Modelling the Complexity of the Coaching Process

INTRODUCTION An examination of the considerable landscape of coaching research [1] reveals a range of theoretical and empirical perspectives and insights into coaching. Despite this, an in-depth understanding of the coaching process appears to remain lacking and we seem as far removed from a consensus about the nature of coaching as ever. Consequently, coach development and coaching practice in North America and the UK operates without reference to any guideline or models for good practice that practitioners undoubtedly crave [2, 3, 4]. It would appear that “a model of coaching is required that has at its heart sound theoretical and research foundations, which are applicable to all sports, coaches and age groups” [5, p.105]. If this is the case, and given the scale of current coaching research activity [1]; why have coaching scholars been unable to develop such a model thus far? Perhaps coaching is an enterprise where a definitive set of concepts and principles will always be elusive and as such a singular all encompassing model may not be possible. Alternatively, if it were possible, would such an approach be in the best interests of developing coaching’s conceptual base? The aim of this paper is to build upon earlier work [2] and explore some of the issues surrounding the generation of and outcomes from a coaching process model. This includes a model for coaching (idealistic representation) and a model of coaching (based on empirical research) [2, 4].


INTRODUCTION
An examination of the considerable landscape of coaching research [1] reveals a range of theoretical and empirical perspectives and insights into coaching. Despite this, an in-depth understanding of the coaching process appears to remain lacking and we seem as far removed from a consensus about the nature of coaching as ever. Consequently, coach development and coaching practice in North America and the UK operates without reference to any guideline or models for good practice that practitioners undoubtedly crave [2,3,4]. It would appear that "a model of coaching is required that has at its heart sound theoretical and research foundations, which are applicable to all sports, coaches and age groups" [5, p.105].
If this is the case, and given the scale of current coaching research activity [1]; why have coaching scholars been unable to develop such a model thus far? Perhaps coaching is an enterprise where a definitive set of concepts and principles will always be elusive and as such a singular all encompassing model may not be possible. Alternatively, if it were possible, would such an approach be in the best interests of developing coaching's conceptual base? The aim of this paper is to build upon earlier work [2] and explore some of the issues surrounding the generation of and outcomes from a coaching process model. This includes a model for coaching (idealistic representation) and a model of coaching (based on empirical research) [2,4].

UNDERSTANDING COACHING AND THE COACHING PROCESS
A fundamental issue to the generation of a coaching process model is the very approach undertaken to such a task. Arguably to date, research has taken an overly simplistic approach to coaching resulting in a dearth of useful research into the conceptual development of the coaching process [2,6]. As Jones and Wallace [7] argue, we do not understand the phenomenon of coaching sufficiently in-depth as a precursor to proscribe. As a result, the complexity has not been acknowledged or sufficiently understood before attempting to produce models and consequently, "oversimplification of the phenomenon and overprecision of prescriptions is the unfortunate price paid" [7, p.123]. The outcome has been that models have been too simplistic and fail to fully encompass coaching practice [2,4] [e.g. 8,9,10] hence their contribution to coaching has been useful though limited.
Of course, the answers one gets are largely dependent on what questions have been asked. The questions coaching research has posed to date have by and large been shaped by the methods and assumptions of the positivist paradigm [2,4,11]. This is important as "paradigmatic allegiances can determine the theories, perspectives, or operationally, the theoretical frameworks that shape the research process" [12, p. 134]. A core concept of the positivistic paradigm is reductionism, which is an attempt to understand the functioning of the whole through an analysis of its individual parts [13]. By its nature, this approach provides a 'mechanistic' guide to understanding; viewing human behaviour as measurable, causally derived and thus predictable and controllable [14]. In addition, the positivist paradigm structures the types of questions asked by researchers [13], with the main goal, in this case, being to establish causal relationships between effective coaching and performer learning in a quest for generalised conceptual models of the coaching process [2].
More recently, qualitative research methods have been engaged to explore coaching practice, and have been worthwhile in recognising more readily the complexity inherent in coaching. This research, however, still remains informed predominantly by a positivist tradition and as such is looking for a single, comprehensive, definitive and generalisable coaching process model or schema. The research acknowledges that the coaching process is multifaceted and cannot be represented as a singular element. However, these 'holistic' views of coaching remain presented as systematic, unproblematic; and inadequately deal with the operational, dynamic or adaptive aspects of coaching [2,15] [e.g., 3,4,16,17]. That said, this body of work, by taking a more sophisticated view of coaching, offers valuable insights, but arguably does not sufficiently grasp the nature of coaching practice and the complexity inherent within it.

