lasted approximately one hour. The interviews were digitally recorded and transcribed verbatim. Most interviews were face-to-face and took place in London, Bristol and Manchester. The interview data were analysed using a qualitative data software program (Nvivo).5 The initial analysis was guided by the research objectives and reported in Healy et al. (2006). The rationality framework emerged from our post hoc deeper engagement with the multiple forms of data and provided the means for an analysis of the politics and complexities of ACs. Quotations are used to illustrate these complexities alongside relevant quantitative data. We draw on quantitative data from a 2005 DDJ competition using ACs. This competition attracted 386 candidates, 43 percent women, 12 percent from BME groups (of whom women were a greater proportion).
Analysis of multiple forms of data
Five propositions are used to analyse our data in the context of four rationality types. In line with Yin's (2003) view, the propositions were helpful in relating the findings to our
Independence and professionalism: o Technical knowledge and expertise o Integrity and independencea o Showing authority o Developing knowledge o Managing workload o Managing selfb Judgement o Weighing and analysing o Resolving and deciding People skills o Communicating o Building relationships
Figure 1. The nine competencies
a b
Used only Recorder competition. Used only in DDJ competition.
Table 1. Judicial applicants to deputy district judge competition by sex and ethnicity Applicants Male Female White BM E Ethnic origin not disclosed N DDJ % 57 43 72 12 15 386
Healy et al. theoretical perspective and to focus the discussion within feasible limits. While we introduce propositions in the context of four rationality types, it should be remembered that in practice these interrelate and overlap.
7
Formal rationality
The implied objectivity and impartiality of ACs associated with formal rationality (and the liberal equality perspective) should lead to equality of treatment of applicants and, all things being equal, lead to the desired end of greater numerical diversity in the judiciary. Thus there is a political instrumental imperative behind the ACs' introduction. This rationality was shared by some of the assessors and applicants, but not unproblematically as will be seen in the next sections. Formal rationality because of its means-end focus partially lends itself to statistical analysis.
Proposition 1: Judicial ACs will lead to greater numerical diversity in the judiciary.
Substantive rationality
The judiciary as a body is underpinned by substantive rationalities of justice and fairness. Moreover, it is run by the body politic that in turn, has political substantive rationalities with respect to diversity. At the same time, members of the judiciary as people will hold substantive rationalities , to a greater or lesser extent reflect their professional and social backgrounds. These rationalities may influence their judicial decisions and have an uneven impact on different groups in society (see Healy et al., 2006). The drive to enhance the diversity of the judiciary was partly to introduce greater awareness and understanding of diversity to the judiciary. Moreover, public substantive rationalities may shape the perception of the judiciary with respect to diversity. Awareness of diversity in the AC is built into the Building Relationships competency and into the Communicating competency. Substantive rationalities also rest in the integrity and independence competency. These are assessed in the interview and the role play.
Proposition 2: The substantive rationality of assessors will shape the diversity of the judicial appointments.
Theoretical rationality
Kalberg (1980) states that theoretical rationalization processes may also be carried out by judges who interpret the incipient world view found in political constitution. The corpus of knowledge of the judiciary is laid down in legislation and case law. It has formed the basis of candidates' professional training and practice. While it may be the outcome of substantive political rationalities, it is a theoretical rationality to be acquired by the legal profession. Unlike the museums in Townley's (2002) study where tacit knowledge is central, judicial work is codified and outputs in the form of judgments are
8
Human Relations XX(X)
evident. Nevertheless, based on case law, judicial work also draws on tacit knowledge in the interpretation. Assessors seek to identify candidates' theoretical knowledge by assessing the competency of technical knowledge and expertise, weighing, analysing and resolving and deciding and developing knowledge.
Proposition 3: ACs will allow theoretical knowledge in written tests to be assessed regardless of sex and/or ethnicity.
Practical rationality
Expediency is a central aspect of practical rationality that always indicates a diffuse tendency to calculate and to solve routine problems by means-end rational patterns of action in reference to pragmatic self-interests. In judicial work, there is an important pragmatic element to its practice. The following quotation from a DDJ judge (also an assessor) illustrates this effectively:
District Judges and DDJs are dealing with people at a very vulnerable level in their life very often . . . there are two decisions which you can take that have the most effect upon them and one, is to take their home away and the other is to take their child away. It's a paradox; the administration allow no more time to deal with [such cases] . . . Now what do you do? Do you cut the thing short or do you take the time that it needs to make sure you are getting your decision right? And if you take the time then you are perceived as not being - how shall I put it - not being a company man. Well, I take that particular one very seriously. . . when children are involved. And if you think that children will grow up and will always remember being thrown out of their home. I mean, that's written indelibly on their minds. And these are [important] decisions that you have got to take quickly. And if you've got people that are indecisive and they dither around, I am afraid you've got a real problem.
This quotation illustrates that workload and time constraints are central to judicial work. Judges are also faced with many diversity based practical rationalities. For example, do they give more time to a person whose first language is not English; how do the demands to make reasonable adjustment for a person with disabilities compete with the time demands outlined in the above quotation? Thus there may be a potential tension between the practical rationalities in the daily work of a judge, and the different needs evident in a diverse society. The AC seeks to assess practical rationality by assessing the competencies of showing authority, managing workload and managing self.
Proposition 4: Candidates will be confident that diversity issues are treated fairly and consistently.
Proposition 5 links the Propositions 1-4 and reflects the second DCA objective to increase candidate confidence in judicial selection processes.
Proposition 5: ACs will not advantage one group of judicial applicants over others.
Before considering each of the propositions in turn, we outline the context of judicial ACs.
Healy et al.
