Antifragile planning

Abstract

We argue that antifragility is a valuable and contentful goal for planning. We present a possible definition and outline the tenets and essential properties of an antifragile planning and compare it with approaches of urban resilience. We further present an argument for the legitimacy of an antifragile planning, by exploring its possible conceptualisation in terms of the capability approach. Hence the recommendation to incorporate antifragility into planning practice and content.

Keywords

antifragile planning antifragility complexity urban resilience capability approach

Introduction

The complexity of urban systems, of which cities are but the most plentiful navels, adds trickiness to the already burdensome responsibility and political implications of planning. Other than making the evolution of urban systems unpredictable in the strong sense (Hilier, 2012; Moroni, 2015a), such complexity also brings about greater uncertainty of ultimate outcomes of policies and actions, raising both operational and deontological concerns for planning (Chettiparamb, 2019; De Roo and Hillier, 2012; Innes et al., 2010; Moroni and Cozzolino, 2019; Portugali, 2006, 2008; Portugali et al., 2012).

To address these concerns, we here advance the proposal that the concept of antifragility (Taleb, 2012) offers a contentful and legitimate partial goal for planning. To appraise the merits of the proposal, in this article we attempt to do the following:

First, we situate our discussion in the more general debate on the problem of policy and agency under conditions of complexity of urban systems;

Second, we outline the essential properties of antifragility, and contrast it with the concept of resilience. We argue that, in general, resilience should be understood as a limit case of antifragility, and therefore that the distinction between antifragility and resilience is theoretically relevant and operationally useful;

Third, we argue that antifragility is a valuable goal for planning, and discuss its affinities with approaches of planning for resilience, to claim that antifragility should be elected as a more general and superordinate goal for the government of urban systems;

Fourth, we sketch out what the main tenets and features of an antifragile planning may be;

Finally, we examine the conditions of legitimacy of antifragile planning, and argue that for some of its tenets antifragile planning can be assumed as a partial condition of legitimacy of planning, beyond being just a question of mere policy preferences.

Before beginning in earnest, we want to clarify that throughout this article we use that overworked word, planning, without reference to any specific set of tools, norms, techniques for the construction and design of spatial plans, but rather more in general as a stand-in for the practices of governing and collective management of urban systems.

Complexity of urban systems and antifragility

Complexity redoubled

The complexity of urban systems is twofold, the two modes of complexity compounding each other. One mode is in the ‘simple’, purely mechanical sense of them being large ‘many-body systems’ (Anderson, 1972), composed of many components interacting in a non-linear fashion. By way of these interactions and feedbacks, such systems are capable of exhibiting forms of spontaneous order and of emergent properties at higher levels of hierarchy. The second mode of complexity of urban systems is due to them being social systems, having autonomous social agents among the systems’ components (Hilier, 2012; Moroni and Cozzolino, 2019; Portugali, 2000, 2006, 2012). If not ultimately ontological, this autonomy cannot but be presupposed as a moral, and an operational assumption.

The redoubled complexity of urban systems brings about an unavoidable uncertainty to the planning practice. It raises problems of what the normative content of planning may legitimately and realistically be, and casts a shadow over certain planning practice overconfident about the effectiveness of too specific goal-oriented attitudes. Among attempts of conceptual overhaul of planning theory and practice to deal with the complexity and uncertainty of urban systems, a prominent, and thriving, approach is that of urban resilience (Davoudi et al., 2013; Davoudi and Porter, 2012; Meerow et al., 2016), together with that, somewhat related, of adaptive planning (Kato and Ahern, 2008; Rauws, 2017; Skrimizea et al., 2018). Our key claim in this work is that the notion of antifragility developed by Nassim N. Taleb (2012) introduces a fruitful conceptual framework to engage and broaden the general coordinates of the debate on planning under complexity and uncertainty.

The idea of antifragility

To define antifragility, we ought to situate the concept within a general examination of the ways in which things (objects, organisms, institutions, systems) respond to uncertainty, to events, perturbations, stressors, volatility, disorder – in short, to time. Here, the conceptual triad ‘fragile, robust/resilient, antifragile’ proposed by Taleb offers a theoretical framework useful to clarify things.

Following Taleb, the essential property of something fragile is that events, perturbations, stressors, volatility, disorder – in short, time – can only harm, damage, break or destroy it, and never benefit it. This does not mean that every perturbation is necessarily harmful. Rather, the above definition of fragility states two conditions which simultaneously need to hold that there is (1) the possibility of harm, and (2) no possibility of gain from perturbations.

