https://www.academia.edu/17543428/Towards_a_Fourth_Generation_Pattern_Language_Patterns_as_Epistemic_Threads_for_Systemic_Orientation
Towards a Fourth Generation Pattern Language:
Patterns as Epistemic Threads for Systemic Orientation
Draft soon to be published in
Proceedings of the Purplsoc (Pursuit of Pattern Languages for Societal Change) Conference 2015. July 2015 at Danube University Krems, Austria Finidori, Helene, Commons Abundance Network, http://commonsabundance.net/ Hfinidori at gmail dot com Borghini, Sayfan G., Holon Institute of Technology, Academic Innovation Hub, 52 Golomb Street, Holon, Israel, Sayfanst at gmail dot com Henfrey, Thomas, Schumacher Institute for Sustainable Systems, Create Centre, Smeaton Road, Bristol, UK Tom at schumacherinstitute dot org dot uk
Abstract
This paper charts the emergence of a fourth generation of pattern languages that continues the generational progression identified by Takashi Iba. In order to characterize Pattern Language 4.0 in the context of societal change, we start by describing some of the complexities of social change processes: the systemic nature of the challenges involved, its pluralistic nature and the consequent need for a semantic approach capable of reconciling multiple perspectives and issue framings. We subsequently describe the systemic potential of pattern languages and outline general features through which fourth generation pattern languages realise this potential and address these complexities. Finally, we describe PLAST (Pattern Languages for Systemic Transformation) as a concrete example of a fourth generation project and in conclusion return to consider prospects for societal transformation and how the use of pattern languages can contribute to this.
Keywords:
Pattern Language; Social Change; Generative Systems; Emergence; Complexity; Wicked Problems
1. Pattern Languages and Societal Change
Pattern Languages originated in architecture, in the work of Christopher Alexander and colleagues at the Centre for Environmental Structure in Berkeley.1 Alexander's team sought to codify the implicit knowledge underlying vernacular architectures and responsible for them being evocative, in culturally specific ways, of a general feeling of beauty and life enhancement.2 Alexander’s intention was to help reproduce in replicable and scalable ways the morphological and moral coherence that makes such environments nurturing for human beings. He sought to create a process that people could use in a self-directed way, adapted to their time- and place-specific needs, and apply in ways that become self-generating after the fashion of organic processes in living systems.3
The barely tangible life-enhancing property sought by Alexander has been defined in various ways. Social activist Tom Atlee refers to it as Goodness,4 physicist David Bohm as Wholeness,5 cultural theorist Jean Gebser as Diaphaneity,6 Gregory Bateson as ‘The Pattern that Connects’.7 Alexander describes it as something desirable, ineffable yet readily perceivable with high levels of intersubjective agreement, calling it the ‘quality without a name’8:
“There is a central quality which is the root criterion of life and spirit in a man, a town, a
building, or a wilderness. This quality is objective and precise, but it cannot be named.”
He goes on to elaborate on its context-dependence...
“[I]t is never twice the same, because it always takes its shape from the particular place in which it occurs.”
He later describes the conditions under which it can arise:
“This quality […] cannot be made, but only generated, indirectly, by the ordinary actions of the people, just as a flower cannot be made, but only generated from the seed.”
Pattern languages as originally formulated, therefore, were vehicles seeking to convey a holistic purpose, to direct designed objects and processes towards a specific quality; hence, instruments for sense-making and purposive design. They aimed to support self-generated societal thriving by articulating fundamental design principles to guide holistic and life-serving applications within a given domain.
Subsequently, setting aside somewhat Alexander's moral preoccupations to concentrate on technical design improvements, pattern languages were quickly and widely adopted as a standard knowledge format in software programming. Their use became standard practice in object-oriented programming, where patterns are a format for exchange and reproduction of building blocks among experts, resolving complex communication issues within this domain.9 This consecrated their role as lingua franca: formalizing tacit knowledge within specific domains of practice.
1 Alexander, C., S. Angel & M. Silverstein, (1977). A Pattern Language. (New York: Oxford University Press).
2 Alexander, C. (1979). The Timeless Way of Building. (New York: Oxford University Press).
3 Alexander, Christopher. 1996. “Patterns in Architecture” presented at OOPSLA ’96, October 8, San Jose, California.
4 Atlee, T. (2014). The Tao of Democracy. (North Atlantic Books). Pp. 35.
5 Bohm, D. (1980). Wholeness and the Implicate Order. (London: Routledge).
6 Gebser, J. (1984). The Ever-‐Present Origin. Authorised translation by Noel Barstad. (Ohio University Press). Pp. 6-‐7.
7 Bateson, G. (1979). Mind and Nature: a Necessary Unity. (New York: Dutton)
8 Alexander, C. (1979). The Timeless Way of Building. (New York: Oxford University Press). Pp. 19 & 26.
9 Lea, D. (1994). “Christopher Alexander: An introduction for object-‐oriented designers.” ACM SIGSOFT Software Engineering Notes,
19(1): 39-‐46. Gamma, E., Helm, R., Johnson, R., & Vlissides, J. (1994). Design patterns: elements of reusable object-‐oriented
software. (Pearson Education).
They later entered interdisciplinary fields such as design of human-computer interfaces10 and technology-enhanced learning.11 In these fields they facilitate communication of expert knowledge across specialized disciplines, partly reflecting Alexander's original intention to undermine architects' control over construction processes by making their specialized technical knowledge accessible to wider groups of stakeholders.
Figure 1: Pattern Languages and the what, why and how of change
Pattern Languages subsequently expanded into many other domains of activity: pattern theorist Helmut Leitner estimates around 30,000 pattern languages and half a million patterns to exist in domains as varied as movie costume design, organizational design, fire fighting, scrum, and music composition (Leitner 2014).12 Of particular importance to this paper are their applications in fields of social change and sustainable innovation.13 These include bioregional development,14 community action on climate change,15 the Transition movement of community-based sustainability initiatives,16 permaculture’s approach to designing sustainable human habitats,17 as well as other human interaction and action domains such as learning, collaboration, co-creation, innovation, and conflict resolution.18 Some applications stress their potential as tools to advance democracy, inclusion, and social justice, notably the Public Sphere Project’s work on pattern languages for use of ICTs as emancipatory tools.19
10 Pauwels, S. L., Hübscher, C., Bargas-‐Avila, J. A., & Opwis, K. (2010). “Building an interaction design pattern language: A case study.” Computers in Human Behavior, 26(3), 452-‐463.