9
Objectives and features of ACs for judicial appointments
The objectives of the judicial ACs were established to:
(i) introduce a 'one-day' selection system, that provides a range of different activities designed to test the criteria for appointment in a fairer and more comprehensive way than interview alone; (ii) collect more information about each applicant, particularly those who receive little or no comment from consultation [with serving judges], on which the decision to appoint or reject can be made; (iii) collect information about applicants using a range of different activities designed to test an applicant's abilities against the criteria for appointment . . . to make decisions on appointment in a fair, systematic and consistent way; (iv) to provide more helpful and specific feedback to the Lord Chancellor about an applicant's performance at the AC on which he can make decisions on appointment; to provide more helpful and specific feedback to unsuccessful applicants about their performance at the AC. (DCA internal report, cited in Healy et al., 2006: 38)
The ACs were tailor made for judicial appointments and consisted of a written technical legal test, written analysis of case studies, practical exercises (role plays) and an interview. For the role play exercises, each candidate, observed by assessors, takes on the role of a judge to handle cases and make decisions. These activities were d e s i g n e d t o s i m u l a t e t h e k e y e l e m e n t s o f j u d i c i a l w o r k a n d w e r e r e l a t e d t o t h e p r e d e- termined competencies. Assessment of all parts of the AC is recorded in writing on forms designed for each activity in order to ensure that decisions are justified. In addition, interviews and role play exercises are tape-recorded - this is meant to increase candidate perceptions of procedural fairness. Assessor panels consist of a serving or retired senior official from the DCA, a lay person (with a human resources background or significant experience of interviewing) and two judicial office holders. In addition, the policy is where possible to have a mixed panel to include male, female and BME members. The broad composition of the assessor panel arguably democratizes the selection decision-making processes and meets a fundamental condition of procedural justice (Young, 1990). It prevents a concentration of power over judicial appointments within the established judiciary (as in the 'secret soundings' approach). All assessors undergo compulsory training in 'sift', inter- view and competency policies and procedures. Mirroring Jewson and Mason's (1986) liberal perspective on equality, most assessors placed great faith in the theoretical ratio- nality of their training to ensure the fairness of the formal rationalities of the AC proce- dures and processes. The design of the process was undertaken by consultants and based on theoretical knowledge of ACs and formal rationalities of the means and ends of the process, whereas, the decision to introduce the ACs was imbued with substantive rationalities underpinned by a complex set of political and equality values.
Findings Proposition 1: Judicial ACs will lead to greater numerical diversity in the judiciary
If Proposition 1 is supported, it will mean that women and/or BME applicants are more likely to be recommended for appointment than white men. Appendix 1 shows that out
10
Human Relations XX(X)
of the 386 candidates, a total of 211 (54.3%) candidates attended the AC, and 78 (37%) were recommended for appointment. The DDJ data were examined to see whether there were any significant differences in performance at sift and at the AC for different groups according to sex and ethnicity. The findings are summarized in Appendix 1. Sex appears to influence overall success at both sift and the AC.6 Women are more likely to be successful at sift and also more likely to be recommended for appointment following the AC than men. Ethnic group did not appear to have any influence at sift7 nor at the AC.8 Further analysis was carried out on the DDJ Competition data to find out whether the influence of sex on the sift and recommendation is moderated by ethnicity. The results are shown in Appendices 2 and 3. Appendix 2 indicates that at sift, the influence of sex is significant both for white and BME applicants,9 although this does not appear to be the case at the AC for both groups.10 In other words women of both BME and white background are more likely to be successful at sift than their male counterparts but are not more or less likely to be successful at the AC. When sex and ethnicity are combined as categories, BME women may have greater success rates at sift than BME men. At the sift stage of selection against the competencies, the data show that women and BME applicants were more successful and that BME women are more successful than BME men. How do we interpret these data? One interpretation might be that the more open and transparent process not only attracted a wider pool of candidates, but also enabled those candidates to demonstrate their suitability using the predefined competencies. In other words, the AC would reflect its underpinning formal rationality. An alternative interpretation might be that the political agenda to increase the diversity of the judiciary had influenced sifters to look more seriously at candidates whom they might previously have rejected. However, of those appointed analysis reveals that women are more likely to be successful but that ethnicity is not significant. Why would this be? Theoretical rationality explanation would be that the system has identified the best candidates. However, it might point to the greater propensity to treat BME candidates as 'other' (see Young, 1990: 60) and an inability to demonstrate sensitivity to diversity that results in BME candidates having greater difficulty demonstrating their competencies, as we see to some extent in this article. Our results do not show unequivocal support for Proposition 1. Some caution is necessary in assuming we can generalize from these results. We have examined only one competition and the numbers for different categories are small. The social context of these competitions is outside the terms of this article. However, statistics show that BME candidates are likely to apply in proportionately greater numbers to lower level courts (e.g. DDJ) than to the higher courts, which have tended to retain their traditional gender and ethnic composition.11 Therefore for Proposition 1 to be properly tested, the exercise should involve a greater number and a diversity of competitions.
Proposition 2:The substantive rationality of the assessors will shape the diversity of judicial appointments
Assessors bring their substantive rationalities into all stages of the assessment process but this is intensified in the interview and the role plays. Our findings revealed a complex picture of contradictory substantive rationalities characterized by both caritas and exclusion.
Healy et al. The Assessor Guide on interviewing states:
11
Most discrimination occurs unwittingly. However unbiased interviewers believe themselves to be, it is natural to hold certain preconceived ideas as to how other people are likely to behave. In the interview open minded objectivity is vital in order to minimize the risk of bringing bias and prejudice into the process. In general this means that questions should not be asked of one group of candidates (e.g. women or members of ethnic minority groups) which are not asked of others. Questions about a candidate's domestic arrangements should be avoided altogether . . . Questions which may be misconstrued by candidates on equal opportunity grounds (or which may give offence) must be avoided.