On the other hand, something is robust or resilient if it is largely indifferent to perturbations. In the case of robustness, the perturbations do not affect it, they leave it as it is. In the case of resilience, it can absorb, bounce back and recover from perturbations. When attributing the property of robustness or resilience to something, it is always in relation to a certain threshold of intensity: things are robust or resilient always only up to a certain intensity of perturbation. Strictly, an urban system should be said to be resilient if it is capable to absorb shocks, perturbations and volatility, and to recover and bounce back to its prior equilibrium or to its functional equivalent. With this, we are deliberately narrowing the definition of resilience from a more extended meaning often encountered in the planning literature, not only limited to recovery, bounce back and restoration of prior equilibrium. Reworking this distinction between ‘strict’ and ‘extended’ resilience, by framing the latter as antifragility, is one of our key points we return to below.

Antifragility is different. Taleb proposes to call something antifragile if it can from time – from events, perturbations, stressors, volatility, disorder – also gain, get stronger, improve, evolve and adapt better. To use Taleb’s brief formula: antifragile gains from disorder. Analogous to fragility, not every perturbation is necessarily gainful; some, perhaps the majority may be inconsequential but – contrary to fragility, robustness and resilience – some can be.¹

In short, to assess if something is fragile, robust, resilient or antifragile means to examine its possible responses to stressors, perturbations and volatility, and to place those responses somewhere along a harm-gain dimension.

Resilience as a limit case of antifragility

However, besides along the harm-gain dimension of response to perturbations, there is another way useful for our discussion to arrange the four terms. In fact, when practically conflating robustness and resilience, Taleb somewhat misses to point out one relevant distinction. Indeed, along the harm-gain dimension, Taleb places:

fragile on one extreme (‘possibility of harm, never gain’);

robustness and resilience together in the middle (‘neither harm, nor gain’);

antifragile on the opposite extreme (‘possibility of gain’).

But the other way to arrange the four terms is to observe that, from the outside where the perturbation is originated, the objects we are classifying may appear either static (irresponsive, passive) or dynamic, endowed with some internal mechanisms of reaction. If we intersect the two dimensions (harm vs gain and static vs dynamic), as in Figure 1, we may derive some noteworthy insights for our subsequent discussion.

Figure 1.

Response (gain-harm) – responsivity (static-dynamic) matrix.

The scheme in Figure 1 helps to observe that robustness is the neighbour of fragility (Box A), while resilience is the neighbour of antifragility (Box B). Specifically:

robustness is the limit case of fragility: both are internally static, but with different thresholds of intensity of perturbation for harm;

while resilient systems are the limit case of antifragile systems: both are internally dynamic, but with different possible gain from perturbations: none or only small for resilient, possibly large for antifragile systems.

Furthermore, the scheme helps to clarify that things Taleb calls antifragile may be of different nature from our static versus dynamic perspective. A system endowed with an internal dynamics and operating mechanisms may be antifragile. But some internally ‘static’ objects (e.g. contracts, bets, heuristics) may also be said to be antifragile. Indeed, Taleb offers many examples of both kinds.² So the scheme in Figure 1 clarifies that not everything antifragile is necessarily internally dynamic nor complex, but also that not everything complex is necessarily antifragile. In the case of urban systems we are of course dealing with the second column, their potential antifragility rooted in their complexity (Blečić and Cecchini, 2017).

Antifragility versus resilience in planning

For planning, the usefulness of the conceptual triad fragile-robust/resilient-antifragile goes beyond its analytical relevance and offers also some operational guidance we want to explore. The problem of efficacy in classical planning arises, among others, out of its overly demiurgic reliance on hard predictions (Hilier, 2012; Moroni, 2015a), that is, on the ambition to predict with accuracy and precision what will and has to happen, when, where, with what intensity, as a consequence of what planning action, often to produce ‘teleocratic’ outcomes (Alexander et al., 2012). If, in the light of the redoubled complexity of urban systems, hard predictions are indeed hard, must we not instead settle for something less? Taleb’s triad points at such a different possible attitude in the planning practice: rather than aspiring to obtain hard predictions, we can, and cannot but, settle for something we could call soft predictions, where the point is not to accurately and precisely predict what will happen, when and where, as a consequence of what planning action; rather the point is to examine what are the system’s possible responses to perturbations, volatility, even low-probability events, in other words to detect its fragility, robustness, resilience or antifragility, and to explore what makes it such. To decision-makers, and thus by extension to planning practitioners, such a soft prediction is, as Taleb rather convincingly lays out, more accessible a practice than making hard prediction.