WHAT IS THE NATURE OF COACHING PRACTICE?
The coaching process and models thereof are a design, and ultimately our designs are hostage to our understanding, perspectives and theories. If we accept that coaches are the diligent implementers of the coaching process and that process operates in an efficient and prescribed manner, then engineering and re-engineering the process in more abstract ways and rolling this model out for implementation is the way forward [e.g., 16,18,19]. However, is this how we genuinely view coaches and understand coaching? Is it instead possible, for example, to conceive of coaches as inventive practitioners who work in such a complex and ambiguous way that can never be captured within a process? These questions and the answers to them reflect our understanding and perspectives on and about coaching. Therefore, it is essential that at the heart of any discussion about the coaching process and models of coaching is the very nature of coaching itself.
Our current understanding demonstrates that coaching is not something that is merely delivered, but is a dynamic social activity that vigorously engages coach and athlete [2,20]. Perhaps even more broadly than this, coaching is not just local events of engagement in certain activities with certain people, but a more encompassing process of being active participants in practices of social communities and constructing identities in relation to these communities [21]. In addition, coaching denotes doing, i.e., it is practical, but not just doing in and of itself, it is doing in a historical and social context that gives structure and in this sense coaching practice is always social [2,21].
Coaching then is a practical, social activity that has as its characteristics "multidimensionality, simultaneity, uncertainty, publicity and historicity" [16, p. 255]. Echoing these sentiments, Saury and Durand [15] argue that coaching can be characterised as complex, uncertain, dynamic, singular, and with conflicting values. Indeed, Saury and Durand [15] suggested that the "actions of coaches were full of context based, opportunist improvisations and extensive management of uncertainty and contradictions" [15, p. 268]. Moreover, these authors among others, argue that each coaching situation carries some degree of novelty, thus practice and hence coaching expertise has limited roots in either planning or reason [15]. Coaching practice can therefore be understood as 'structured improvisation' [22,23] which means that reducing coaching to generic rules and processes becomes, at best, hugely problematic.
Indeed, coaching practice includes not only the explicit (language, roles, tools documents, etc.), but also the implicit (relationships, tacit conventions, subtle cues, untold rules of thumb, recognizable intuitions, specific perceptions, well-tuned sensitivities, embodied understandings, underlying assumptions, shared world views, etc.) [21]. While most of the latter can never be articulated, they are unmistakable signs of coaching practice and arguably crucial to its effectiveness. The tacit (implicit), uncertain and contradictory nature of coaching has been conceptually well developed by Robyn Jones [7,20] who highlights both the pathos and ambiguity of and in coaching. His work reaches beyond the problems of capturing the complexity of coaching arguing that coaching is in fact largely uncontrollable, incomprehensible and imbued with contradictory values [7,20].
Engagement with the detail of coaching practice reveals much about the construction and complexity evident within it [24,25]. So while inconsistency and ambiguity are observed, there is arguably the beginning of coherence to our understanding of practice [2,15]. Structured improvisation, or the interaction of order and chaos, suggests that continuity in coaching comes not from stability but from adaptability. The ever changing nature of coaching practice means we must focus on the totality of that practice and the practitioner, rather than simply on 'episodes' that occur in the process. In this sense, the coaching process becomes an instrument rather than an object of analysis. Such an approach perhaps offers the potential to grasp the 'particular' of coaching within the 'general', and the 'general' of coaching within the 'particular'.

ISSUES AND CONCERNS ABOUT MODELS
It could be suggested that the appeal of modelling the coaching process and, indeed, the desire for such a model, seems to be accepted almost uncritically despite the enormity of the task given the complexities of coaching. Therefore, it is worthwhile giving critical consideration to some of the issues surrounding modelling the coaching process, based on existing research, current thinking, and understanding of coaching practice.
A consistent issue in grasping the complexity of coaching is its representation. Models, diagrams, and schemata, can only be represented two dimensionally and as a result appear as composites of logical episodes. Models are consequently unproblematic representations of what are complex actions and, as such, can only plot hierarchical relationships and interactions without generating an understanding of the functional complexity that lies behind it [7]. Moreover, because coaching can be readily represented as 'episodes' and therefore parts of it described in individual terms, it is easy to overlook the degree to which the inter-relatedness and interconnectedness of coaching sustains the process [2]. As a result, it becomes (and has become) easy to take an asocial, linear view of coaching [3,8]. Indeed, conceiving a social structure such as coaching as the "mere aggregate of individual strategies and acts of classification makes it impossible to account for their resilience as well as for the emergent objective configurations these strategies perpetuate or challenge" [26, p. 9-10].
Coaching being understood as a relational, dynamic social microcosm that is contingent and ever changing has the implication that to think of coaching and the coaching process, one should think relationally or dialectically. The ever changing nature of coaching practice means that the coaching process has to be thought of differently to a 'system', since systems postulate common function, internal cohesion and self-regulation [15]. This relational nature makes the actuality of coaching and its process significantly different to representations of it. Indeed, representations lead us to reduce the effect of the context and reduce coaching to the outcome of direct action actualised during an interaction. As such, we understand much about the 'what' of coaching, but less about the 'why' and 'how'.
This somewhat monological thinking results in 'modelling' as ultimately limiting in that it is a way to enforce a specific and exportable interpretation on coaching to create encompassing models that appear to reveal broad patterns. Despite the apparent utility of this, ultimately the outcome remains a very restricted representation that is abstracted, because through being 'generic' and transportable from one practice to another inevitably entails a loss of context and content. The broad view then, involves trade-offs in complexity; we see more by seeing less and we end up knowing something different with its own relevance, but which does not subsume the perspectives it attempts to incorporate [e.g., 3, 4 16].
There remains much to be learned about coaching practice (good and bad) with large areas of "unexplored foggy practicalities" [2, p. 14]. Against this backdrop of incomplete knowledge, developing models prematurely and with limited understanding could have a detrimental effect to the evolution of coaching's conceptual base. For example, Wenger [21] argues that while practice models with their succinctness and portability have a focussing effect, they can also ossify practice around their inertness, and thus hinder the very conceptual and practical development they are designed to promote. Moreover, practice models developed from an immature or limited understanding can hide meaning in blind sequences of operation, with the knowledge of a formula or schema leading to the illusion that one fully understands the process that it describes. A situation that some coaching scholars [2,7,20] argues is currently upon us.
Arguably, the outcome of the 'modelling' process does not currently serve coaching well, but is instead illustrative of the conservative function of coaching research. Consequently, pre-constructed facts are taken to create and represent something that is actually far more diffuse and intangible in practice. Limited or decontexualised, models, formulae, and schemata dressed in "theoretical tinsel" [27, p. 58] can gain a concreteness once framed. To the practitioner though, these representations of the coaching process often seem disconnected from the coaching context and frozen into a text that does not capture the richness of lived experience. By reinforcing the coaching process as a natural bounded object, it is possible to speak in the language of a 'co-ordinated system' driven by 'common purpose' [28]; an idea that serves to obscure both the social and historical roots of coaching and those involved within it. As a focus of attention essentially detached from practice, models may even be viewed with cynicism by practitioners and scholars alike as a somewhat ironic substitute for that it is intended to reflect [21].