This advice reflects the nature of formal rationality and of the liberal equality perspective, that is, that all candidates should be treated the same. In addition, assessors are provided with a set of possible interview questions, but are given discretion in the choice of questions posed to individual candidates and in the way that questions are asked. This gives assessors control over the perceived 'best way' to identify competencies. An exam- ple of how such control was used to shift the direction of an interview was given by one assessor who talked about his feeling that encouraging candidates to bring in examples from broader life experiences (as the competency approach allows) might actually give women an unfair advantage over men. He explained:
In [assessing the competence] 'building relationships' . . . sometimes you get a better response from women candidates particularly when you ask, 'members of the judiciary have sometimes been criticized for being too distant, do you agree with that?' Then the follow-up could be 'what steps do you take to ensure that you keep in touch with people?' Usually then women come out much better [than men] and that may well be through school, with children, social groupings, and interaction with people in the community. I am usually conscious of that and if that is the case, I try to get them off that and move them on somewhere else because that's advantaging them.
Such gendered value-based redirection was not found when candidates attempted to demonstrate their competency by way of more traditional work based examples. This illustrates how assessors can control the interview and therefore how they can actively shape the outcome; that is, if a candidate is prevented from demonstrating (gendered) strengths, then it is hardly surprising if their rating is low and in turn a low rating justifies non-appointment. Since judicial ACs are shaped by HRM approaches to diversity management, underpinned by principles of formal and theoretical rationality, there may be an expectation that there would be transparency about questions used to evaluate candidates' competencies. Yet some BME candidates recounted questions that seemed unrelated to any espoused competency; for instance, a question was asked about the candidates' view of racial disturbances in a particular area. Asking extraneous and even politically charged questions of one group of candidates enables interviewers to impose their substantive rationalities on the outcomes of interviews - 'inappropriate' answers clearly attracting low ratings. This clearly deviates from the formal rationality of the AC. Further, some assessors conveyed their own substantive values about the roles BME people should play in the community. Their ethnocentric assumption may be that candidates' ethnicity overruled their professional middle class status, for example:
12
Human Relations XX(X)
Well, again, if you ask them what they are doing in their community because, I think of an Asian candidate, telling me he was a member of Rotary wasn't actually what I was looking for.
The fact that the candidate did not play out the stereotype of the Asian person solely rooted in their ethnic community (rather than the middle-class England of their white professional contemporaries) meant that the assessor's value based substantive rational- ity might militate against awarding a high competency rating. Thus, paradoxically, the agenda to build diversity could work against some BME candidates who do not play up their ethnic difference. Whereas another assessor recognized the intersectional connection between class and ethnicity but viewed class as overriding ethnic disadvantage:
I mean most of them [candidates] weren't the sort of black people that I would meet in my Court. They were highly educated. Some public school, grammar school, graduates so that they weren't - they weren't ordinary and they'd never had to struggle with diversity because of their colour in society. It had not affected them.
Thus the one assessor would see BME lawyers as 'other' in the sense of having entirely different experiences than white lawyers. The other saw them first and foremost as middleclass professionals who would have no connection with the black people who appear in his court, in this case denying the intersectional identities and disadvantage that BME professional workers experience in life and work in Britain (Heath et al., 2008; Oikelome and Healy, 2007; Skellington, 1996). Crucial to these different substantive values are the potential interpretations that result. In contrast, assessors also demonstrated the caritas associated with Weber's substantive rationality. Scrutiny of assessors' documentation exposed evaluations of candidates' insensitivity to diversity, for example in assessing 'building relationships' an assessor wrote '. . . his descriptions of some clients suggested either personal distaste or naïve surprise at other people's ways of life'. This was reflected in the low mark allocated for this competency. Turning to the role-play exercises, when assessors talked about their evaluation of different 'types' of candidates it was clear that some gender stereotyping occurred, but the outcome for candidates was not necessarily negative:
Often you get a complete surprise. For example a woman can be really interventionist, really assertive and you get a male candidate who is very sensitive, empathizes, caring. I found that with them all, you can never predict by the sex of the candidate that that's how they are going to play it.
Again, a complete mixture of response . . . many [women] are very, very good; quite surprising really. Some of them really empathize. But you would expect that, from their experiences and so on, you would expect them to be good.
Assessors' lack of understanding of stereotyping was evident; sex or ethnicity seemed to foreground applicants who were not white male. Nevertheless, assessor judgments could both advantage and disadvantage candidates from under-represented groups and
Healy et al.
13
yet ostensibly assessors were adhering to the procedures stipulated in the AC guidance documents. The AC is a political process where gendered and ethnicized dynamics may play out. Thus, the power relationship between the lay and the judicial assessor shape the way that the formal rationality requirements are interpreted. Moreover, the applicant too may enter these power dynamics, when it can also take a gendered turn. An assessor reported:
My experience is that they [the applicants], especially if there is a man and a woman on the panel . . . will address all their comments to the man.
Influenced by their substantive rationalities, applicants may unconsciously augment a gender power imbalance and potentially undermine the importance of the woman assessor thereby creating an additional and unnecessary hurdle for the woman assessor to assert her authority and perhaps influence on the process. Those who were lay assessors were more likely to uphold substantive values of e q u a l i t y a n d f a i r n e s s a r o u n d t h e d i v e r s i t y a s p e c t s o f t h e A C , w h e r e a s t h e j u d i ci a l a s s e s- sors were constantly balancing the formal rationalities with the substantive rationali- ties absorbed from their judicial work and professional socialization. How are these differences in substantive rationalities resolved in a practice? On the one hand, we learn that:
Obviously the judicial member will carry greater weight, if he [sic] says 'I think it's very good the way they dealt with that subject'.
In the case where there was disagreement:
. . . we did have a chat round the table about the three different candidates and there was a bit of give and take then because we'd all made very comprehensive notes.
Thus, the parties seek a compromise acceptable to them all, the outcome of which may of course reflect the power relations between the parties. Therefore, the variability of assessor interpretations was not only informed by the formal and theoretical rationalities and their requirement to undertake the practical rationalities of the process, but also and crucially by the different substantive rationalities that the assessors held as mediated by power relations. Proposition 2 holds.