In comparing antifragility with resilience, it is beyond our scope here to explore with greater detail what makes systems antifragile in general. Both antifragility and resilience are richer in meanings, implications and uses, and there are still notable theoretical and operational issues to the application of resilience in planning (Davoudi et al., 2013; Davoudi and Porter, 2012; Meerow et al., 2016) and in public policy in general (Duit, 2016). However, the essential property of ‘gaining from disorder’ in antifragility, absent from resilience, should be consequential enough to let us lay out the argument that antifragility is a contentful goal for planning, distinct from resilience.

To be clear at the outset, we share the view that resilience is a justifiable policy goal. Kolers (2016) has persuasively shown that resilience can and should be incorporated as a necessary condition of legitimacy and justice of public policy. So urban planning as a field of public policy should incorporate resilience into both its methodology and content.

Our first point on antifragility vis-a-vis resilience in planning is a plea for greater terminological precision. It is often the case that things get called resilient without them being cases of resilience. It may be a question of nuances and, after all, everyone is free to choose the words to use and to come to an agreement with others on their meaning. However, it should make one pause a moment when often one notices the need to supplement the term resilience with certain adjectives – such as ‘adaptive’, ‘transformative’, ‘evolutionary’ (Davoudi et al., 2013; Davoudi and Porter, 2012) – which, strictly speaking, pertain to antifragility. We already mentioned this distinction between a ‘strict’ and an ‘extended’ use of the term resilience. There indeed seems to be a significant encroachment between what in the current planning literature is understood as planning for resilience on one hand, and what we propose to call antifragile planning on the other. So, should we dismiss this distinction as just a nominalist squabble, a sterile pedantry, so that we better throw in the towel, drop antifragility and embrace the extended meaning of resilience?

It though does seem to us appropriate to advance a modest plea to sharpen up our common conceptual apparatuses, if only to enrich our lexicon and make a more perspicuous use of it, and to start employing the term ‘antifragile’ when we essentially talk about antifragility, and ‘resilient’ when we talk about resilience. There are of course many worthwhile nuances, but we hold that the guiding principle to settle the distinction should be that resilience proper be understood as a limit case of antifragility, as we suggested above.

This, to some extent, may also be a way to ‘save’ resilience from the danger of becoming an empty buzzword, so voracious that after swallowing everything it means nothing. As Davoudi and Porter (2012) have already observed:

resilience appears to be fast replacing sustainability as the buzzword of the moment. It may well follow a similar fate and become a hollow concept for planning: an empty signiﬁer which can be filled to justify almost any ends. [. . .] Anything that is uncritically and easily accommodated into our lexicon demands close investigation to clarify what ends it is serving. (p. 329)

In this sense, for instance, it seems to us more productive to frame the so called ‘evolutionary resilience’ (Davoudi et al., 2013; Simmie and Martin, 2010) as antifragility. Indeed, if, following Davoudi, ‘evolutionary resilience’ is characterised by (1) persistence (being robust), (2) preparedness (learning capacity), (3) adaptability (being flexible) and (4) transformability (being innovative), the last three terms clearly refer to salient features of antifragility, that ‘building capacity for envisaging and embracing transformation through creativity and imagination at institutional, community and individual levels and thus cultivate flexibility, resourcefulness and cooperative networks at various scales’ (Davoudi et al., 2013: 319).

This hope for greater terminological clarity leads us to our second and more substantial claim on antifragility vis-a-vis resilience. To pinpoint one fundamental difference between antifragility and resilience, we want to put focus on one core property of antifragility: optionality. It is the property of having options (possibilities, rights, entitlements, capabilities to do, to have, to become, to change the course of action, to reverse prior decisions, etc.), but not obligations. The greater the optionality embedded in a course of action, the larger the possibility of a favourable asymmetry between the action’s upsides and downsides. Optionality is an essential property of antifragility because having options, also after undertaking a course of action, allows to benefit from unpredicted and unpredictable opportunities, while limiting the possible harm arising from threats.

