**4. Criticisms**

Fundamental to maturity models' raison d'être is the claim that adoption of a maturity model will translate into actual value for individuals or organizations. It is not clear the degree to which these models deliver on that promise [47].

For a maturity model to claim efficacy, there must be comprehensible underlying theoretical constructs and mechanisms of action. Central to the matter is the assertion that concepts called *best practices* do exist, and that the assumptions and conditions necessary to attain them are clearly understood. Saying that *maturity* is the state necessary for this attainment is insufficient; models claiming to address maturity should present a well-developed characterization of the construct. Unfortunately, "the central term of maturity is seldom defined explicitly", despite the number of entrants into the field of maturity models [54] (p. 338). It is unsurprising then, that questions have been raised about how accurately maturity is being codified [47], how effectively it can be measured [50], and whether the claims of its existence in hierarchical structures are a representation of reality that is accurate [20]. Maturity is a complex phenomenon involving many intricately related factors. As systems science has demonstrated, interaction among factors is central to understanding complex phenomena. Theorists and developers of maturity models have been criticized for having inadequately considered such interactions in developing the maturity construct and developing models that claim to account for it [20]. To *best practices* and *maturity* itself, we can add notions of *transformation* to the list of undertheorized aspects of maturity models. The allure of these models is the idea that they help a person or organization achieve ever-greater levels of maturity. Models vary in characterizing this transformation as *change*, *development*, or *evolution*. However, the mechanisms of action driving such transformations toward maturity are little understood. From psychology, for example, "[o]bserved relations between stages of moral development and various forms of social conduct do not establish that the structures of moral reasoning that define stages of moral development exert a significant causal impact on moral behaviour" [63] (p. 672). In the field of business process management, dynamics that generate movement among stages seem even less well understood: "All models implicitly expect organizations to eventually reach the top of the maturity ladder" [54] (p. 339), reflecting little understanding of the processes by which this destination is to be reached. Where mechanisms of action that drive the maturation process are described, maturity modelers argue that these, and not others, are to be followed. This has the effect of devaluing alternative approaches and rendering the modes of thinking used by all but a certain (i.e., "mature") segmen<sup>t</sup> of people "deviant and atypical, rather than reasonable and relevant" [47] (p. 172).<sup>3</sup> Enthusiasm for the idea of maturity models notwithstanding, a slipshod approach to building a solid theoretical infrastructure has left such models open to justified criticism.

Maturity models are particularly vulnerable in this respect: the literature generally agrees that maturity models do a good job of describing various degrees of maturity, but where models claim prescriptive insight, they often fail to meet users' expectations. We see this disappointment in users' claims that maturity models are oversimplifications of the lived complexities that users experience [53]—surely a concern of systems thinkers—presenting optimistic messages that maturity is a state eventually reached, ye<sup>t</sup> vague on the details of how this actually occurs [67]. Where such details are forthcoming, maturity models attract criticism that the attention they do give to movement between maturity stages is vague or prescribes "step-by-step recipes" [54] that often do not work. Conversely, some models are so complex that they likewise fail to provide the promised rewards [20]. Too simple or too complex, if maturity models are to achieve fitness of use, the complexity of the frameworks they offer must reflect the needs of users. Likewise, the prescriptions they offer must fit the resources available to individuals or organizations. When models are too costly to adopt relative to the rewards

<sup>3</sup> For example, theorists have discussed such concerns in the project managemen<sup>t</sup> discipline's critique of its own attempts to codify best practice and the means by which it is to be attained (e.g., [64–66]).

they claim to offer, no one wins. When they rigidify the maturity pathways they espouse, individuals or organizations whose problems and environments differ from those envisioned by model designers are left to try force fitting their way to maturity, usually unsuccessfully, or to customize models in ad hoc ways that may also fail [20]. Describing levels of maturity is relatively easy. Recommending pragmatic pathways by which it is to be developed demands attention to real-world impediments to mature behaviour, the difficulties involved in overcoming such impediments, and the need for feasible, flexible guidance that works.

"What works" is, of course, a matter of evidentiary support, and here is where the most damning critiques of maturity models arise. The increasing numbers of maturity models sugges<sup>t</sup> interest in authoritative insight and expertise on how people in varying jobs can operate more maturely; this enthusiasm has, however, been unmatched by actual scientific study to validate such claims. Despite the intuitive appeal of maturity models and anecdotal confirmation that they are useful, research that studies their rigor, validity, or usefulness in correlating model prescriptions with actual success is scarce [45]. When enthusiasm outweighs empirical evidence, the value of a particular model, and maturity models in general, is called into question.