Propositon 3:ACs will allow theoretical rationalities in written tests to be assessed regardless of sex and/or ethnicity
At the AC, theoretical rationalities assessing the required competencies are assessed by a) a written paper in which applicants analyse case papers and document their observations, and b) a technical paper that is a test of knowledge of relevant jurisdiction or potential to acquire it. Criticisms of the tests were articulated by candidates
14
Human Relations XX(X)
(e.g. insufficient time; extent of preparation required), although there were no apparent gendered nor ethnicized differences in these comments. Critical comments were more likely to emerge from those who had not taken examinations for a long time and/or who worked in legal specialisms that do not require day to day engagement with a breadth of theoretical knowledge. Generally, candidates felt that the tests were important, objective measures of their competency and suitability for judicial appointment, which reinforced the formal rationality of that part of the AC. Candidates were divided on whether these were appropriate means to assess the corpus of legal knowledge necessary for the judiciary often depending on how well they had done in the assessments. A common comment was that they were not fully prepared for the content of such tests. Nevertheless candidates accepted that the competencies assessed by the written tests (in technical knowledge and expertise and weighing and analysing) are a necessary requirement for judicial appointments. Furthermore, we found no difference in perceptions of the legitimacy of these processes regardless of sex and/or ethnicity. However, differences between the candidates emerged with respect to the lack of anonymity on the papers. Women and BME candidates (but not white men) did identify concerns that candidate names rather than identifying numbers were on the paper, which is the common practice of UK universities. Assessors were divided in the possible implications of this practice. While some thought that its introduction might have the benefit that candidates would perceive that their paper was treated fairly; in essence the anonymizing process would be tantamount to a rejection of any arbitrariness in the way that theoretical tests would be activated through formal rationalities associated with universally applied rules, laws and regulations (as cited earlier by Kalberg, 1980). Other assessors thought that anonymity was unnecessary, reflecting a belief in the formal rationality of the process. They argued that the pressure to mark the papers within the day meant that it was unlikely that they would notice who wrote which paper and therefore unlikely to bias the interpretation of answers. In addition, the technical paper was multiple choice where knowing the name would not alter the outcome. Whereas, women and BME candidates were more likely to identify this as a questionable practice; this suggests their greater awareness of potential discrimination. The following BME woman applicant responded:
I don't think [they] should know your identity. I mean for the same reason that examiners should not be aware of your identity when marking papers.
Proposition 3 is unproven. We find that the content of written tests is not permeated with a gendered or ethnicized bias. However, it is possible that such bias could creep in to the assessment owing to the lack of anonymity of the papers.
Proposition 4: Candidates will be confident that diversity issues are treated fairly and consistently
One of the ends of the formal rationality of ACs is that they should be seen to be fair; propriety of questions is one of the measures of fairness in recruitment and selection (Gilliland, 1993). Just as the assessors identified unevenness in the type of questions asked, particularly 'diversity questions', this was confirmed in applicant interviews. On
Healy et al.
15
the one hand we noted some applicants felt that the process was very formalistic yet on the other they noted that there were no questions on diversity, for example:
. . . it was a very mechanical because they had set questions. So they weren't diverting too much. And I had no diversity questions at all. Which I had been warned about, you know, expect really difficult questions on diversity. I didn't have one single question. (BME applicant)
For such candidates it would be difficult to demonstrate their knowledge and experience of diversity issues. In contrast, a number of BME and women candidates had the impression that their experiences of diversity issues were valued. One applicant reported that her application was strengthened by the difficult career route she had navigated and because:
. . . the fact that I was a woman and I think, finally and lastly, that I was a black woman. It wasn't the first thing that was said - it was like at the end that this was because I would bring something different again, as a result of my own cultural background. (BME applicant)
This example is important as it turns on its head the belief that being a woman or from a BME community will have a negative effect on an application; on the contrary, for this woman, these personal attributes were seen as definite advantages. This perspective emerged from a number of candidates and it can be directly related to the political substantive rationalities that were espoused by the Lord Chancellor at that time. These espoused values gave underrepresented applicants confidence that their applications would be treated fairly. However, others felt uncomfortable with the confessional nature of the diversityrelated questions, giving this area of questioning a different nuance for them as opposed to white male candidates. For example:
I find it strange that they would even have to ask the [BME] person about diversity issues. I live it, I live the life. So I don't criticize the question, it was part of the process, but I think it is unnecessary. It's like asking someone who is in a wheelchair how they can relate to other people in wheelchairs, why on earth would you need to do that? (BME candidate)
AC candidates in particular have to give away vast amounts of personal information, thoughts and feelings (Stansfield and Day, 1998). Women and BME applicants suggested that there were more subtle and indirect lines of questioning that were trying to uncover the proximity of candidates to diversity issues, which could potentially be disadvantageous. For example:
You did rather get the impression that if you lay claim to actually being close to them [diversity issues] that was going to be actually better than just knowing or being informed [about them]. (Female candidate)
Despite the claim made in the candidate guidance notes that broader life experiences would be valued, a small number of women indicated that they felt, from the reactions of the panel (body language, facial expressions, etc.), that competencies gained from
16
Human Relations XX(X)
experiences, for example, time out of career to care for children, were not properly valued or understood, for example:
I think it is an important thing to bring up children. I think the experience teaches you an awful lot. I get the impression that they don't think that.