Optionality draws a watershed between antifragility and resilience: while at the core of antifragility, optionality in the strong sense is absent in resilience. We say ‘in the strong sense’ to be more careful in defining the distinction. Indeed, the goal of resilience may be pursued for institutions, services, infrastructures and environmental systems which are valuable also for providing certain optionality to people. A public transit system quite obviously increases the optionality to move and reach different parts of a city, and the availability of nearby natural environments offer options for recreation. But in this weak sense, resilience at its face value does not contemplate the possibility that these institutions, services, infrastructures and systems evolve, improve and gain, even in terms of their purpose of providing optionality, specifically from the unexpected opportunities with time. For that, the goal of antifragility must be put to work.

Let us further clarify the distinction with an example. There is one notable piece of ‘planning theory’ which we hold helps to mark the difference between resilience and antifragility. When, in her prodigious masterpiece The Life and Death of Great American Cities, Jane Jacobs (1961) presents the conditions for vibrant and safe neighbourhoods and examines the cases which possess these conditions, she goes on cautioning us against possible jeopardies, and suggests defensive measures in order to preserve the neighbourhoods as vibrant, lively and safe. In other words, she advises us on how to pursue a goal of resilience.

But when Jacobs recommends on building new neighbourhoods and on unslumming the existing ones, and suggests how to increase the chances of getting there, she is reasoning in terms of antifragility. In fact, she proposes a repertoire of measures³ for creating favourable conditions so that autonomous individual and collective actions may have the greatest possible chances to bring about the evolution in the desirable direction (Cozzolino, 2018). If the aim is to have such-and-such neighbourhoods, we need to build in resilience in the neighbourhoods that are already such-and-such, and we need to build in antifragility if they yet need to evolve and become such-and-such.

But to all this, there is an additional, normative level also present in Jacobs’ argument. She does not only tell us how to make neighbourhoods vibrant, but also why we would want them vibrant in the first place. Apart from just creating hip, cappuccino-sipping Stadtluft for the sake of it, Jacobs tells us that such neighbourhoods are generators of diversity, innovation, business opportunities and culture. Places where new things, fostered by the encounter of diverse urban populations and by entrepreneurial and social experimentation, are more likely to emerge, to be experimented and to flourish. In this sense, such neighbourhoods are engines of antifragility.

So antifragility in Jacobs shows up at two levels. On one level, to create new neighbourhoods or to unslum the existing ones, she recommends we proceed by strengthening their antifragility, their ability to gain from the unpredictable, from the spontaneous individual actions and self-organising. On the second, normative level, if we indeed need to work in the direction of making successful neighbourhoods resilient, by protecting their vibrancy against the jeopardy of specialisation and loss of diversity due to their very ‘success’, it is in view of that antifragility of successful neighbourhoods at the second level, of them being engines of antifragility.

In our view, Jacobs does not only clearly envision cities in terms of antifragility, but also maintains the distinction, we are attempting to bring to fore, of antifragility and resilience being two different objectives. In general, antifragility is promoted by whatever opens and increases the possibility of local experimentation and tinkering, of new combination of uses in relation to new demands and pressures, of learning from trial-and-error, without introducing the threats of systemic risks from failure, when this experimenting is not just a natural random process but due to actions of individuals who are themselves risk-takers. Conversely, whatever reduces such possibilities, either through prohibitions, regulations, or due to the lack of diversity and of a sufficiently granular plurality of uses and practice on the ground, can be said to fragilise neighbourhoods and cities.

In Table 1, we attempt to summarise the distinction between resilience and antifragility by contrasting some key concepts, phenomena and practices we discussed in the text, or drawn from Taleb.

Table 1.

Resilience versus antifragility.

	Resilience	Antifragility
General	‘No harm, no gain from disorder’ (bounces back to prior equilibrium or a functional equivalent)	‘Gains (sometimes) from disorder’
	Provides no optionality (in the strong sense)	Provides optionality (in the strong sense)
	Redundancy^a	Functional redundancy (degeneracy in biology)^a
	Phoenix (mythology)^a	Hydra (mythology)^a
Examples in planning	Persistence, preparedness, reactive adaptation	Transformative, evolutionary, proactive adaptation, innovation
	Jane Jacobs’ ‘defensive measures’ against jeopardies to urban diversity	Jane Jacobs’ unslumming strategy
	Restoration of buildings	Exaptation of buildings
	Participatory design	Tactical urbanism
	Consensus building	Agonistic pluralism

From Taleb (2012: Table 1).