11 Winters, N. & Y. Mor (2008). “IDR: A participatory methodology for interdisciplinary design in technology enhanced learning.” Computers and Education 50: 579-‐600.
12 Leitner, H (2014). Christopher Alexander – “Introduction and Crash Course” <http://bit.ly/16iRiEw> and <http://bit.ly/1Bw3ls2>
13 Schuler, D. (2014). “Pattern Languages — An Approach to Holistic Knowledge Representation” <http://slidesha.re/1IMvMa5> [Retrieved 13 December 2014]
14 Reliable Prosperity website <http://www.reliableprosperity.net/> [Retrieved April 5th 2015)
15 Community Pathways website <http://communitypathways.org.uk/> [Retrieved April 5th 2015)
16 Hopkins, R. (2011). The Transition Companion. (Totnes: Green Books).
17 Jacke, D. & Toensmeier, E. (2006). Edible Forest Gardening. Volume 2: Ecological Design and Practice for Temperate Climate Permaculture. (Vermont: Chelsea Green Publishing). Pp. 63-‐115. Bane, P. (2012). The Permaculture Handbook. (Gabriola Island: New Society Publishers). Pp. 57-‐95.
18 Leitner, H. (2015). Pattern Theory. Introductions and Perspectives on the Tracks of Christopher Alexander. (HLS Software).
19 Schuler, D. (2008). Liberating voices: A pattern language for communication revolution. (MIT Press).
2. Three Generations of Pattern Languages
Takashi Iba identifies three distinct phases, or generations in the development of Pattern Languages since the end of the 1970s.20 He distinguishes these in relation to their 'object of design’, ‘act of design’ and ‘purpose’
Figure 2: Three Generations of Pattern Languages
2.1. Pattern Language 1.0
The first generation includes Alexander's original pattern languages for settlement and building design and construction.21 The objects of design are mainly material forms. The act of design consists of a discrete and bounded initiative, such as a specific construction project. Patterns’ main purpose is to aid communication (and hence overcome power disparities) between designers and users. As descriptions or hypotheses of sets of invariant properties and systems of forces to resolve given problems,22 they are sufficiently abstract to be adapted to users’ preferences and local conditions, with each principle amenable to diverse forms of manifestation. Pattern languages of the first generation are oriented towards the practice of participatory design with a set purpose.
20 Iba, T. (2013). “Pattern Languages As Media For The Creative Society.” In: Proceedings for Coins13. <http://bit.ly/1Bw0ors> [Retrieved 13 December 2014]
21 Alexander, C., M. Silverstein, S. Angel, S. Ishikawa & D. Abrams (1975). The Oregon Experiment. (New York: Oxford University Press).
22 Alexander, C., S. Angel & M. Silverstein, (1977). A Pattern Language. (New York: Oxford University Press)
2.2. Pattern Language 2.0
The objects of second-generation pattern languages are non-material, intangible forms or structures such as organizations, learning programs, software or interfaces. Design acts incorporate multiple successive iterations over time. The purpose of patterns is to bridge gaps of understanding between expert and non-expert designers; they differ from those in first generation pattern languages in the possibility that non-material, intangible forms hold for successive improvement through incremental and iterative redesign. Patterns contribute to revealing, interpreting and communicating domain-specific tacit knowledge, allowing it to be understood, unpacked, and recomposed to address specific challenges and needs by interdisciplinary groups of collaborators lacking any other form of common technical language. They are themselves regularly redesigned to assist and follow adaptations and keep their edge in technical performance. Examples include those cited above in object oriented
programming and related fields where programmers work collaboratively within interdisciplinary teams.
2.3. Pattern Language 3.0
Third generation pattern languages have as their objects of design social processes, being oriented towards process and interaction among actors and the forms they design, the ‘how’ of design. Acts of design are again iterative, consisting of self-referential cycles of action learning through which actors learn to identify and express their own tacit knowledge, and collaborate on designing their own interactions. The purpose of patterns is to bridge gaps between people with different backgrounds, experiences, perspectives, and knowledge. They function as a medium for conversation and self-reflection, introducing feedback loops into design processes and allowing structural coupling of design and practice. Such processes foster individual development of consciousness and awareness, new forms of communication within and among groups, and new modes of innovation and action.23 Examples include Takashi Iba's pattern languages on creative collaboration24 and learning,25 and the Group works pattern language for group processes.26
Figure 3: Three generations of pattern languages and the why, what, and how of design
Development from Pattern Languages 1.0 through 2.0 to 3.0 can be viewed as a shift from attention to the intent of design and emergent properties (the 'why' of design) in the first generation, to the pattern as an element of an agreed solution to a complex problem (the 'what' of design) in the second generation, to patterns as characterising social processes (the 'how' of design), each generation bringing a new dimension to designing for transformation.
23 Iba, T. (2010). “An autopoietic systems theory for creativity”. Procedia-‐Social and Behavioral Sciences, 2(4): 6610-‐6625.
24 Iba, T. (2014). Collaboration Patterns: A Pattern Language for Creative Collaborations. (Kanagawa: CreativeShift Lab).
25 Iba, T. (2014). Learning Patterns: A Pattern Language for Creative Learning. (Kanagawa: CreativeShift Lab).
26 Groupworks deck: Pattern language for group processes <http://groupworksdeck.org/>
Pattern Language 4.0, whose emergence we propose to chart in this paper, links the why, how and what
and explicitly seeks to employ pattern languages as tools in social change processes that self-evaluate, in ongoing fashion, in relation to a set of common underlying goals. In order to characterize pattern language 4.0, upcoming sections describe some of the complexities of social change processes: the systemic challenges it faces (Section 3), and its pluralistic nature and the consequent need for a semantic approach capable of reconciling multiple perspectives and issue framings (Section 4). We subsequently describe the systemic potential of pattern languages (Section 5) and outline general features through which fourth generation pattern languages realise this potential and address these complexities (Section 6), describing PLAST as a concrete example of a fourth generation project. In the conclusion we return to consider prospects for societal transformation and how the use of pattern languages can contribute to this.
