In approaching this book, we begin by asking ourselves how much progress has been made since the publication of the early attempts to conceptualise the field (Lyle, 1996 and Cross and Lyle, 1999). This research attempted to explore the coaching process and coaching practice and move the focus from sports performance and the purely ‘technical’ aspects of coaching delivery onto the search for a conceptual framework that represented coaching sufficiently well to underpin coach education. Indeed, publications such as those above (see also Abraham & Collins 1998) were stimulated by the realisation that research and academic writing had made little if any impact on coach education.

Therefore, we have posed the question, ‘what progress has been made in “filling out” a conceptual framework for coaching?’, and it is this question that we asked our fellow authors to address in their approach to each chapter. Thus, we were driven by issues such as: has a theory of coaching begun to emerge?, is there a consensual position within academic writing?, how has conceptualisation dealt with the clearly distinctive domains within which coaching is practised?, what is the relationship between coaches’ practice and our concept of coaching?, and what has been the impact of a developing conceptualisation on coaching research? In this introductory chapter we have adopted the same challenging approach that we encouraged in our fellow authors.

In their recent review of coaching research Gilbert and Trudel (2004a) identified in excess of 1000 coaching-related publications. An examination of this considerable landscape of coaching research reveals a bewildering range of theoretical and empirical perspectives and insights into coaching. Despite this seeming depth of empirical work and the recognition within it for two decades (Lyle 1984) of the existence of a coaching process, we conclude that an in-depth understanding of coaching and a conceptual underpinning with which to inform practice is absent (Cushion, 2006 and Cushion, 2007a). Regardless of our research efforts we seem as far removed from consensus or clarity about the nature of coaching as ever (Cushion 2007a) and hence have no clear conceptual framework to inform practice or the development of practice.

More broadly, the test of the utility and value of research to a community is the extent to which its findings are (a) used as recommended practices in the preparation of practitioners and (b) incorporated by practitioners in everyday practice (Ward & Barrett 2002). Research tends to suggest that many coaches work without any reference to a coaching process model, but we have to distinguish between our knowledge of how coaches operate and our capacity to describe this, or indeed how we educate and develop coaches. In reality, the research findings suggest that coaches base their coaching on feelings, intuitions, events and previous experience that trigger actions (Lyle, 1992, Cross, 1995a, Cross, 1995b, Gilbert and Trudel, 2001, Cushion et al., 2003 and Cushion et al., 2006). Furthermore, despite some positive research examples (e.g. Smoll & Smith 2006) there is no evidence for the systematic application of these, or any other findings, in the development of coaching practice or coach education (Abraham and Collins, 1998, Abraham et al., 2006 and Lyle, 2007a) in terms of either methodology or results (Cushion 2007b). As both Gilbert, 2007 and Cushion, 2007b remark they are yet to ‘meet a coach that referenced a coaching model when describing what they do’. Indeed, a somewhat disheartening picture of the ‘effectiveness’ of coaching research to impact practice can be drawn particularly considering the pragmatic nature of coaching practitioners; models or theories that do not work are quickly discarded or ignored (Brewer 2007).

In passing, however, we must note that the criticism of existing ‘models’ needs to be placed in context, and we must be careful not to criticise models for what they were never intended to do. It would be strange indeed if a coach described practice in terms of a coaching process model, given the complex environment and the contingencies and exigencies within which they operate. On the other hand, coaches might be expected to be exposed to models as a means of helping them to reflect on their practice. Conceptualisations of the coaching process are not necessarily there to be ‘drawn upon’ directly and overtly by practitioners. They should, however, inform coach education and development, in addition to underpinning research. Prescriptions for practice that coaches can relate to are likely to be much more focused, simply presented, and couched in appropriate technical language; albeit underpinned by sound theoretical concepts.

These limitations notwithstanding, the literature has broadly come to acknowledge that coaching is a social activity built on a web of complex, context-dependent and interdependent activities that come together to form a holistic process (Lyle, 2002, Jones et al., 2004 and Cushion et al., 2006). It is a remarkably complex, intricate yet coherent process incorporating myriad individual variations that each coach, player and environment adds to the blend (Cushion et al., 2006 and Cushion, 2007a). Perhaps it is this very complexity that has resulted in a dearth of ‘research that has explored the conceptual development of the coaching process’, possibly because it is too complex to research neatly or about which to draw straightforward conclusions (Lyle 1999 p 13, Cushion et al 2006).