The criticisms that maturity models face are fundamental and appropriate. If such models are to achieve what they set out to achieve, academic communities must undertake serious reflection about the characteristics of maturation to replace the vague belief that it is associated with development in a good direction. (For this reason, scholars in fields such as information science have called for the development of research standards for model designers [53].) Despite the warranted misgivings, there are models that are believed to be relevant and worthwhile: "Certain maturity and competency models might be robust enough to become the global standard for certification purposes" [6] (p. 11). This possibility, that maturity models can stand as international standards of how effective functioning can be measured and developed, has inspired this brief overview.

#### **5. Design Considerations for a Maturity Model of Systems Thinking Competence (MMSTC)**

As discussed earlier, knowledge about systems is foundational to systems thinking. Broadly speaking, knowledge generally tends to be associated with thinking. However, it has been argued that a person can know facts about systems without being a systems thinker [26]. This argumen<sup>t</sup> is consistent with Kruglanski's epistemic theory, which recognizes that possessing a knowledge domain is necessary, ye<sup>t</sup> insufficient [27]. Another key element of epistemics is the particular modes of thinking that are conducive to perceiving something accurately. This he calls "welcoming cognitive conditions"—the mental skills and cognitive stances that one requires to focus one's understanding in order to apprehend a thing. Here, we would say that being a systems thinker requires knowing facts and also utilizing particular ways of perceiving those facts [37]. Coming to know systems facts is a task readily handled by universities; institutions training one on how to do systems thinking are lacking [68].

This is not to say that all existing maturity models for systems thinking have overemphasized knowledge and neglected cognition. The engineering field has several exemplars of models that include cognitive skills, e.g., [33,38,58] and others. However, fields of systems thinking beyond systems engineering are largely deficient in articulating what we might term welcoming cognitive conditions for systems thinking. A Maturity Model for Systems Thinking Competence beyond engineering, then, would incorporate key elements of the domain of general systems knowledge as it is presently understood. For any model to be considered worthwhile, it should also include cognitive orientations deemed necessary for systems thinkers in any discipline to perform systems thinking in competent ways. The facility and sophistication with which one uses them would be indicative of one's level of systems thinking maturity, and without such orientations, it is questionable whether one could be said to be using systems thinking at all. What might be welcoming cognitive conditions necessary for systems thinking? The following could be considered:

An orientation toward causality: A system's structure is made up of causally linked variables [31]. A focus of systems thinking is identifying both those variables and the nature of the causal relations among them. To Checkland, focusing one's attention on simple cause and effect sequences is an inferior form of thinking [24], or at least, not one to be understood as *systems* thinking. Rather, the degree to which systems thinkers succeed in orienting themselves to multiplicities of causal relationships suggests the degree to which a system's complexity will be appropriately understood.

An orientation toward logic: Systems thinkers regularly face phenomena that do not make sense, or rather, the sense underlying a system's behaviour is not always readily apparent. The logic of a system—the "set of principles underlying the arrangemen<sup>t</sup> of elements" (Oxford Dictionary)—is what must be grasped if one is to understand the way a particular system coheres, what makes it robust, and how it maintains its equilibrium [24,31]. In the face of complexity, people tend to oversimplify why a system is behaving as it is or to dismiss it as illogical [36]. Neither cognitive strategy supports the ability to accurately discern the logic underlying a particular system's behaviour.

An orientation toward explicit and implicit structures: Systems can appear explicit, comprised of obvious rules giving rise to understandable behaviours. The awareness that structure generates behaviour [30] enables a systems thinker to avoid attribution errors common among people who do not understand systems (for example, assuming that a particular person's motivation or actions has caused a system's "problem"). Knowing that behaviour is expressive of structure, a systems thinker wanting to change counterproductive behaviour will imagine potential changes to structural design, rather than assuming that a simple substitution of "problematic" elements for others will overcome the confluence of implicit structural factors delimiting how those elements will likely behave [69,70].

An orientation toward subjectivity: Subjectivities are particularly potent features of human systems [71]. Each person possesses mental models of reality that, taken together, create and sustain a system's identity [70] and the degree to which that system is able to learn and change. Mental models eliminate feelings of ambiguity and influence how a system's members think and act; "unless you understand them, you will not understand the system" [70] (p. 147). Yet, inherently, mental models are an aspect of human subjectivity not available for direct measurement [72]. Thus, systems thinkers must orient themselves toward the difficult work of eliciting communication about mental models,<sup>4</sup> translating people's tacit perceptions into discussable language [31,73]. While doing this, they must also refrain from imposing their own judgments on the models of others [25]—striving for a "rigorous approach to the subjective" [24] (p. A43).

An orientation toward self-reflection: In systems thinking, understanding the mental models of members of a system—from *their* perspective—is necessary [70], and necessarily difficult. Always, it must be acknowledged that human subjectivity—others' and our own—is incomplete. While subjectivity enables us to function in reality, it does not provide a full or fully accurate representation of a system in which we are operating. Systems thinkers themselves hold incomplete (and thus inaccurate) understandings of a system. They, no less than anyone, hold preconceived values, "taken as given" assumptions, and their own personalized logics [24,31], all of which can obstruct understanding and clear communication.