We use the term fourth generation here neither to be prescriptive as to future directions of pattern language research and practice, nor to imply superiority of any generation or generations over any other. The term identifies one emerging trend among possible potentially useful and valid directions in pattern research. Fourth generation builds upon and seeks to integrate insights from the entire history of Pattern Language research.
3. Systemic Challenges and Societal Change
Most present-day societal challenges are systemic in nature. Lacking clear resolution or even problem definition, they exist as ‘wicked problems’, intractable to causal analysis and impervious to simple and concrete remedies.27 Wicked problems defy attempts at definitive formulation because they accommodate many different perspectives on an issue, and proposed pathways to change.28 Multiple intervention points exist; some problems can be symptoms of others; solutions are partial and contingent, and may be contradictory and involve trade-offs. Proven experience, practice and expert knowledge do not exist, data are uncertain and often absent, and the information necessary to understand the problem is dispersed across multiple different contexts.
Lock-ins, path dependencies, and unpredictable unintended outcomes are all common features of complex societal problems, emergent upon dynamic processes affected by the behaviours of multiple agents (human and non-human). Consequences of these dynamics often manifest at levels other than those at which they originate. Initially invisible, they may over time gather momentum and become self-reinforcing, and in turn become structural features that shape the choices available to agents in the system. This is the source of phenomena like carbon lock-in:29 a set of linked technological, institutional and political barriers to decarbonisation of industrialised economies that ensure that, regardless of levels of apparent political will and investment, dependencies on fossil fuels and associated infrastructure remain entrenched.30 This highly undesirable state subsequently propagates as path dependencies built in to new or rapidly changing systems, for example as emerging economies like China and India preferentially invest in market-ready, technologically proven energy conversion technologies with high levels of technical understanding and infrastructural support.31 Innovation that can escape situations of lock-in is restricted either to the margins of the system, or to localized niches deliberately insulated from wider societal pressures.32
7 Rittel, H. & Melvin W. (1973). “Dilemmas in a General Theory of Planning”. Policy Sciences: 155–169 Leach, M., I. Scoones & A. Stirling (2010). Dynamic Sustainabilities. (London: Earthscan).
29 Unruh, G. C. (2000). “Understanding carbon lock-‐in”. Energy Policy 28(12): 817-‐830.
30 Mitchell, C. (2008). The Political Economy of Sustainable Energy. (Basingstoke: Palgrave MacMillan). Leggett, J. (2014) The Energy of Nations. (London: Earthscan).
31 Unruh, G. C. & Carrillo-‐Hermosilla, J. (2006). “Globalizing carbon lock-‐in”. Energy Policy, 34(10): 1185-‐1197.
32 Grin, J., J. Rotmans & J. Schot (2010). Transitions to Sustainable Development. (London: Routledge).
Our examination of the relationship between pattern languages and systemic constraints upon change is informed by the concept of stigmergy: indirect coordination of action among different agents based on the changes they make in the landscape.33 Francis Heylighen describes how agents shape their environment, or medium, as their tracks aggregate and consolidate, up to a point where the medium itself becomes the mediator that directs the agents.34 For example, ants leave pheromone traces on the ground to direct others to food sources, creating a positive feedback loop as the pheromone signal strengthens with increased traffic and in turn attracts more ants. Similarly, cross-country trails start as barely distinguishable paths made by the first people to walk them, become more visible as people travel them and eventually become persistent roads and highways that funnel the majority of traffic. Heylighen notes that the medium effectively manages communication processes, retaining the most useful pathways while the others are abandoned.
Lock-ins and path dependencies arise when the most travelled roads solidify as a dominant infrastructure from which escape is difficult: in more general terms, when norms and structures restrict choice and agency. When a behavior or a strategy is acknowledged as the fittest it becomes a benchmark, standard, or other form of institutionalized model, in such a way that other actors maximise their benefit by seeking to emulate it or directly cooperating. The major criterion for evaluating performance becomes the rate of application of the winning strategy, not its outcomes, and there are no embedded feedback mechanisms that would assess whether it remains fit for purpose through time. Obscured by the strength of this dominant trajectory, alternative routes towards system renewal and regeneration in the system become difficult to identify as such. These mechanisms underlie the establishment of internet monopolies such as Amazon, Facebook and Google, which when achieving critical thresholds of user numbers became able to shape user behaviour over the internet more widely. Those same mechanisms are also the source of many of the unwelcome and even perverse outcomes of a global economics based on the scaling up of structures and behaviours optimised for smaller scales and under very different circumstances.35 {How realistic is the behavioral pattern change? In the case with Internet monopolies, the situation may change with the development of "thinking out of the box", towards knowledge sharing. - Knowledge Sharing Feasibility - DVS}
In the forms of techno-social environments in which most present-day humans live, institutionalized models very soon become embedded in cultural codes: sometimes non-materially as accepted customs, norms and habits; sometimes, particularly as computers and computer programs increasingly structure socio-technical realities, physically as algorithms that remain hidden from view in black boxes. Where humans differ from ants is in the difficulty of changing what has become solidified into structure or coded into algorithm. Once they deplete their food source, ants begin to forage elsewhere and the pheromone track dissipates leaving room for new pathways to emerge. Human tracks are more persistent. As suggested by Heylighen, positive feedback that characterizes goal-oriented stigmergy is a great driver for both action and outcome. This however only applies until the continued feedback of following the most travelled tracks and overshadowing alternative opportunities destroys the function of the medium and so endangers the whole system.