At the same time, we have to acknowledge that this is an interpretation, and the view may differ depending on perspectives and underlying assumptions about coaching. However, much of the writing in this book will demonstrate that the sociocultural and pedagogical aspect of coaching has been given insufficient attention, both in coach education and in our conceptualisation of the process. In addition, coaching is also a cognitive activity (Abraham and Collins, 1998, Lyle, 2002 and Abraham et al., 2006). The apparent reliance on intuition, the everyday ‘management’ of interventions, and the capacity to take non-deliberative decisions in the ‘heat’ of practice suggests a cognitive expertise that is certainly not well understood, and is also not a commonplace feature of coach education and training.

With increased research attention devoted to coaching and yet little apparent impact on coaching practice or coach education, this chapter attempts to give an overview and critical evaluation of ‘what we currently know’ about coaching. It is of course beyond the scope of this chapter to ‘review’ coaching research in its entirety, but drilling into key issues and linking these arguments with others presented throughout the book has the potential to provide a broad and comprehensive analysis of the substantive nature of inquiry into coaching. We suggest that a critical examination of the state of the field in terms of conceptual development, research direction and research evidence provides a framework to understand and bridge the ‘theory–practice’ gap (Abraham & Collins 1998). Indeed, Gilbert and Trudel (2004a) argue that utilising such guiding frameworks has the potential to allow a number of things to be achieved; first, researchers to set research agendas, and second, coaches to access and realise the potential of coaching research and finally, to allow the full scope of research findings to be integrated into coach development programmes.

The conceptual features of the coaching process are wide-ranging and multifaceted, and there remains a need to clarify this process so that effective coaching methods can be established (Mathers, 1997 and Lyle, 2002). Conceptualisation creates a mechanism for representing the coaching process, without which there is no adequate basis of understanding, analysis or modelling. Such a representation is required to underpin research and education, and to appreciate the impact of interpersonal and environmental factors. Conceptual schema will address questions about terminology, purpose, variability in practice, meaning, genericism versus specificity, and domain distinctions; and these understandings form the basis of subsequent assumptions about effectiveness, expertise, and good practice prescriptions (Lyle 2002).

Note, however, that conceptualisation is not value-free. Any particular interpretation of sports coaching has the potential to influence coach education [a current example would be the linkage between Long-Term Coach Development and Long-Term Player Development (Stafford 2005)], research validity, and accountability measures. For example a sociological perspective (Jones 2000), a pedagogical perspective (Armour 2004), instructional perspective (Sherman 1997), humanistic perspective (Kidman 2005), or science of performance perspective (Johns & Johns, 2000) each makes assumptions that lead to practical consequences. Naturally there are barriers to conceptualisation that derive from coaching itself. The most obvious is the vast range of coaching contexts in which coaching is said to take place. The challenge of linking in a meaningful way domains characterised by school sport, children’s activities, recreational sport, elite Olympic competition, highly commercialised professional sport, and so on, is considerable, and made even more so by considerations of team and individual sports, environmental differences, and the role of the coach in competition. Would or should a conceptualisation imply that a generic coaching process exists, or should the conceptualisation embrace more than one concept of coaching?

One particular issue has been the difficulty in dealing with the complexity of the coaching process: coaching practice exists within such a variable and dynamic environment of conflicting goals, socio-pedagogical delivery, sports specificity, non-consensual values, coaching traditions, and mores. This complexity has implications for researchers and the validity and utility of their research. Indeed, a number of researchers have argued that without studies specifically directed toward describing the complexity inherent in coaching, knowledge informing the coaching process is likely to largely remain ambiguous and a matter of conjecture (Saury and Durand, 1998, Jones et al., 2004 and Cushion et al., 2006). The danger is that attempts to simplify the complexity appear not to represent adequately the complexity of the environment or coaching practice. Indeed, it could be argued that to date research approaches have taken an overly simplistic approach to coaching resulting in a dearth of useful research into the conceptual development of the coaching process (Jones et al., 2002 and Cushion et al., 2006). As Lyle, 1999 and Lyle, 2002 points out, a fragmented or episodic approach to coaching knowledge tends to underestimate the complexity of the coaching process, and because coaching can be represented as ‘episodes’ and therefore parts of the process described in individual terms, it is easy to overlook the degree to which the inter-relatedness and interconnectedness of coaching sustains the process (Cushion et al., 2006, Cushion, 2007a and Jones, 2007). Consequently, it becomes (and has become) easy to take an asocial, linear view of coaching. This, in turn, leads to immature or limited understanding that hides meaning but gives the illusion of a more complete understanding (Cushion 2007a). It could be argued therefore that this hinders genuine conceptual development and growth.