<sup>4</sup> A variety of techniques have been used to help elicit mental models so that they can be more openly communicated and understood. These range from the use of diagrammatic techniques akin to fuzzy cognitive mapping and systems dynamics approaches, to rich pictures and other participatory modeling approaches that both surface differences and enable consensus analysis [72].

As such, systems thinkers must orient themselves outwardly toward the systems they seek to understand and also inwardly toward themselves [26]. Discomfort is an inherent part of such dual orientation; oppositional emotions (i.e., disbelief, disagreement, etc.) that emerge in systems thinkers are reliable signals that their personal subjectivities are being challenged [69] and must be consciously reflected upon [24] so that they can be usefully discussed with others.

These orientations toward causality, logic, explicit/implicit structures, subjectivity, and self-reflection could be considered mental stances without which systems thinking is not possible. With verification (to be discussed below), these orientations and/or others could come to be understood as welcoming cognitive conditions uniquely important to systems thinkers and as standards without which a Maturity Model for Systems Thinking Competence would be insufficient.

Were a Maturity Model of Systems Thinking Competence to be developed that would be relevant and worthwhile to systems thinkers in various professions worldwide, systems research communities would do well to glean lessons from modelers and theorists from other disciplines. I turn my attention now to highlighting key considerations to be addressed in any future initiatives to develop competence models for systems thinkers.

Consistent with the move underway to codify the nascent science of systemology [1,2], this author agrees with arguments against maturity models that rely solely on anecdotal assertions (including the welcoming cognitive conditions synthesized above). As Edson and Metcalf [74] have written, good systems research responds to the need to marry scientific discernment with lived experience. Any research initiative to establish a model of systems research competence must consider this. Models that describe levels of maturity solely based on anecdote will fail to meet the rigours of good science and will run the risk of misleading systems thinking practitioners who trust them. There is irony in the fact that most maturity models claim that one cannot skip steps on the path to mature standards of competency, while modelers in most disciplines have skipped the crucial step of empirically validating their own models. Without such validation, claims that experts understand the nature of systems thinking maturity, and that systems communities should measure themselves by those claims, tread on shaky ground. Thus, the development of a maturity model with an explicit theoretical base is vital.

A Maturity Model of Systems Thinking Competence should define constructs like *maturity* and *maturation* and must identify observable indicators of maturity levels and the characteristics of paths that lay between them [54]. Numerous bodies of theory could provide useful guidance in the development of a systems thinking maturity model:


Much discussion about maturity could lead to the implicit assumption that maturity models focus on the maturity of *individuals*. While many do, it is also the case that some models focus on the maturity of *organizations* with respect to their competency in business processes, project management, agile software development, etc. It may be possible to create maturity models for individual systems thinkers, working in education, management, or the social sciences, for example. It may also be useful to create maturity models for teams, for example, working in fields such as artificial intelligence. Were there an initiative to develop a broadly applicable maturity model of *collective* competency in systems thinking, organizational theories can be useful in developing models for organizations in which systems thinking takes place [53]:


Systems thinking communities have at their disposal numerous theory candidates that can assist in the development of sound maturity models for both individuals and groups.

Building on a solid theoretical base, all the strategies and methods demanded of good systems research should be applied to any initiative to develop a Maturity Model of Systems Thinking Competence. Model development must include particular care to rigorously differentiate relationships of inference and causality [45] that anticipate the criticisms leveled at maturity models in other fields, whose claims about what actions can reliably move one to greater maturity rely on scant evidence or none at all.

In any scientific endeavour, care must be taken to avoid generalizing findings from one instance to all conceivable contexts. Röglinger, Pöppelbuß, and Becker have noted that maturity models often do not translate well in all situational contexts their users face [54]. Maturity models have struggled to account for the idiosyncrasies of the problem spaces in which users work. Differences in the size of projects, technical complexity, and organizational culture greatly affect the work people do and the ways they do or do not develop maturity [6]. In particular, work that demands unique processes are hard for maturity modelers to predict and take into account. This makes it difficult to imagine the kinds of skills and behaviours to be called forth from users, which makes it difficult to legitimize certain skills and behaviours as exemplars of maturity [51]. While it is problematic to overstate the number of settings to which a maturity model should apply, so too is it problematic to prescribe qualities—in the name of maturity—that implicitly privilege a too-narrow number of people based on moral typologies [76], gender roles [77], reputations of being proven stellar in particular environments [21], or preference toward particular schools of systems work (e.g., systems dynamics, systems modeling, etc.) [38].