33 Bonabeau, E. (1992). "Editor's introduction: stigmergy." Artificial Life 5(2): 95-‐96.
34 Heylighen, F. (2008). “Accelerating socio-‐technological evolution: from ephemeralization and stigmergy to the global brain”. In Globalization as evolutionary process: modeling global change: p. 284 (London: Routledge). <http://bit.ly/1yfQyZN > [Retrieved 15 November 2014]
35 Boyle, D. & A. Simms (2009). The New Economics: A Bigger Picture. (London: Earthscan).
36 Alexander, C., HJ. Neis & M. Alexander (2012). The Battle for the Life and Beauty of the Earth. (New York: Oxford University Press).
Alexander has experienced this in his own work, when established perspectives and practices in the mainstream construction industry directly conflict with the life-enhancing aims of his user-centred projects.36 The new architectural paradigm implied by Alexander’s approach has failed to appear, not for reasons to do with its inherent features, but because the discipline is trapped in the tracks of processes that simply can not accommodate it.37 More generally, the anti-patterns identified by the Public Sphere Project, describing factors responsible for the creation and persistence of various forms of injustice, inequality and oppression, illustrate how thoroughly pervasive they can be in global society, despite obviously being socially dysfunctional.38
The normalization of specific forms creates system dynamics that may be hard to perceive. The underlying structures and practices that continue to generate and reproduce the systems in which we live, whatever their initial purpose, may limit possibilities for engagement and the effects of active agency in projecting and shaping future trajectories. Consequently, actions focused on setting and achieving defined goals without relating to underlying patterns may be ineffective. Alexander opened this discussion by highlighting correlations among built forms and systemic social outcomes, and such discourse has now become evident in multiple areas of social change. Design patterns, especially when located in networks of interrelationships as pattern languages, have the potential to capture and operationalize systemic properties of relevant domains. This is illustrated in the first elements of the pattern language for the productive city, integrating principles regenerative of natural systems, presented by Howard Davis at the Purplsoc 2015 conference.39 Systemic properties and the dynamics they generate need to be made visible, and over time monitored for truth to purpose and regularly challenged. Weak signals that may announce breakthroughs as well as breakdowns, and indicate possible alternative system trajectories, need to be identified in order to disrupt predetermined dynamics and deliberately design for the Quality Without a Name. Douglas Schuler drew attention to this when including systemic orientation among the key features of pattern languages for social change in his Purplsoc 2014 keynote.
4. The Multiple Facets of Social Change
While weak signals may be inconspicuous in the face of entrenched dominant societal norms, they are nonetheless present at the margins. Edgar Morin describes this multiplicity of pathways in La Voie (the Way):
“On each continent and in each nation one can find creative bubbling, a multitude of political initiatives in the direction of economic, social, political, cognitive, educational, ethical or existential regeneration.
But everything that must be connected is yet dispersed, compartmented, separated. These initiatives are not aware of each other, no institution enumerates them, and no one is familiar with them. They are nonetheless the livestock for the future. It is now a matter of recognizing, aggregating, enlisting them in order to open up transformational paths.
These multiple paths, jointly developing, will intermesh to form a new Path which will decompose into the paths each of us will follow and which will guide us toward the still invisible and inconceivable metamorphosis.”40
For systemic change to arise, and to be meaningful and endure in the long term, it needs to occur in a variety of ways and arise from many different locations, interconnected as networks and networks of networks. Isolated silos of thought and action need to become linked in a fabric of interventions that is compatible with the diversity of appropriate forms of action, creating synergies that leverage their variety, their complementarity and the richness of possible pathways they create. {Why we need to talk about, locate and study patterns when the method and practice of Wholeness is already realised?! I can only see it as Pattern Researchers following a behavioral pattern of all other researchers showing in such a way inconsistency of their research: Inconsistency of Theory - DVS}
37 Grabow, S. (1983). Christopher Alexander: the search for a new paradigm in architecture. (Stocksfield: Oriel Press).
Public Sphere Project -‐ Anti-‐patterns < http://bit.ly/1Nz7i5O>[Retrieved 15 September 2015]
39 Davis, H. (2015). “Pattern Languages and the New Productive City”. Keynote presentation at Purplsoc 2015 Conference, Krems.
40 Morin, E. (2011). La Voie: Pour l'avenir de l'Humanité (Paris: Fayard). p 34. Citation translated by H. Finidori.
Plurality of interventions is a vital condition for systemic transformation.41 This needs to be acknowledged, and represented in a way that allows coordination of action across domains without imposing any centralized form of organization, or necessarily demanding coalition among events. Fourth generation pattern languages can accommodate multiple problem framings and describe the consequent emergent complexity in ways that allow change agents to take advantage of and build upon existing links across domains of action, whether or not those domains are aligned in terms of their internal descriptions and defined aims.
Figure 4 illustrates one way to represent and group approaches within this necessarily multidimensional landscape.
Figure 4: Interconnecting islands of agency and engagement42
41 Leach, M., I. Scoones & A. Stirling (2010). Dynamic Sustainabilities. (London: Earthscan).
42 Centola, D. 2015. “The Social Origins of Networks and Diffusion” in American Journal of Sociology Vol. 120, No. 5, pp. 1295-‐1338
Achieving cognitive understanding within cohesive domains of action is easier than achieving cultural understanding across actions logics, but not necessarily realized in practice. For example, in addition to any differences of perspective, particular groups of change agents may prioritize intervention on the ‘target’ itself (the ‘what’: quality and effectiveness of the object transformed). Others may focus on practices and processes (the ‘how’: quality of the process and experience), others still on the intentions and motivations of stakeholders (the ‘why’: quality of the intention and designs for emergence). Accordingly, the same phenomenon may be a strategic/tactical concern (how) or justification (why) for one group, and at the same time the primary object of attention and care (the what) for another. Equally, development of a tactic (how) or an object of attention (what) may be invoked as a motivating reason (why). All this makes coordination for change rather complicated. What we consequently observe is a multitude of islands of agency and engagement. Each has its own tacit knowledge and language that are effective in the specific domain on which it focuses. As Figure 4 shows, different islands either can not or do not need to align on any of the dimensions of how, why and what. However, they may converge in several of their applications.