As Jones and Wallace (2005) suggest, we do not understand the phenomenon of coaching sufficiently in-depth as a precursor to proscribing coaching practice. As a result, the complexity has not been acknowledged or sufficiently understood before attempting to produce models. Consequently, ‘oversimplification of the phenomenon and over-precision of prescriptions is the unfortunate price paid’ (Jones & Wallace 2005 p 123). The outcome has been an approach that fails to fully encompass coaching practice (Lyle, 2002 and Cushion et al., 2006); hence the contribution of such an approach to coaching has been useful but limited. It was for this reason that Lyle (2007b) argues that we should distinguish the intent from the practice – or at least attempt to appreciate the latter in terms of the former. It might, for example, be argued that the coaching intervention is an instrumental, managed process that can be planned in a more or less systematic fashion. Implementation of those intentions, and the on-going process of negotiation and accommodation, becomes susceptible to the complex dynamic environment within which it is delivered.

Although the conceptual development and understanding of the coaching process has been somewhat limited, a promising and growing body of work that explores coaching practice, and the debate that it has stimulated, has begun to emerge. This line of enquiry has been worthwhile in recognising more explicitly the complexity inherent in coaching. This body of work demonstrates that coaching is not something that is merely ‘delivered’ but is a dynamic social activity that engages coach and athlete in an active way (Cushion et al., 2006 and Jones, 2006). The conceptual inference that could be drawn from this is that the coaching process is evident at two levels. On one level is the social context/meaning and flow of delivery created by coach and performer (at all levels and stages of sport), but at the same time, there is the planned intention/direction/delivery from the coach (and potentially influenced or designed with the performer). An interesting conjecture is whether the nature of such ‘levels’ changes with sports domains.

Coaching then is a practical, social activity that has as its characteristics ‘multidimensionality, simultaneity, uncertainty, publicity and historicity’ (Côté et al 1995 p 255). Echoing these sentiments, Saury and Durand (1998) argue that coaching can be characterised as complex, uncertain, dynamic, singular, and with conflicting values. Indeed, Saury and Durand (1998) suggested that the ‘actions of coaches were full of context based, opportunist improvisations and extensive management of uncertainty and contradictions’ (p 268). Increasingly, research is suggesting that each coaching situation carries some degree of novelty, thus practice is characterised in terms of ‘structured improvisation’ that is neither entirely reason-based or planned, and is far from systematic but highly problematic and individual; a set of reciprocal relations between athlete, coach and context (Saury and Durand, 1998, Sève and Durand, 1999, d’Arripe et al., 2001, Cushion, 2001 and Poczwardowski et al., 2002, Jones et al 2003, 2004).

However, Lyle (2007b) has argued that the novelty element can be misunderstood, and to some extent could be dependent on the sport. There is no doubt that the coach’s intervention requires active ‘application’ and the management of the activities/drills/exercises involves a continuous process of adjustment (or not). This management of the situation, embracing the evolving history of the processes and people involved, may be described as the developed expertise of the coach. When perceived this way, the ‘improvisation’ looks much more like a skilful application of more or less anticipated decisions about progression. This is not novelty as such, particularly for the experienced coach, but perhaps a better description is the ‘degrees of freedom’ that might be expected in sports preparation and performance, given the complexity involved.

Accounts of coaching practice such as those mentioned earlier tend to be sport-specific, although the research is from a range of sports, both team and individual. The coaches involved in the research were all operating in the ‘performance domain’ (Lyle 2002) and thus fully engaged with the coaching process. The degree to which the cumulative findings of this body of work can be generalised to coaching in all domains is a matter for debate; but there is arguably enough in the findings to challenge the more ‘traditional’ conceptions of coaching (we take to mean an unproblematic, acontextual, linear assumption, within a performance science framework). Indeed, the utility of the work perhaps lies in its use as a gateway to a more sophisticated and rich view of coaching practice. It is able to identify in coaching what Wenger (1998) identifies as the explicit: language, roles, tools, documents; and also the implicit: relationships, tacit conventions, subtle cues, untold rules of thumb, recognisable intuitions, specific perceptions, well-tuned sensitivities, embodied understandings, underlying assumptions and shared world views. While most of these can never be articulated, they are unmistakable signs of coaching practice and arguably crucial to its effectiveness (Cushion 2007a).