Some women attempted to demonstrate various competencies by reference to family life, but their subsequent time poverty meant they also were unable to demonstrate other competencies by reference to voluntary work, from which they sensed some disapproval. In such cases candidates reported a sense that assessors were trying to trip up candidates who might be demonstrating strengths in one area, by making attempts to get them to reveal weaknesses in others. In contrast, we noted women who felt that diversity was 'in' at the moment and they should play to that, even though they would normally never mention family and children in an application. For some the political rationalities influenced their actions:
I was busy mentioning that I was a woman with children, which I would normally never put on an application. Or I wouldn't try and shoehorn it in, but I felt that it was, I did it with some misgivings. But in the round, I felt that I should try and get that in somehow. (Female applicant)
Moreover, this woman was confident in her experience and sensed that the female assessor empathized with her and understood that with two children and a busy career at the bar, it would be difficult for her to do voluntary work. Just as analysis of the assessor interviews demonstrated that assessors may not share a unitary substantive rationality (see Proposition 2), the applicants' interviews demonstrated a wide range of interview experiences with respect to diversity. With regard to the role play exercises where practical rationality is important, the overall perception of candidates was mixed. Candidates were required to take the part of judges in a simulated court room scene with actors taking the parts of barristers, the accused and witnesses. One case involved an asylum seeker; who in the words of an assessor 'doesn't speak very good English, wears a headscarf and looks rather poor'. The assessor questioned whether applicants would be comfortable with this woman. Yet the candidates reported that their efforts to integrate diversity concerns, such as adjustment to disability issues or to language deficits in the role play were judged negatively by assessors. Thus, where candidates demonstrated practical rationalities in, for example, giving more time to a person judged to require it because of poor English; this was considered poor time management of the managing workload competency. Some female and BME interviewees considered the role plays to be a useful vehicle for displaying their competencies in a practical way. They argued that the role plays allow the applicants to put themselves in the role of a judge and demonstrate competencies such as resolving and deciding and showing authority, which cannot be easily measured through the other methods of assessment. However, several interviewees considered that potential variations in style of authority and decisiveness, adopted by women, men, minority and majority ethnic groups as well as across other social divides, could be an
Healy et al.
17
issue that might give rise to disadvantage or assessor bias. They feared that such differences can be misconstrued as weaknesses. Yet white males reported that when discussing diversity issues (where asked), the interview seemed to become more interesting:
I think perhaps I found the interview seemed to flow better when we were perhaps touching on issues regarding diversity and that's not assuming too much about the entire process but . . . they seemed to be far more interested in what I was saying there.
The above quotation suggests that (white) assessors felt more comfortable discussing diversity with a white male because it was about the 'other' rather than with the 'other'. A comment from an assessor may go some way to explain the difficulty of discussing what is perceived to be 'otherness':
No, I think you stay pretty clear of race. Perhaps . . . so as we didn't put our feet in it. We had a couple of lesbian applicants and we decided we weren't going to go anywhere near that field in case we upset someone and so the black applicants were treated no differently. They were asked about diversity and the responses were very much the same as the white candidates because they'd never thought about it.
The above comment suggests that the substantive rationalities of the assessors are limited by narrow insights into diversity and that they may be uncomfortable discussing diversity issues, other than in the abstract with the 'other'. This tendency is compounded by lack of clarity in a common understanding and indeed confusion of diversity and what the formal system was really requiring. There seemed to be a different expectation from BME applicants than white applicants and as stated above seemed to be difficult for applicants to untangle. Proposition 4 is not supported. Candidates revealed inconsistencies in the way that diversity was addressed and interpreted.
Proposition 5:ACs will not advantage one group of judicial applicants over other groups
From an applicant perspective, many felt that the AC did not advantage one group over others, but there was a small number of women and BME interviewees who offered critical reflections. For example, responding to a question about whether the AC approach advantages or disadvantages any particular group:
I think it might in certain respects. The people conducting the interview are white and middle class, to that extent yes. (Female candidate)
While the dominant political substantive rationality may be about widening diversity, many applicants were yet to be convinced that the officials of the AC process would share these substantive values.
18
Human Relations XX(X)
The perception of relative advantage or disadvantage may be influenced by applicants' personal biographies that will shape their own substantive values and their perception of the AC. Women, for example, reported past experiences of overt sexism. A striking example was an applicant who stated that she was told that she was not taken on by a set of chambers because 'We wanted somebody with the looks of Diana and the brain of Athena'. The interlocking nature of gender and racism was overcome by some applicants in their early education, for example:
I was told at school that I could never be a lawyer . . . I think at the time the expectations of a black girl were factory or secretary but I thought that's rubbish. I can do whatever I want if I've got the determination to.
Moreover, many applicants, particularly those educated in post-1992 institutions thought they would not get to the AC stage and were pleasantly surprised that their educational background had not appeared to disadvantage them, a view held by both successful and unsuccessful applicants. It is not only their educational background that has the potential to advantage, or disadvantage, them; applicants were also con- scious of a hierarchy of legal experience. Indeed this hierarchy may be ethnicized. The Commission for Judicial Appointments (CJA) (2005) sought to explain the divisions:
. . . graduates who do not attend the elite universities often have a more pressing need for an immediate income, and as a result, gravitate towards Social Security or Immigration law. Consequently, a disproportionate number of ethnic minority practitioners work in these areas - which are as complex and important as any others, but not as high profile, as highly thought of or, in the long term, as highly remunerated as commercial or Chancery Law.
The legal hierarchy (including some AC assessors) is imbued with material substantive rationalities. Such rationalities shape the value given to particular kinds of legal work. For example, pro bono work is an important way of gaining experience and of making a contribution to society, and this may be used as evidence in judicial applications, as an assessor explained:
Pro bono work is another thing; ethnic minority candidates do a lot of law centre work for which they give up their time freely really.
When asked how pro bono work was valued, it was clear that the substantive values driving the assessment of such work reflected the hierarchical status within the profession, that is, that which tended to reflect the white male majority.
Well, it's fine if it's an add-on but if they see it as their life's work and central then I would say it is not thought of as being of the same level. It's not dismissed as being irrelevant . . . but it's not comparable with somebody who is a high-flying and adored barrister in a £500,000 a year fee earning chambers.
Healy et al.
19
The above response reveals how status is closely associated with material rewards and imbued in substantive rationalities. Lawyers from lower status (i.e. lower paid) legal work may find it harder to convince assessors of their professional competence than lawyers in high paid legal work. The consequence of this hierarchy and indeed of the legal fragmentation imposes additional burdens on those lawyers from less privileged specialties:
Because I think that it would generally be people who aren't from ethnic minority and from smaller practices where they would not have as much resources available to them or mentors and it would be difficult for them to prepare for the process than for someone who comes from a different background. (BME candidate)
Legal hierarchies and their associated substantive rationalities will also shape the AC context. A number of applicants reported how they used their networks to demys- tify the judicial selection process. The benefits that accrue to those in more privi- leged positions were spelt out by a white male barrister who is in court two or three times a week:
I am in court sometimes two, three times a week . . . to be seen by quite a few judicial styles and I know quite a few judges socially.