Pattern languages as they have evolved to date enable change agents, whatever their core areas of interest, to identify their own goals and design their own preferred pathways towards them. Various types of pattern language exist, catering for multiple forms of why, what and how in diverse domains of knowledge and action. Each nurtures opportunities for learning and collaboration, and possibilities for mutually supportive action. By making visible previously covert processes and the tacit knowledge that underlies them, a pattern language provides a common vocabulary for design and so opens up new channels of communication and understanding.
The question we now pose is whether there are valid ways to interconnect pattern languages in order to provide a basis for aggregating diverse separate bodies of actionable knowledge and enabling coordinated action towards common higher purposes that reflect deeper systemic features shared across domains (Figure 5). Douglas Schuler mentions in his Pattern Languages for Public Problem Solving the need to accommodate greater levels of interconnection and sharing through syntactic formalization and creation of shared tag sets {Tag Clouds in IVAN terminology - DVS} to mark up the systemic components of patterns.43 The next step would involve finding meta-languages for tagging, according to their systemic implications, a multitude of patterns belonging to pattern languages of different types, across domains of know-how and cultures.
Figure 5: Interconnecting Pattern languages for synergy and coordinated action
43 Schuler, D. 2014. “Pattern Languages for Public Problem Solving. Seven Seeds for Theory and Practice.” Presentation to the Purplsoc Workshop 2014 <http://bit.ly/1GBPQIT> [Retrieved 10 October 2015]
5. The Systemic Orientation of Pattern Languages
Understanding the way pattern theory builds upon systems theory is fundamental to appreciating how elaboration of a fourth generation of pattern languages can support the interlaced meta-network of social action described in the previous section. Across all three prior generations of pattern languages, and in Alexander’s core conceptualization, pattern languages provide interrelated units of organization (whether physical, immaterial or relational) that aim to generate systems recognized by their users/designers as creative, healthy, living wholes sustaining multiple types of function.
Alexander wrote in a 1968 article Systems generating Systems:44
“There are two ideas hidden in the word system: the idea of a system as a whole and the idea of a generating system.”
This distinction is crucial to understanding pattern languages. The same article continues:
“A system as a whole is not an object but a way of looking at an object. It focuses on some holistic property which can only be understood as a product of interaction among parts.
A generating system is not a view of a single thing. It is a kit of parts, with rules about the way these parts may be combined.
Almost every ‘system as a whole’ is generated by a ‘generative system’. If we wish to make things which function as wholes we shall have to invent generating systems to create them.”
The article predates A Pattern Language but it clearly articulates how this systems perspective influenced Alexander’s later work45 and his notion of pattern languages as evocations and instances of generative processes fundamental to the unfolding of natural phenomena, operating in conjunction with intentional behaviours.
Alexander indeed questioned later on46 in his work whether pattern languages such as derived from A Pattern Language (in their context - problem - solution format) didn’t actually limit design to assembling fragmented structural ideas or objects (i.e. sequences of good ideas) into static forms rather than helping to create living generative entities and diachronic models (i.e. incorporating a time dimension) able to effectively address contemporary societal challenges. The question of the adequacy of the APL pattern approach for societal change was at the center of David West’s Purplsoc conference talk on Missing Pattern Languages.47 This talk provided a reminder of lessons from Alexander’s earlier book Notes on the Synthesis of Form48, in particular the idea of patterns as composable diagrams, key to the process of creating form.
The pattern as, “[A]bstract pattern of physical relationships which resolves a small system in interacting and conflicting forces,” of the Notes became a problem-solution construct as a means towards descriptive simplicity. More than ever, however, we need patterns to support our understanding of complexity and the richness of our diverse perceptions, interpretations and representations, in order to help us expand opportunities for collective interventions. Recent decades have achieved greater maturity in the understanding of organized complexity, allowing us to begin to reformulate and reintroduce the process of orientation inherent in the idea of patterns
and give it an explicit presence.
Most important for the purpose of this paper is the concept of ‘generative’, which points at emergent structure and behaviour. The notion of emergence is invoked when a system’s observed properties cannot be predicted from the known properties of and interactions among its elements.
44 Alexander, C. (2011). “Systems Generating Systems.” In Computational Design Thinking, edited by Achim Mengers and Sean Ahlquist, 58–67. Chichester, England: John Wiley & Sons. Reproduced from “Systems Generating Systems”, Architectural Design, volume 38 (December), John Wiley & Sons Ltd (London), 1968, pp. 605-‐610. Originally published in Systemat, a journal of the Inland Steel Products Company.
45 Alexander, C. (2001-‐2005). The Nature of Order. (Berkeley: Center for Environmental Structure). Alexander, C. (2003). “New concepts in complexity theory arising from studies in the field of architecture an overview of the four books of The Nature of Order with emphasis on the scientific problems which are raised”. <http://bit.ly/1MmRBli>[Retrieved 15 September 2015]
46 Alexander, Christopher. 1996. “Patterns in Architecture” presented at OOPSLA ’96, October 8, San Jose, California. <http://bit.ly/1LnXvyP> [Retrieved 15 October 2015]
47 West, D. 2015. “Missing Pattern Languages”. Talk at the Purplsoc 2015 conference <http://bit.ly/1OIC3bG> [Retrieved 15 September 2015]
48 Alexander, C. 1964 Notes on the synthesis of Form (Harvard University Press)
Emergence therefore refers in particular to the onset of a new level of functional properties in a system. A system presenting such a property is defined as complex and described as having complex behaviour.49 Complexity is a basic organizational feature of all living systems;50 its emergence is a central feature of biological and social evolution.51 Many complex behaviors are generated by simple rules of interaction, as observed for example when colonies of social insects build and maintain extremely sophisticated nests,52 or in the ‘nearest neighbour’ rules of flocking birds.53
Figure 6: Adaptive development of systems
Using the notion of emergence, “Generative patterns can be understood to work indirectly; they work on the underlying structure of a problem (which may not be manifest in the problem) rather than attacking the problem directly.”54 The pattern approach to systems, while maintaining awareness of inherent uncertainty, allows identification of relevant points of contact and interaction (articulated by the patterns) at levels other than the problem or phenomenon as a whole. It thus orients intentional behaviour in relation to systemic phenomena. Generative patterns address the physical connections, designed forms, organizing protocols and processes55 that give rise to emergence, synergies and other complex properties responsible for the wicked nature of societal
challenges.