The tacit, uncertain and contradictory nature of coaching has been conceptually well developed by Robyn Jones (e.g. Jones and Wallace, 2005 and Jones, 2007) who highlights both the pathos and ambiguity of and in coaching. His work attempts to reach beyond the problems of capturing the complexity of coaching arguing that coaching is in fact largely uncontrollable, incomprehensible and imbued with contradictory values (Jones and Wallace, 2005 and Jones, 2007). This work draws on some empirical research to support it, and while interesting and indeed appealing, the concepts that Jones develops need further empirical exploration and scrutiny. Empirical research might demonstrate that this ‘uncontrollability’ is overstated and perhaps significantly environment-dependent, while the contradictory values are no more evident than in other similar enterprises. However, perhaps the greatest benefit of Jones’ work has been to highlight and problematise many underlying assumptions made about coaching practice. Indeed, the work has highlighted the challenge in identifying the coach’s expertise when the links between intentions and practice are often compromised or blurred. Indeed, the ability to orchestrate and maximise (perhaps, optimise) detailed technical analysis, planning and performance, in the context of the various social, psychological, and organisation milieus, may be the emerging mark of coaching expertise.

Immersion in, and engagement with, the detail of coaching practice reveals much about the construction and complexity evident within it (Cushion 2007a). We detect a movement towards some clearer and coherent concepts around, ironically, understanding ambiguity and inconsistencies in practice. Structured improvisation (managing the degrees of freedom), or the interaction of order and chaos, suggests that continuity in coaching in fact may not come from stability but from adaptability (Cushion 2007a). The notion of structured improvisation and the degrees of freedom therein, themselves require some considerable elaboration. The terms are useful, however, because they imply that coaching practice is not without an intentional framework, but identifies the continual process of application that appears to characterise the coach’s actions. As we have already argued, the term improvisation may suggest compromise, and an absence of pre-determination that may also not adequately capture the coach’s practice. However, what has become clear is that the ever-changing nature of coaching practice means that we must focus on the totality of that practice and the practitioner, and not simply on ‘episodes’ that occur in the process.

It is difficult to conceptualise coaching without mention of coaching domains, particularly if recognising these specific contexts helps both to understand the complexity, and perhaps ameliorate some of it. We would argue that coaching domains are distinctive sporting milieus in which the environmental demands lead to more or less coherent coaching practice, and a community of coaches recognisable within it. These coaching domains place specific demands on the coach’s expertise and behaviours, and require domain-specific knowledge and understanding to operate within them. The environmental demands refer to the meeting place of agent and structure, that is, an accommodation of the tensions between the performers’ aspirations, abilities and developmental stage, the consequent development and preparation provision, the coach, and the organisational and social context. In a sense, coaching domains are simply a conceptual device; each domain can be thought of as constituted by an aggregation of countless coaching contexts, processes and episodes that exhibit a similar characteristic pattern. The very complexity of the domain practice is recognisable in those elements that are likely to differ: for example, intensity of participation, complexity of performance, coach recruitment, interpersonal skills, value systems, competition emphasis, and the strength of the community of practice.

The result of these differences is differential demands on the coach in terms of knowledge, behaviour, practice and expertise. For example, the coaching process that emerges from each set of circumstances will require different planning skills, tactical preparation, decision making, refining/teaching skills, resource management, interpersonal skills and so on. In addition, the knowledge base and associated skills will differ across domains. It is difficult to avoid suggesting that the higher the levels of athlete performance the greater the sophistication required of the coaches’ knowledge and skills. However, this should be avoided. The coaching processes in each domain should be thought of as different, and presenting their distinctive challenges. Nevertheless, when thought of in these terms, the arguments for distinctive perspectives on coaching are persuasive. Perhaps most persuasive is the notion that coaches will learn to ‘frame’ their roles and expectations within a particular set of personal, educational, and experiential circumstances.