Towards an antifragile planning

Our Jane Jacobs example should suffice to make it clear we do not claim there is a new, yet unseen mode of governing urban systems we should dub ‘antifragile planning’. Rather, our claim is that antifragility, and its distinction from resilience, offers a valuable conceptual framework. In this sense, its role is similar to that which Swanstrom (2008) assigns to the concept of resilience itself: ‘more than a metaphor but less than a theory. At best it is a conceptual framework’. Indeed, many existing planning practices and theoretical proposals, on different spatial scales, implicitly practice antifragility. So, altogether, our proposition is more modest, to assess the usefulness of the concept for the debate on planning under uncertainty, and to suggest it as an umbrella term embracing parts of different proposals which in our view on at least some accounts pursue the goal of antifragility. These go from some formulations of strategic planning (e.g. Albrechts, 2010; Albrechts and Balducci, 2013), to evolutionary resilience (Davoudi et al., 2013; Davoudi and Porter, 2012; Meerow et al., 2016), to adaptive planning (Ahern, 2011; Kato and Ahern, 2008; Rauws, 2017; Skrimizea et al., 2018), to tactical urbanism (Silva, 2016; Wohl, 2018), to those which explicitly reference antifragility (Fusco et al., 2017; Las Casas and Scorza, 2017; Roggema, 2019).

Let us attempt to delineate some essential properties of an antifragile planning.⁴ The goal of antifragility in planning could have two different meanings. One is to understand it as pursuing antifragility for the object of planning. In this sense, an antifragile planning would mean to do planning to make urban systems more antifragile, and to avoid making them more fragile. We may call this way of understanding antifragile planning ‘planning for antifragility’.

The second way of understanding antifragile planning is to think of pursuing the goal of antifragility for the planning itself, for its practices and outputs. In this second sense, the pursuit would be to make urban planning decisions, their decision-making processes, procedures, policies, regulations, management and spatial plans, themselves more antifragile and less fragile. We may call this second way of understanding antifragile planning ‘antifragility for planning’.

The two meanings are not operationally unrelated. An antifragile planning policy – designed sufficiently open-ended and flexible, and embedding enough optionality, in order to embrace and benefit from opportunities, spontaneous social energies, entrepreneurship and from yet unpredictable or low-probability, circumstances – could both advance the goal of antifragility of the urban system it ‘acts upon’, while being itself antifragile. Therefore, while the distinction may be analytically useful, and there could be separation between operational criteria for ‘planning for antifragility’ on one hand and for ‘antifragility for planning’ on the other, what we propose to call antifragile planning should be antifragile in both our senses. Conversely, a ‘fragile planning’ is fragile in both these senses: it fragilises the urban systems it acts upon, and its processes, outputs and actions are themselves fragile.

Fragilisers

Since other than what to do, antifragile planning is much about what better avoid doing, we want to begin from a set of attitudes and practices of intervening on social systems in general, and on urban systems in particular, which may fragilise them. Here is a non-exhaustive repertoire of such fragilisers, largely known in the planning and public policy literature.

Plans and policies based on fragile predictions. This is a category of fragilisers implicit in many of the more specific ones that follow. This category of fragilisers is due to the idea that the purpose and task of planning is to strongly determine specific spatial and social arrangements and outcomes, therefore requiring hard predictions of what the city will be like and how it will evolve and react to planning interventions in the mid- to long-run. Given that urban systems are in many aspects intrinsically unpredictable, should the efficacy of planning decisions hinge on hard and hence fragile predictions (Batty, 2007; Moroni, 2015a),⁵ the decisions would be themselves fragile, in both the senses we have given above.

Excess of centralisation-cum-micromanagement. Centralisation increases the risk of blow-ups, catastrophic failures and jeopardies to the survival: one bad decision applying to all can wreak havoc on the entire system. This also applies to large investments and megaprojects (Flyvbjerg, 2017; Flyvbjerg et al., 2003), where holds that ‘big is fragile’ (Ansar et al., 2017). Instead, decentralised decision-making may lead to only local errors, which in principle are harder to propagate to provoke systemic failures. Furthermore, decentralised decisions favour greater experimentation and tinkering and foster innovation through learning by trials and errors (Taleb, 2012). To be clear, a policy preference may make centralisation justifiable for the efficacy and expediency of pursuing certain policy goals, for example, equity or uniformity of conditions, opportunities or outcomes, or due to large-scale economies, network effects and fixed costs of investment. However, centralisation threatens to fragilise the system and to jeopardise its antifragility, especially when the centralised action aims not only to design a general frame of reference for individual and local action, to grant rights, to provide universal public goods and to solve collective action problems in the case of infrastructures with large network effects, but also to micromanage the detailed workings of the system, in all its parts.