49 Prigonine, I. & I. Stengers, (1984). Order out of Chaos. (New York: Bantam Books).
50 Jacob, F. (1989). The Logic of Life. (London: Penguin).
51 Kauffman, S.A., (1993). The Origins of Order. (Oxford: Oxford University Press).
52 Theraulaz G., Gautrais J, Camazine S, Deneubourg JL, (2003), “The formation of spatial patterns in social insects: from simple behaviours to complex structures”. Phil. Trans. R. Soc. Lond. A 361, 1263–1282.
53 Reynolds, C. W. (1987) “Flocks, Herds, and Schools: A Distributed Behavioral Model, in Computer Graphics”, 21(4) (SIGGRAPH '87 Conference Proceedings) pages 25-‐34.
54 Coplien, J. O. (1996). Software Patterns. SIGS Books. From Ing, David “Systems generating systems -‐ the architecture of Christopher Alexander (1968) < http://bit.ly/1Eq8N3A> [Retrieved 15 September 2015]
55 Jones, P.H. “Systemic Design Principles for Complex Social Systems” Chapter 4 in: Social Systems and Design, Gary Metcalf (editor) Volume 1 of the Translational Systems Science Series, Springer Verlag
Understanding how complexity acts together with intentional design in social systems also requires examination of the telos - purpose or directionality - of complex systems and how this can be influenced through design. Considering biological systems, O’Grady and Brooks distinguished three types of complex behaviours: teleological (end is defined by human or other external agent); teleonomic (end is 'programmatic', setting defined limits within which possibilities can unfold in many different ways, such as those of the genetic code) and teleomatic (end is contingent, the consequence of concomitant forces, like in ecosystems).56 Similarly from a social design standpoint, systems thinker and design thinker Peter Jones distinguished purposive systems - well structured or institutionalized social systems that embed deterministic mechanisms dedicated to prescribed outcomes - from those which are purposeful or purpose-seeking. Purposeful systems target pre-determined outcomes and fulfil specified purposes through differentiated means. A purpose-seeking system seeks to converge towards an ideal future state, and upon attainment of any of its intermediate goals then seeks another goal which more closely approximates its ideal.57 {Looks close to STA Methodology postulated "Way"-, or method-focused, and "Goal"-focused methodology of development? - DVS}
Living, non-human, systems exhibit behavior that is teleonomic/purposeful and/or teleomatic/purpose-seeking rather than teleological/purposive. Our techno-social, socio-economic and socio-environmental systems are hybrid complex systems whose directionality derives from a variety of underlying processes and forms of telos, generating different combinations deterministic and intentional behaviours, allowing variable degrees of agency.
In relation to the foregoing formulation, pattern languages provide the basis for the type of design for emergence Jenny Quillien proposes for addressing organized complexity.58 They attempt to replace purposive processes, prescriptive as far as interventions are concerned (such as prior specification by an architect of all details of how a building will be built and used), with generative processes whose nature is either purposeful (for example, when designers, builders and users co-design a building within specific parameters to produce certain types of behavior through inclusive and organic processes) or purpose-seeking (if no end is specified at the onset, but rather a ‘quality’ or property as an ideal that is then pursued through intermediate stages).
Present-day understanding of ecological systems has moved beyond static equilibrium models based on homeostasis and the teleological notion of progress towards climax formations to more dynamic notions incorporating phases of rapid change whose outcomes are contingent on interactions between the system’s internal dynamics and factors in its external environment.59 Similarly, in a social process, the development over time of hidden informal structures and the inherently unpredictable nature of emergent behaviour, particularly in turbulent contexts, mean we find ourselves in the area of purpose-seeking, needing to orient and adapt processes through a number of intermediate stages.
While searching for generative mechanisms, we can neither map causal processes to structures in any simple fashion, nor can we assume that the relation of processes to structure will remain unchanged over time. The option we do have is continuously to explore the landscape of possible solutions as it evolves. This is where the system may continue to develop new functionality, which we can evaluate in relation to desired properties. In nature, we observe a massive parallel exploration of strategies that constantly optimizes structures in relation to the fitness landscape.60 Artificial systems perform similar processes (if parameters can be accurately formulated) through simulations by recursive algorithms that explore multiple options in parallel towards successful generative relations. Studying a sufficiently wide selection of existing case studies or data, when they are available, will provide a similar type of output.
56 Mayr, E. (2006). “Teleological and Teleonomic: A New(er) Analysis” <http://bit.ly/1GilErG> [Retrieved 15 September]; O’Grady, R. T., & D. R. Brooks, (1988). “Teleology and biology”. Pp. 285-‐316 in B.H. Weber, D. J. Depew, and J. D. Smith (eds.) Entropy, information, and evolution: New Perspectives on Physical and Biological Evolution. (Cambridge, MA: The MIT Press). Unseen, cited in Abel, T. (1998). “Complex adaptive systems, evolutionism, and ecology within anthropology: interdisciplinary research for understanding cultural and ecological dynamics”. Journal of Ecological Anthropology, 2(1), 6-‐29.
57 Jones, P.H. Ibid
58 Quillien, J. 2015 “Parsing systemic change: And the Three begot the ten thousand things” in: Spanda Journal VI,1/2015. Systemic Change, Finidori, H. (Ed) (The Hague: Spanda Foundation)
59 Gunderson, L. & C.S. Holling (eds.) 2002. Panarchy. (Washington DC: Island Press).
60 Kauffman, S., 2007. “Beyond reductionism: Reinventing the sacred”. Zygon, Journal of Religion and Science 42: 903–914.