A case can be made that the discipline of systemology is uniquely well-placed to develop frameworks that *can* be generalized in rigorously defensible ways. As Midgley [78] and others have written, a strength of the systems field is the way it encompasses a very diverse collection of perspectives, priorities, and tools. However, since the caution against overgeneralizing applies also to systems science endeavours (W. Varey, personal communication), one might conclude that any initiative to develop maturity measures for systems thinkers would require different models for every one of the widely differing systems approaches. Recently, however, Hammond [79] reminded us about the origins of the modern systems movement that was motivated by the desire to identify patterns common across the boundaries that typically divide academic inquiry. (A central text in the field does, after all, characterize the movement as a quest for a *general* systems theory [22], and organizations like the International Society for the Systems Sciences have been established "to foster the investigation of the analogy or isomorphy of concepts, laws, and models in various disciplines and professions" http://isss.org/world/administration/bylaws). Further, it has been argued that certain perceptual and behavioural competencies are common across multiple systems traditions and methodologies [80].

A credible case can be made that a unified Maturity Model of Systems Thinking Competency is possible. In its creation, designers should be aided by the contributions of systems theorists who have contributed to the field by calling for implicit biases to be surfaced and critiqued in systems work (e.g., [81,82]). For a Maturity Model of Systems Thinking Competencies to be ethical and effective, such biases must be a focus of attention.

Other characteristics of good maturity models would serve systems practitioners well. User-friendly design is important. Systems thinkers operating in different cultures and problem domains should have assessment tools that are accessible and comprehensible. Model theorists have

stressed the importance of well-structured and easily applicable self-assessment tools [54]. Some have advocated for tests that are "quick" [83]. Others have pointed out the usefulness of models that include templates and checklists for users to collect evidence and artifacts of competent activity at each level of maturity [17]. Should systems communities elect to computerize a maturity model, it should feature intuitive graphical interface and easy report-generating capabilities aligned with principles of good software design [17]. Should systems communities choose to go beyond a descriptive model to actual evidence-based recommendations on advancing one's level of systems competence, then "relevant drivers and best practices for a roadmap to [increasing] maturity" [50] (p. 141) in systems thinking should be provided in concrete, actionable language that is commensurate with a level of granularity suitable to each maturity level [54]. An emphasis on pragmatic tools, technology, and developmental plans for a Maturity Model of Systems Thinking Competence would have the effect of meeting systems practitioners in their lived experience, while providing transparency about the qualities and components believed to be indicative of competent skills and behaviours at each stage of systems thinking maturity [54].

It is worthwhile to remember critiques that maturity models imply that adherence to particular schemes of behaviour, uniform techniques, and particular decision-making strategies can automatize and guarantee sure progress toward maturity (e.g., [63]). This trivializes the situational complexities users face and would do systems practitioners ill service. It is axiomatic that systems workers grapple with systems that are messy—wicked, even [84,85]. The grappling would be no less for those attempting to develop a maturity model for competencies relevant to systems thinkers working in complex contexts. Competent systems thinking cannot be routinized; the nature of systems work defies this possibility. Mature systems thinkers are aware of the ways the systems they study are interdependent with the environment and aware of the ways in which they themselves are likewise interdependent [86].

The competing forces of unity and plurality that are central to systems work are mirrored in the structure of maturity development evident in existing models. Every model presents its maturity stages as comprised of multiple interacting factors. Those factors include knowledge, skills, and metacognitive abilities [52]; they involve the interplay of cognition, emotional development, moral development, and decision-making capacities able to resolve difficult psychosocial conflicts (Wikipedia.com—"Maturity"). In other words, any single stage of maturity operates as a system of interdependent elements. Maturity models are complex, involving dynamic interactions unfolding in ways that can shift a person into progressively more mature levels of functioning—i.e., the development of maturity is a phenomenon involving the emergence of successively higher orders of coherence in a person's capabilities.

"A static or prescriptive model of maturity cannot hope to provide the level of guidance that organizations require in making effective choices" [47] (p. 181). Similarly, "the development and refinement of a [theoretical] construct is an ongoing process that requires attention to clarifying the constructs' definition and parts" [87]. The work of developing a maturity model for competent systems thinking must be iterative. Research design for a maturity model project should be both rigorously planned and intentionally modified throughout the research life cycle [88], acknowledging that systems research is a circular process that builds upon previously obtained knowledge and responds to experience gained through the course of the study [89]. The project of developing a maturity model for systems thinkers ought to proceed as would any sound systems research initiative. Careful attention should be paid to problem structuring [89]. How the task is framed should be adjusted as the project unfolds and modelers reflect on what they are learning [90]. Central to the development of a Maturity Model for Systems Thinking Competence would be identification of success factors—for example, education, knowledge networks, use of systems tools and techniques, organizational climate, and the support of leaders are all factors identified as conducive or obstructive to maturity in other domains of knowledge work [5]. The relative contribution of these and other factors to systems thinking would need to be evaluated [6], enabling us to clarify the nature of maturity as it pertains to systems work.