Efficiency and optimisation. There are cases in which it is worthwhile and possible to pursue efficiency and optimisation of certain subsystems, services and processes. But this is an uncontroversial effort only if the following three conditions are simultaneously satisfied: (a) the general purpose of the subsystems, services and processes to optimise and make efficient is uncontroversial (uncontroversial purpose), (b) their relevant performance is measurable on only a single criterion (mono-criterion performance) and (c) the outcomes of the effort to optimise and make them efficient are predictable (outcome predictability). Even letting aside the condition of outcome predictability, is such a simplifying teleology, grounded on uncontroversial purpose and mono-criterion performance, a meaningful and feasible pursuit in the case of urban systems in general? Urban systems are rather a common platform for the ‘heterogony of ends’, a shared spatial, social, cultural and economic context within which different agents choose and pursue their ends and life plans through the encounter of human agency with the multiplicity of opportunities offered by the urban system, by its physical fabric, by the way in which it is organised, designed and regulated, and by the social practices of use of space.

The fragilising power behind the thrust for efficiency and optimisation arises especially when it is focused only on the immediate first-order effects, because such efforts may trim down the optionality, remove safeguards and protective redundancies, reduce the opportunities of exaptation (Johnson, 2010) and thus shrink the possibility of evolutionary adaptations, of change of uses, and of embracing the multiplicity and heterogeneity of ends and needs – present and future (Zellner and Campbell, 2015).

4. Specialisation. Excess of specialisation is a special case of trimming down the optionality, for it makes the system fragile to external perturbations and reduces its capacity to adapt and evolve with the changes in the environment. This has long been pointed out by Jane Jacobs (Ellerman, 2005). Examples of one-company towns or places with tourism monocultures come to mind. The case of the historical city of Venice overran by tourism, functioning less and less as a city and more and more as a one big open-air amusement park, is an exemplary contemporary cautionary tale.

5. ‘Extractive’ (rather than inclusive) political and economic institutions, and the crumbling of the ‘cement of society’. The first fragiliser (Acemoglu and Robinson, 2012) removes the conditions for political and economic participation, does not provide incentives for investment and innovation, and violates the principles of accountability and publicity (Rawls, 1996). The second is a fragiliser of social cohesion, by which we do not mean a stationary state of ‘harmony’, but a dynamic, ultimately precarious, outcome of conflicts, reciprocal accommodations and partisan mutual adjustments. An example of such a fragiliser is when excessive economic inequality, coupled with particular institutional arrangements, spills over into the inequality of real opportunities, capabilities and possibility to meaningfully participate in the democratic political process, undermining the social cohesion from within (Sandel, 2012).

The three planes of antifragile planning

The perspective of antifragility allows us to distinguish three operational planes for the planning practice: (1) the via negativa, (2) the shared vision and the ‘coordination by means of future’ and (3) the space of the projects. We envision these three operational planes operating on different time, spatial and institutional scales, from long-run and high-level (regional and above) of the via negativa, to short-term and strictly local of the space of the projects.

Via negativa

Building on Taleb, by via negativa in planning we mean a set of mainly negative general rules providing ‘external’ restrictions and prohibitions delimiting the space of possible actions. Largely nomocratic in nature (Alexander et al., 2012; Moroni, 2010), they follow the logic of via negativa insofar as they provide proscriptive rather than prescriptive rules (Hakim, 2014), do not directly predetermine the outcomes, do not impose performative behaviours, do not indicate what to do and what should be done, do not tell what should happen, but mainly what is forbidden and what should not be done. Moreover, the concept of via negativa also refers to the removal of what could be harmful and of the superfluous, which wastes social and human energies (Moroni, 2015b), from counterproductive constraints to procedural and normative superfetation.