His experience exposes him to different judicial approaches and importantly puts him in the social network of a number of judges from whom he will undoubtedly gain a confidence not available to those in other sectors of the law profession. Moreover, he will gain insight into the practical and theoretical rationalities of judicial work and become aligned with the substantive rationalities of the judicial mentors. The evidence indicates that the fragmentation of the profession results in uneven access to professional, social and time resources to give to the AC preparation and presentation and therefore some judicial applicants would be better able to demonstrate the competencies in a more confident manner than those from less privileged backgrounds. From Proposition 2, it is evident that assessors will bring their substantive values to the AC. This raises the question as to what attempts are made to ensure these values are aligned with the espoused political rationality to build a more diverse judiciary to ensure that all groups are treated equally. Training of assessors is one means of seeking to ensure consistency of practice in the operation of the AC in order to achieve the end (associated with formal rationality) of greater judicial diversity. Yet, our interviews with both the consultants, who delivered the AC training, and with the assessors demonstrated an unstructured and superficial approach to the diversity content of the AC training. The consultants reported that:
We have a very limited amount of time for training; it took the form more of a discussion with the group so there probably wouldn't be that much formal material behind it because it was more thrown open to them. (Consultant)
20
Human Relations XX(X)
This view was confirmed in many of the quotations from the assessors, who commented for example, 'it [the training] was fairly unstructured in terms of talking about attendees experiences really' and 'if someone said to me what training did you get in diversity, I would have said nil'. That the training content is likely to be relative to the group receiving the training suggests inconsistent experience between different cohorts. In addition, this is an approach that is likely to reflect the values and experiences of the most dominant members of the group, potentially those who were part of the judiciary. Bearing in mind that it is well recognized that the judiciary is unrepresentative, it was somewhat surprising that more effort was not exerted to ensure that assessors are more aware of the way that their substantive rationalities might act to the detriment of a particular group. That little attempt was made to deconstruct the meanings associated with diversity and its theoretical rationality in the context of judicial ACs is revealing and suggests a limitation of means used to achieve the espoused end of a more diverse judiciary. We acknowledge that such deconstruction of diversity is rare in organizational studies (see Kirton, 2008). Notwithstanding the critical comments, on balance most candidates felt that the rounded approach of the AC, that is, not depending solely on an interview, provided an opportunity for a broader range of people to demonstrate their suitability for judicial appointment:
I think the AC approach is better because . . . somebody who, with a less middle class type background doesn't do so well in . . . interviews, But it doesn't matter who you are. If you're going to be good at doing your job, you get the opportunity to demonstrate it in the other parts of the assessment day. (Female candidate)
The data indicated that for many candidates the theoretical underpinning of the AC supported by the much publicized substantive political rationalities lent the selection process a certain social validity, which meant that they felt confident that regardless of outcome, they had been treated fairly. Proposition 5 is not supported nor refuted since the evidence is contradictory.
Conclusions
This article has investigated the multiple rationalities in judicial ACs and shown how different and competing rationalities interrelate with the judicial diversity project. The analysis reveals a range of theoretical, methodological and empirical contributions and implications of the study. Using a rationalities framework has enhanced our understanding of ACs but has also offered a theoretical purchase. It allows a multidimensional and multi-layered perspective to be employed that reduces the domination of formal rationalities as the focus of analysis at the expense of substantive, theoretical and practical rationalities and allows insights into the nature of rationalities in a political context. Moreover, it introduces the concept of diversity with respect to a particular change intervention (AC) into the rationalities debate. By locating Kalberg's (1980) four-fold rationalities framework in a diversity perspective, particularly with respect to gender and ethnicity, we reveal the contradictory nature of rationalities with respect to the diversity project in ACs and how the complex social
Healy et al.
21
and relational processes of the AC are imbued with competing rationalities. Moreover, our findings indicate that the use of a rationalities framework in organizational analysis will be partial unless a diversity perspective is incorporated. The importance of different levels of analysis is central to this approach. Thus, the gendered and racist nature of our society forms the backdrop to initially explain the unrepresentative nature of the UK judiciary and the political consequence that shapes the substantive rationality that drives reform at the high levels of government. This in turn led to the use of a formal rationality approach to change the nature of the judiciary to more closely reflect British society. Thus political substantive rationality has been key to understanding the complexities and contradictions in the operation of the AC. The rationalities framework reveals the different substantive rationalities that come into play in change initiatives. In AC judicial selection, the political substantive rationalities of diversity (promoted by a political imperative) were embraced, evidenced in the formal rationality of the official documentation and in our discussions with officials and assessors. Nevertheless, it was evident that the substantive rationalities of some assessors in different ways took precedence over the formal rule making. Whereas others, particularly officials who were accountable to the body politic, were more likely to align with political substantive rationalities thereby reflecting an instrumental formal rationality, that is, that such action is in the self-interest of the officials themselves. It was evident that the resolution of the resulting differences in assessor evaluations in the AC led to what Bolan (1999) (drawn from Habermas) called communicative action, oriented toward reaching understanding in the sense of reaching agreement on meaning, value, and the guidance of human association. Thus, there is some reconciliation of differing assessor views through communicative action, thereby denying the objectivity of the formal rational approach. Furthermore, the formal rationality of the AC was also undermined in the training process that avoided any systematic discussion of the theoretical rationality underpinning the diversity concept. Methodologically, we have found it helpful to identify five propositions drawing on multiple research methods to draw out the way that rationalities overlap, interrelate and compete in judicial ACs. If formal rationality is measured by numerical diversity, then we might tentatively suggest that ACs have the potential to 'work'. From this perspec- tive, recent figures provide some verification for this optimism. The proportion of women appointed to judicial office had increased to 41 percent of new appointments in year to March 2006 (in 1999 the figure was 24 percent). The proportion of BME appointments rose to 14 percent (in 1999 the figure was 5.4 percent) (DCA, 2007). T h e p o l i t i c a l s u b s t a n t i v e r a t i o n a l i t i e s a r e i m p o r t a n t f a c t o r s i n t h i s i n c r e a s e , a rg u ably more important than the mere existence of the ACs. ACs are not universally used and as our quantitative investigation of a DDJ study demonstrates, the results are inconclusive. The findings indicate that there is some consensus with respect to assessor and candidate perceptions of procedural justice and in this way, by affording candidates and the assessors multiple opportunities to demonstrate or evaluate a breadth of knowledge and skills. This would provide some verification for the instrumental means-end aspect of formal rationality.