Furthermore, the idea of domains is important as Trudel and Gilbert (2006) argue that a single typology of coaching contexts/domains is required to facilitate research within a meaningful framework, and to assist with the design of coach education. The authors reviewed a number of coaching context classifications, characterised by terms such as community, instruction, competition, professional, volunteer and school. Following the review, Trudel and Gilbert (2006) decided upon a typology of recreational sport, developmental sport and elite sport. This they acknowledge is most analogous to Lyle’s (2002) typology, which they describe as the most thoroughly described, grounded in concepts of the coaching process, and consistent with empirical research on stages of performance development. It may also be the case that the incorporation of developmental strategies such as the Long-Term Athlete Development model (Stafford 2005) will, de facto, create coaching domains.

Understanding coaches and coaching practice across domains remains the cornerstone to conceptual development and engaging practitioners (Cushion 2007b). For example, how coaching impacts the subjectivities of those involved and how coaching is experienced as both a social space and a social structure offers fertile ground for further conceptualising coaching (Cushion 2007b). Against this backdrop, as Cushion (2007b) argues, any consideration of interaction and discourse within the coaching process, and of the coaching process itself that is devoid of context is both flawed and limited. Nevertheless, we reinforce this change in emphasis in the knowledge that the central purpose of sport coaching (improvement of sports performance) and how that can be achieved is sufficiently well understood in its domains to provide a core process against which the social construction of practice can be understood. Our thinking then should not be focused on the production of all-embracing definitions, but about enquiring with greater breadth, depth and detail in order that we increase our understanding about practice. As a consequence, the search for an adequate conceptualisation of sports coaching has implications for how we research sports coaching. For this reason we now turn to a commentary on coaching research.

Despite the growing body of work to which we have already alluded, coaching unfortunately remains relatively speaking, largely under-researched. The existing research tends to be ‘sparse, unfocussed and subjective’ (LeUnes 2007 p 403). We might argue that there are two key reasons for this. First the coaching research agenda is too often driven by personal research interest, with coaches and coaching a convenient data set for some other issue. Second, despite a seemingly compelling argument for a ‘paradigm shift’ in coaching research (Cushion et al., 2006 and Jones, 2007), perhaps reflecting the wider sport science research field, there remains a predominantly narrow, reductionist, rationalistic and bio-scientific approach to coaching research (Jones et al., 2002, Jones et al., 2004, Cushion et al., 2003 and Cushion et al., 2006).

This is despite the significant changes taking place in the methods employed. Gilbert and Trudel (2004) found that the balance of qualitative to quantitative research had moved from 11%/81% to 28%/70% between 1990–93 and 1998–2004. Gilbert and Trudel (2004) note that there had been marked changes into interview-based and observation-based research away from questionnaire studies. This methodological shift perhaps reflected the change away from ‘characteristics’ and ‘career development’ research (emanating from North America) to coach behaviour and cognitive organisation. Despite this shift, the research remains largely positivistic.

However, we have to be careful about generalising across all coaching research, and about criticising research for not being something that it didn’t set out to be. Therefore, we might be critical of the research community for paying less attention to the complexity of the coaching milieu, but we also need to address the methodological challenges that this brings. The criticism that positivism is reductionist and attempts to reduce ‘interference’ (for example, systematic observation of coaching behaviours) is apt; more interpretive methodologies are able to identify the complex interweaving of personal, performance, and environmental factors, but may not, as yet, have contributed substantively to theory building or practice prescriptions.

Coaching research itself may be a misnomer, and we ourselves may be guilty of perpetuating the concept of a unified field or consensual purpose that we will later argue does not exist. Without wishing at this stage to offer a taxonomy of research fields, we can point to coaching practice, both environment and career (e.g. Mallett and Côté, 2006 and Jones, 2007), coaches’ behaviours, both intervention/delivery and interpersonal (e.g. Smith and Cushion, 2006 and Jowett and Poczwardowski, 2007), coaches’ cognitions, both decision policies and decision making (e.g. Vergeer & Lyle 2007), coaches’ expertise (Schempp et al 2006), and coach education and training (e.g. Trudel and Gilbert, 2006 and Nelson et al., 2006). This catalogue of emerging coaching research fields does not, of course, absolve us from reflecting in their methods the generally ‘messy’ nature of coaching for which we have argued.