A prima facie content of such a via negativa in planning is to be looked for in avoiding fragilisers we described before, and in embracing the ‘coding turn’ in planning (Alfasi, 2018). For that, the via negativa constitutes a core of what it takes to have an antifragile planning, and its political legitimacy may be, at least partially, independent from mere policy preferences.

However, under the tenet of via negativa there is also place for policy options subject to democratic deliberation. The idea of via negativa does not imply a withdrawal into a planning night-watchman state or the maintenance of the status quo in the legal and institutional settings. As long and the regulation observes principles of generality, and retracts from short-term contingencies and conveniences, the via negativa does not exclude the possibility of structural transitions and ‘changes of regime’ if that removes the sources of fragility. In other words, if we are operating through general and abstract rules – without the pretence to bridle, over-control and micro-manage the dynamicity of the system, its capacity of self-organisation and the capability of agents for dynamic adaptation – different policy options become admissible. For example, a different regime of the ius aedificandi and land property rights, or the introduction of fiscal tools for land value capture (Ingram and Hong, 2012).

Shared vision and ‘the coordination by means of future’

A planning firmly tied to hard predictions is fragile, but a planning which does not ‘tend towards future’ and does not ‘create future’ is a contradiction in terms. It is reasonable to expect that a political community would care for its future collective outcomes, at least on a time horizon accessible to one’s imagination and circuit of care of three to four generations.

For this, there is space for deliberation on a reasonably shared vision of desirable futures and of those to avoid, through a strategic decision (Albrechts, 2010; Albrechts and Balducci, 2013). Such a shared vision is a concrete declination, in a precise historic moment, based on available resources, of the set of different freedoms which compose the right to the city. To define such freedoms, we hold that the Senian capability approach (Sen, 2009) can provide a theoretically and practically feasible normative focus, compatible with the goal of antifragility. That is so, because extending capabilities provides space for actions and projects by economic and social agents and by the public institutions, thus making urban systems more antifragile, without fragilising individuals, and indeed expanding their optionality. This shift from an outcome-oriented to a capability-oriented planning could be operationalised through the idea of ‘urban capabilities’ (Basta, 2016; Blečić et al., 2013, 2018). Besides providing a framework to cast into the normative content of an antifragile planning, we argue in the next section that the capability approach can also play the pivotal role in providing the basis for the legitimacy of antifragility in planning.

For the deliberation on the shared vision, the practice of construction of future scenarios may prove useful. To clarify how one could devise the purpose of ‘imaginary’ future scenarios if the future is in principle unpredictable, we reiterate the proposition (Blečić, 2013) that the practice of construction of future scenarios is something different from future scenarios in themselves: the purpose of the practice of construction of scenarios is not only, nor primarily, to predict, but rather the construction of meaning and shared knowledge of possible futures, which the agents participating in the process of scenario construction can appropriate, in order to mobilise for action and for collective coordination (Godet, 2001). The exceptionality of social systems resides in the unavoidable orientation towards the future of the agents operating within those systems. Only in social systems there may exist an idea of the future ‘operating’ on agents in the present, since agents react to expectations and incorporate them into their behaviours and plans of action. This ‘causal retroaction’ of an idea of the future (Dupuy, 2014) is indeed typical of social systems, and which purely mechanical complex systems are entirely devoid of. By means of that idea of the future, agents can effectively coordinate, sometimes spontaneously, to make that future possible. So, even if the future is unpredictable in the strong sense, it remains the essential means of coordination among agents. In this sense, the effectiveness of a future scenario constructed as a shared vision hinges on its capacity to successfully incorporate its impact on the agents, while avoiding ‘prediction paradoxes’ of planning (Alfasi and Portugali, 2007: 169; Portugali, 2008). This possibility of ‘coordination by means of future’ (Dupuy, 2014) is, on close reflection, the essential condition of possibility and efficacy of any planning.⁶

The space of the projects

Finally, there is a flexible plane for actions and projects of the public planning and of individuals in the social forms they decide to choose. At the level of this plane resides the possibility of coordination along the backbone of the via negativa and the shared vision. Since the via negativa is not meant to pursue ‘positive’ goals possibly heralded by the shared vision, this drive towards positive objectives is offered by what we call the space of the projects, a flexible space with stable boundaries and constraints for the individual action and for public intervention. A space which combines top-down with bottom-up, short and medium-term, possibly reversible, modular, even ephemeral, actions. This ‘space of the project’ is, as it were, a sort of via positiva, a space for action (1) constrained by the via negativa and (2) compatible with the shared vision. Such space contemplates private transformations, public planning (Moroni and Cozzolino, 2019), private–public partnerships, but also embraces bottom-up practices of self-management, of temporary uses, and of tactical interventions (Silva, 2016).