22
Human Relations XX(X)
Engaging with Brubaker's (1984) argument that there are conflicting values between calculability, efficiency and impersonality on the one hand and fraternity, equality and caritas on the other' provided a link between rationalities and diversity. However, we suggest this conflict is not inevitable. Weber's (1978) ambiguities in substantive rationalities indicates a complexity in that some groups' substantive val- ues uphold calculability and may in fact reflect principles of exclusion, rather than caritas and inclusion, thus providing consensus between formal and substantive ratio- nalities as well as conflict. Accounting for the intricacies in our understanding of rationalities lies very much in the relational aspects of Weber's thesis. While all asses- sors shared a common understanding of ACs and understood the political imperative to increase the diversity of the judiciary, at the same time, they imposed their gen- dered and ethnicized substantive values on the means of the formal rationalities. In this way, their interpretations were able in subtle ways to influence the ratings of individuals and the appointment decisions made according. The limitations to the liberal equality approach (Jewson and Mason, 1986) underpinned by ideas of formal rationalities are evident. Thus by contrasting the caritas of ACs with the desired numerical formal end, we find a further contention between a diversity sensitive judi- ciary and simply greater numerical diversity in the judiciary. It is unsurprising that in some cases assessors' substantive rationalities conflicted with the political substantive rationalities. The existence of occupational segregation by gender, race and class within the legal profession and the associated asymmetrical nature of the wider relations of power limit the possibilities for a 'quick-fix' implied by such for- mal rationalities. The introduction of any particular selection method, such as ACs, could only ever have a partial impact on diversity, because essentially the method is seeking to eliminate discrimination not oppression by subjecting everyone to the same tests of ability and performance. Whereas Young (1990) would argue that we need not equal treatment but differential treatment to tackle oppression. ACs for judicial appointments do not and cannot tackle the institutional and structural framework of oppression that gives rise to occupational segregation by gender, race and class generally. In this case, in a highly systematic HRM approach, we did not find equal treatment, but uneven and unsystematic treatment that had the potential to lead to inequitable treatment. Our findings are contrary to those of Townley (2002) who found that formal r a t i o n a l i t i e s u n d e r m i n e d s u b s t a n t i v e r a t i o n a l i t i e s . T h e d i ff e r e n c e m a y b e e x p l a i n e d b y t h e d i ff e r e n t i n s t i t u t i o n a l c o n t e x t s a n d t h e p o w e r b a s e s i n f l u e n c i n g t h e i m p a c t o f t h e s u b s t a n t i v e p o l i t i c a l r a t i o n a l i t i e s . I n t h i s c a s e , i t i s t h e h i g h l e v e l p o l i t i c a l i m p e r- ative that drives the desire to increase the diversity of the judiciary. This leads to c l e a r s e l f i n t e r e s t a m o n g j u d g e s a n d o ff i c i a l s t h e m s e l v e s t o b e i m m u n e f r o m c r i t i - cisms resulting from the exclusion of certain groups. Nevertheless, the evidence indicated that competing substantive rationalities influenced assessors' actions and judgements at the micro-level of the ACs so that some assessors sought to give 'minority' applicants opportunities to display their capabilities, whereas others dis- played ethnocentric and masculine biased approaches that could disadvantage BME and female applicants' scores. Such action might be what Weber (1978) would
Healy et al.
23
characterize as traditional, that is determined by ingrained habituation but equally it could be the outcome of conscious value-rational action or a combination of both. The value of employing rationality concepts illuminates political processes and helps explain decisions and outcomes in particular institutional contexts that cannot be fully explained by formal rationality alone. Moreover, they reveal that politics are in the delivery at all levels of the AC. Indeed, this single study leads us to go some way to concurring with Bolan (1999) that practical rationality, theoretical rationality and formal rationality all presuppose and derive from substantive ratio- nality, however, we would argue the importance of seeing substantive rationalities in the plural and as relational. It is by revealing the multiple substantive rationali- t i e s t h a t e n a b l e s u s t o g r a s p t h e i n t e r r e l a t i o n s h i p b e t w e e n t h e d i ff e r e n t r a t i o n a l i t y forms. Understanding the complexities of the rationalities that come into play in a change intervention enriches our understanding, demonstrates their interrelation- s h i p a n d t h e d y n a m i c i n t e r a c t i o n a n d i m p a c t t h a t d i ff e r e n t s u b s t a n t i v e r a t i o n a l i t i e s have on organizational interventions. Moreover, the importance of the societal and i n s t i t u t i o n a l c o n t e x t p l a y s o u t t h r o u g h d i ff e r e n t a c t o r s ' s u b s t a n t i v e r a t i o n a l i t i e s a n d their dynamic interrelationship. It was evident that the candidates perceived irrationalities in the practical implementation of the AC, particularly with respect to the interviews and role plays. However, it was accepted, on balance, that overall the AC was a fairer system of selection than earlier methods. Meyer and Rowan's (1977) point that the assumption t h a t f o r m a l s t r u c t u r e s a r e r e a l l y w o r k i n g i s b u ff e r e d f r o m t h e i n c o n s i s t e n c i e s a n d anomalies involved in technical activities is apposite in this case. Moreover, it was evident that candidates brought different substantive rationalities to their understanding and experience of the AC and that ethnicity and gender shaped these perceptions. Organizations that attempt to reduce discrimination in recruitment and selection should be applauded, but a deeper investigation of what methods such as ACs can actually achieve, has to conclude that ACs should not be seen as a panacea to enhance diversity. While they may be introduced as an HRM initiative to challenge the reproduction of unfair discriminatory practices, such interventions are inevitably partial and context specific. It is suggested that ACs might be an 'elaborate charade' (Garavan and Morley, 1997: 170) because of the lack of evidence that they actually reduce selection errors. However, this could be missing the point. If the purpose of ACs is situated within the macro-political context as in this article, the key issue is the substantive political rationality to be seen to be doing something about the unrepresentative composition of the judiciary. Acknowledgement
The article is draws on original research undertaken for the DCA (Healy et al., 2006). We are grateful to those who made this work possible including the applicants, the assessors, Moira Calveley, Cynthia Forson and DCA officials. Thanks are also due to the anonymous referees and Paul Edwards for their helpful comments.