The evolution of pattern languages can also be understood from this perspective. Alexander’s meticulous study of recurrent structural patterns in vernacular architecture while constructing the first ever Pattern Language, provided a wide set of case studies, with demonstrated capacity to maintain fitness, from which to formalize knowledge in the relevant domain. The second generation built upon the format thus developed, in order to allow successful transfer and production of adaptive knowledge across different domains. This allowed exploration and optimization of solutions over an extremely wide and diverse range of users and problems. The third generation introduces self-referential processes that reconnect the structures being designed to the systems they produce, in real time and easy to observe at human scale. It thus creates an iterative loop that is both generated by the structure (relational) and generative of it, and so brings reflexivity about social processes to center stage.
This understanding is fundamental for work in domains changing so rapidly that past change in similar systems cannot alone provide the information necessary to inform effective strategies. Even if we are able to define the quality without a name as an end, we have to reintroduce processes of orientation and attention to telos, evident in Alexander’s original thinking, in order to create frameworks for guiding unpredictable dynamic processes.
Once the dimension of orientation is acknowledged and actively iterated, the broad plurality of active agency and the multiplicity of approaches discussed earlier allow a wide-ranging exploration of the landscape of possibilities that can orient the search for functionally sound generative interventions. The next step is to reveal the potential for active and adaptive knowledge development and transfer across wide and heterogeneous networks, and monitoring and orientation of the development of the systems thus generated.
6. Fourth Generation Pattern Languages
Fourth generation pattern languages respond specifically to these needs for interconnection and diachronic orientation. Different pattern languages can be semantically interconnected to provide an epistemological thread and a collective coordination medium connecting capacity for action across separate domains of application. {Realised in IVAN as man-generated Tag Clouds - DVS} As Figure 7 shows, in this way they help set the conditions for effective societal change by leveraging collective intelligence and catalyzing agency at aggregated levels.
Figure 7: Leveraging Collective Intelligence and Catalyzing Agency at Aggregated Levels
Completing Takashi Iba’s framework, the objects of design of fourth generation pattern languages are dynamic systemic forms (structures, systemic behaviours, relationships and effects). The act of design is generative: emergent, stigmergetic and self-organizing over time. The purpose of design is to connect different types of agencies and intentions across domains of practice in order that the search for functionally sound generative patterns converges on optimal solutions.
We see Pattern Language 4.0 as a system of knowledge to:
(1) Help decode and encode systems’ directionality (potential emergent behaviour and evolution of design objects)
while
(2) Mediating among multiple knowledges and cultural contexts, perceptions, interpretations and representations by semantically connecting existing pattern languages.
We here relate to pattern languages primarily as operating generative structures in recursive interaction with the systems they are intended to produce, and thus having strong systemic connotations. Summarizing the above arguments, we describe them as empirical and phenomenological tools with the capacity to:
- Investigate and capture interacting components of situations and systems, and formalize tacit practices that have generative properties;
- Create suitable media for the collective processing of knowledge about constructive responses to complex challenges;
- Render visible and accessible for iteration the feedback loops that support inquiry into the desired functionality of context-related systems;
- Thus allow collective orientation of emerging systems, at the same time facilitating the formalization, monitoring and iteration of the desired functions adaptively provided by a system.
To accommodate the last aspect we added a fourth attribute to Takashi Iba’s comparison matrix: orientation. It refers to the ability to actively monitor and orient our interventions within the systems being produced, in a process of successive approximation to desired and iteratively formulated qualities that are understood as emergent. It can also be understood as the capacity for collective interpretation of the evolving directionality of the resulting system (which is different from the purpose of the design). This is a fundamental aspect of working with increasingly complex systems and wicked problems - as is the case in societal change - where direct interventions are part of knowledge building in the field and demand continuous processes of observation, interpretation, orientation, design, action, monitoring and adaptation. Processes of this kind are involved in frameworks such as Charles Pierce’s cycle of pragmatism61 (observation, induction, abduction, deduction, testing, action), John Boyd’s OODA loop for situational awareness62 (observe, orient, decide, act) and more recently, Dave Snowden’s Cynefin approach to complex adaptive systems63 (probe, sense, respond).
61 Peirce, C. S. (1903). “Pragmatism as a Principle and Method of Right Thinking”, The 1903 Lectures on Pragmatism, ed. by P. A. Turrisi, SUNY Press, Albany, 1997. Also in [18], pp. 131-‐241.
62 Boyd, J. (1996). “The Essence of Winning and Loosing”. Slideshow <http://bit.ly/1MomGyU>[Retrieved 10 October 2015]
63 Kurtz, C. F. & Snowden, D. J. (2003). “The new dynamics of strategy: Sense-‐making in a complex and complicated world” in IBM Systems Journal, Vol 42, NO 3 pp 462-‐483
Figure 8: The Four Generations of Pattern Languages
In the newly proposed matrix of comparison we see that while
Pattern Language 1.0 is about orientation of the practice of design as a system of meaning with a purpose,
Pattern Language 2.0 is about orientation of collective problem-solving while dealing with technical performance and efficiency of the objects transformed,
Pattern Language 3.0 is about orientation of the continuous design of interactions as a form of engendering creativity (like in collaborative discovery and sense-making),
Pattern Language 4.0 is about orientation of the collective interpretation and navigation of system directionality as an enactment of agency.
PLAST: a Concrete Example of a Fourth Generation Project
The PLAST project aims to develop a pattern language of the fourth generation.
PLAST is a collaboration of individuals and communities of practice (pattern language practitioners, permaculturists, change agents, facilitators, educators and designers) and scholars in various disciplines (pattern languages, complexity theory, network theory, systems thinking, mathematics, algorithmic design, economics, linguistics, graphic design...) gathered in an open source network to create tools and methodologies for collective interpretation of systems’ behaviour and directionality, and to share experiences, while understanding each other across domains of knowledge and culture, in order to have more impact in bringing about societal
transformation.
PLAST - A Collective Interpretation and Systemic Orientation System
The sets of tools and methodologies PLAST aims to develop comprise a visual language (vocabulary, grammar, syntax) composed of systemic interpretation elements;64 systemic patterns and various semantic connectors to encode and decode systemic orientations, structure and interactions, emergent dynamics and behaviors; and associated heuristics in order to assess and evolve systemic design.