UNDERSTANDING THE COACHING PROCESS: A WAY FORWARD
Lyle [29] argues that, "improvements to coach education and to coaching practice depend on a sound understanding of the coaching process" (p. 29). Clearly the type of knowledge a sophisticated insight into coaching practice offers is important, as the practical context is the context in which the coaching process exists; it is fundamental knowledge to understanding how to enhance coaching and coach education [2,29,30]. Unfortunately, this approach to the analysis of the coaching process is under-developed, with relatively few empirical studies of this nature having emerged.
However, the findings from studies that have been undertaken [e.g., 15,24,25] of the coaching process that have had as a focus the detailed, contextualised analyses of practice, suggest a need to question existing conceptions of the coaching process [8] and develop evidence for others [4]. Indeed, paying attention to the detail of coaching practice, the forces that shape coaching practice and the interconnections that run between them, has revealed much about the construction and complexity inherent in the coaching process [2]. In particular, and in conflict with currently recognised models of coaching and coach education, it seems worthwhile reiterating that it is unlikely that coaching practice and the coaching process can be reduced to the application of generic rules.
This focus upon the level of practice is not to glorify the local, but to see the processes of negotiation of meaning, learning, development of practices, and the formation of identities and social configurations [21] as involving complex interactions between the particular and the general, and the local and the global. As practice unfolds, it evolves in organic ways that tend to escape formal description and control. As a result, the landscape of coaching practice is not, arguably, congruent with structures of division and boundaries. That is not to say that coaching is not independent of structures, but crucially neither is it reducible to them. Indeed, boundaries of coaching practice are constantly renegotiated in order to define much more fluid and textured forms of activity and that require both a sophisticated and detailed analysis.
Consequently, without studies specifically oriented toward describing the complexity inherent in coaching, knowledge of the coaching process is likely to remain imprecise and speculative [2,15,20]. Moreover, by considering the existing research evidence concerning the nature of the coaching process, it is possible to begin to develop a conceptualisation of coaching based on the process as the instrument of analysis rather than the object of it. Cushion et al. [2] outline five features of the coaching process that could be used instrumentally to further our understanding of coaching practice and the coaching process: 1. The coaching process is not necessarily cyclical, but is continuous and interdependent. 2. This process (and practice) is continually constrained by a range of 'objectives' that derive from the club, the coach and the athletes. 3. The process is a constantly dynamic set of intra-and inter-group interpersonal relationships. These relationships are locally dialectic between and amongst agent (coach, player) and structure (club, culture) and are subject to a wide range of pressures. 4. The coaching process is embedded within external constraints, only some of which are controllable (see [7] for further discussion). 5. A pervasive cultural dimension infuses the coaching process through the coach, the club, the players, and their interaction.

CONCLUSION
It could be argued that this approach has the potential to provide a more sophisticated overview of what is involved within coaching practice. With this approach, we also need to be clear about our assumptions concerning the individual and the coaching world and the relationship between the two [31]. Without grappling with fundamental questions about the nature of coaching practice and being immersed in that practice, we are likely to develop representations of the coaching process that are systematic distortions of both knowledge and understanding [2,23].