24 Appendix
Human Relations XX(X)
Appendix 1. Analysis of deputy district judge candidates by diversity and performance in AC Successful at sift Gender Male Female Total =15.819; Unsuccessful at sift Successful at assessment centre 32 (41.0%) 46 (59.0%) 78 (100%) Rejected after Statistics assessment centre 68 (53.5%) 59 (46.5%) 127 (100%) Sift: Difference significant (Chi-S p?? .000; Recommendation: Difference significant (Chi-S = 3.030; p?? .087) 157 (84.9%) 122 (85.9%) 28 (15.1%) 20 (14.1%) 185 (100) 142 (100%) 56 (83.6%) 11 (16.4%) 67 (100%) 96 (85.7%) 16 (14.3%) 112 (100%) Sift: Difference not significant (Chi-S = 0.71; p?? .875; Recommendation: Difference not significant (Chi-S = .149; p?? .829)
101 (47.9%) 119 (68.0%) 110 (52.1%) 56 (32.0%) 211 (100%) 175 (100%)
Ethnic group White BME Total
Appendix 2. Percentage of successful deputy district judge applicants at sift by gender and ethnicity Ethnic group Sift Gender Male BM E Successful at sift Not successful at sift Total Successful at sift Not successful at sift Total 8 (36.4%) 14 (63.6%) 22 (100%) 80 (49.7%) 81 (50.3%) 161 (100%) Female 20 (76.9%) 6 (23.1%) 26 (100%) 77 (65.3%) 41 (34.7%) 118 (100%)
White
Chi-square tests: BME: Chi-S = 8.066, p?? .008. White: Chi-S = 6.704, p?? .011.
Healy et al.
Appendix 3. Percentage of recommended deputy district judge applicants by gender and ethnicity Ethnic group Assessment Gender Male BM E Recommended Not recommended Total Recommended Not recommended Total 3 (37.5%) 5 (62.5%) 8 (100%) 25 (31.6%) 54 (68.4%) 79 (100%) Female 8 (42.1%) 11 (57.9%) 19 (100%) 31 (42.5%) 42 (57.5%) 73 (100%)
25
White
Chi-square tests: BME: Chi-S = .049, p?? 1.000. White: Chi-S = 1.909, p?? .182.
Notes
1. In 1975, 7 percent of solicitors on the Roll of the Law Society and 6 percent of practising certificate holders were women. By 2004, women made up 42 percent of solicitors on the Roll, 41 percent of practising certificate holders and 62 percent of trainees. By 2004 of those who identified their ethnicity, 8.3 percent were from BME groups (11 percent women, 6 percent men) (Law Society, 2005). Gender segregation has proved more resilient at the Bar, while BME people appear to be slightly better represented when compared with solicitors. Women comprise 26.5 percent and BME people 8.6 percent of self-employed barristers (the latter 12.6 percent women, and 7 percent men) and BME people 13.2 percent of barristers at the employed Bar (16 percent women, 11 percent men) (General Council of the Bar, 2004). 2. Seehttp://wwwdca.gov.uk/. In 1997 the responsibilities of the DCA were passed to the Ministry of Justice and to the Judicial Appointments Commission. 3. The Recorder is a more senior appointment than a District Judge. Recorders are appointed by the Queen on the recommendation of the Lord Chancellor. The statutory qualification for appointment as a Recorder is a 10-year Crown Court or 10-year County Court qualification. District Judges are appointed by the Lord Chancellor. The statutory qualification is a sevenyear general qualification. Both appointments are drawn from the legal profession, barristers and solicitors. Seehttp://www.dca.gov.uk/judicial/appointments/jappinfr.htm (accessed 14 March 2007). 4. The judicial candidates were selected to achieve as balanced a representation as possible according to sex and ethnicity. It is noteworthy that the candidate interviewees were from diverse backgrounds and did not generally fit the white male, Oxbridge stereotype. Of the 39, 14 were successful and 17 were rejected following the assessment centre, and eight were rejected following the sift (the shortlisting). Women and BME people are under-represented among assessors and a BME assessor was not available for interview. 5. The qualitative analysis software package NVivo, was used to sort and prepare the transcripts for thematic analysis. Initially, the transcripts were coded according to the responses to the interview questions (which included information on application preparation, opinions on the
26
Human Relations XX(X)
competency approach and experiences and perceptions of the various elements of the AC. A number of themes were then identified as emerging from these responses and the transcripts were then more finely coded to reflect the themes embedded in the questions and answers including issues of legitimacy, fairness, transparency and importance of diversity within the AC process. The use of NVivo also allowed us to conduct searches across the attributes of gender and race to enable comparison of BME/white and male/female responses. This approach facilitated the testing of the five propositions. However, it was not sufficient. In addition all transcripts were iteratively read in order to avoid the mechanistic dangers inherent in such analytical software packages. These processes were important making the links with other forms of documentation. Chi-S = 15.819; p< .000; Chi-S = 3.030; p