PLAST will be operationalized offline through design and facilitation methodologies that will support both the design of the PLAST language and artifacts and a process for hermeneutic inquiry based on action research, shared discovery and mutual recognition across communities of practice and experience. It will thus enhance systemic awareness and literacy. The approach is part of a peer-to-peer learning process that will help communities distinguish and understand different perspectives and engagement logics in order to appreciate their differences and complementarities and federate a diversity of actions.
A first set of components sourced both from theory and pilot practice has been tested in workshops, including during the PURPLSOC conference, which helped generate systemic stories that enabled comparisons and rich discussions about various systemic conditions and situations. PLAST will grow organically by expanding through community encounters during conferences, workshops, and hackathons.
The language itself - and in particular the systemic interpretation elements - will be used for indexing and connecting various elements, and for mapping the web of pattern languages and domains of social change. Envisioned technological applications include standards for indexing solutions and pattern databases {How it relates to the Server of Ready Solutions at IVAN? - DVS}, qualification or tagging of projects and models, and indexing algorithms as a systemic interoperability standard, in addition to indexing pattern languages.
64 Finidori, H. (2014). “A Pattern LAnguage for Systemic Transformation (PLAST) -‐ (re) Generative of Commons”, in Baumgartner, P & Sickinger, R (Eds) PURPLSOC: The Workshop 2014 (Krems: Conference Edition) <http://bit.ly/11xD2oF> [Retrieved 13 December 2014]
PLAST will be operationalized online mainly as a digital platform comprising a repository of various layers of patterns and pattern languages by domains of application or function (such as PatternPedia65), with a social network component to evaluate and discuss problematiques and patterns, along with design tools that support synchronous and asynchronous co-design (graphic as well as text/code), forking and merging, migration, combining, sequencing, such as currently enabled by Github66 or Ward Cunningham’s Smallest Federated Wiki.67
PLAST will strive to enable the following experiences and capabilities: Towards a Fourth Generation Pattern Language.
Case Study: PLAST
Conclusion: A Code for Systemic Transformation
Fourth generation pattern languages such as PLAST can become emancipatory tools, as they can bring to light knowledge and understanding about the workings of interconnected techno-social, socio-economic and socio-environmental systems that are normally hidden from view and which, when consciously articulated, allow people to propose and act upon ways to transform the directionality of these systems and, hence, their own situation. The approach of decoding and encoding is a way of ‘hacking the system’, to understand its components and interactions in order to build anew or transform.
65 Fehling, C., Barzen, J., Falkenthal, M., & Leymann, F. (2014). “PatternPedia. Collaborative Pattern Identification and Authoring”. In: Proceedings of Pursuit of Pattern Languages for Societal Change -‐ Preparatory Workshop 2014.
66 Github features: <https://github.com/features>
67 Smallest Federate Wiki Welcome Visitors page <http://bit.ly/1kri2tI> [Retrieved 15 September 2015]
Alexander saw pattern languages as tools for supporting what he referred to as System-A, a life-centred, human scale alternative to the globally dominant social, political, economic and cultural configuration he called System-B.68 Discussions on the feasibility of systemic transformation, on structure versus agency, have been going on for a long time in many disciplines.69 French philosopher and sociologist Jean Baudrillard defined an ungraspable order in the System, a simulacrum made of indefinite chainings of simulations he called ‘the code’.70 This code, in its unbounded (de-finalized in Baudrillard’s own terms) space-time, dissolves and absorbs any final cause for change such as political economy or revolution. In this logic, trying to ‘fight’ The System from within is vain, because any attempt will be neutralised by dilution in the code: the medium taking control of the agent as described earlier in the paper.71
For Baudrillard, dissent could only succeed by transcending ‘the code’: when applied at a higher logical order than that it wishes to overcome. It is difficult to imagine in this logic, with an open ended medium that is not ‘circumscribable' and that we can’t step out of to 'transcend' or 'overcome’, that this could happen through anything else than a totalitarian enterprise. Hence the deadlock he evoked. We suggest that dissent could materialise through the insertion of new pieces of code at multiple levels and scales via distributed praxis, in an approach akin to hacking, in order to 'dilute' the existing dominant code in an emergent process.
This is probably what Alexander had in mind when he noted,72 in his talk at the Object Oriented Programming, Systems, Languages and Applications conference in San José in 1996, the increasing influence of computer programs in shaping almost all aspects of the world in which we now live in (for example in manufacturing, transportation, construction management, diagnosis, printing or publishing…even before the advent of social networks and their increasing power to orient choices and shape behaviour), and deplored the lack of discussion about how programs and therefore code could help change the world. Software programmers, he argued, have a
responsibility to generate living structures in the world and could write code as sets of instructions for generative processes and procedures.
As a set of ‘meta’ languages, used to encode and decode systemic directionality, emergent behaviour and evolution of design objects, pattern languages of the fourth generation could well pave the way towards achieving this purpose. They would help make visible the various hidden ‘codes’ of the system and enable a hermeneutical inquiry of systemic, semantic and paradigmatic nature that would help the emergence of new more generative encodings. Unlike conventional code, which usually operates in black boxes, this ‘code’ would operate in plain view.
68 Alexander, C., HJ. Neis & M. Alexander, (2012). The Battle for the Life and Beauty of the Earth. (New York: Oxford University
Press).
69 Bateson, G. (1979). Mind and Nature: a necessary unity. (New York: Dutton). Giddens, A. (1984). The constitution of society: Outline of the theory of structuration. (Cambridge: Polity Press). Sewell Jr, W. H. (1992). “A theory of structure: Duality, agency, and transformation". American Journal of Sociology, 1-‐29.
70 Poster, M. 1988. Jean Baudrillard Selected Writings. 1st edition Stanford University Press p 110 to 148
71 Poster, M. Ibid
72 Alexander, Christopher. 1996. “Patterns in Architecture” presented at OOPSLA ’96, October 8, San Jose, California.
<http://bit.ly/1LnXvyP> [Retrieved 15 September 2015]
Comments (0)
You don't have permission to comment on this page.