On legitimacy of antifragile planning: antifragility of what to what?

Before concluding, we want to examine the question of legitimacy of antifragility as a public policy goal. We especially want to address this question in response to Kolers (2016). While making the case for resilience as a condition of legitimacy and justice of public policy, Kolers denies this status to antifragility. His proposition that antifragility is illiberal is grounded in the fact that antifragility is not just contextual but also level-relative:

The citizens’ affairs cannot all be anti-fragile, because in many cases the anti-fragility of some involves capitalizing on the fragility of others. And the state or community cannot itself be anti-fragile because part of its function is to absorb some of its citizens’ fragility. As a state aim, anti-fragility is therefore illiberal. (Kolers, 2016: 95)

Kolers is correct to point out, as Taleb makes it explicit, that antifragility on the aggregate systemic level often emerges at the expense of fragility at the individual level. In such cases, the system benefits from the fragility (i.e. exposition to harm) of the individuals. For example, in biological evolution the source of antifragility of a species, of an ecosystem, and more so of the entire ecosphere, largely depends on the relative fragility of individual organisms. This indeed is an essential feature of the mechanisms of natural selection. Closer to social systems, this condition is similar to Schumpeter’s (1942) reflections on ‘creative destruction’ through entrepreneurship and innovation in industrial-entrepreneurial districts, and in capitalism in general. So indeed, a system is antifragile if it gains from the fragility of its components. Such antifragility is obtained because in this way, the system incorporates mechanisms of learning, it advances through trials and errors, and it improves through tinkering, local experimentations and failures. Here Kolers is correct to raise the core issue with antifragility as a political ideal, since in social systems some of those low-level components are humans and social groups, whose fragility can be a legitimate concern of the state, and whose treatment, liberties, well-being and security are valuable goals, not on an unconstrained disposal for being fragilised, let alone right off sacrificed, for the sake of ‘the greatest antifragility for the greatest number’.

In the case of urban systems, one could wonder if their antifragility may be at the expense of the fragility and survival of some individuals and social groups. We can offer two answers. The first is that sometimes it happens in the evolution of cities that social groups, functions and economic activities disappear, get transformed, change their nature, and there could be limits to what a public policy can and is admissible to do about it. The second answer, though, is that a city without its people does not exist, and that people are citizens, a fact with a series of implications in terms of individual rights, social justice and public policy.

But perhaps Kolers’ assessment of antifragility is too hasty. Taleb does not actually attempt, nor was that his aim in Antifragile, to work out a feasible political theory. But even if he were to lay out one with excessive shades of social Darwinism, up to jeopardising essential liberal principles, we wouldn’t be bounded to be fundamentalist Talebians, nor would that be the only possible, let alone the most productive theory. In essence, we object that the level-relative nature of antifragility poses an insurmountable obstacle to the construction of a workable legitimisation of antifragility as a political ideal.

In general, political theory is not unfamiliar with dealing with level-relative concerns. For instance, the two-tier structure of Rawls’ principles of justice operates with such concerns in mind, when the priority of the liberty over the difference principle wards off certain treatments on the ‘lower’ level (i.e. violation of equal right to individual liberties) at the expense of the performance on the ‘higher’ level (i.e. maximisation of the benefit of the least-advantaged members of society). In the context of planning theory, Moroni (2019) has convincingly argued that the distinction between such level-relative concerns is not only possible, but unavoidable and necessary.

So, rather than following Kolers to altogether dismiss antifragility as illiberal, we claim that we need to examine what kind of antifragility may be legitimate to pursue, and, considering its inevitable level-dependency, at what level should it be pursued. To paraphrase the question others have asked about resilience itself (Carpenter et al., 2001; Davoudi and Porter, 2012), the point is to ask ‘antifragility of what to what?’ Indeed, the desirability of resilience depends on how one answers the question ‘resilience of what to what?’ (Carpenter et al., 2001): it depends if the system to be made resilient is functional and valuable as-is and what its permanent disruption would bring about. If the status quo is undesirable, making it resilient may not be a valuable goal. To justify resilience as a legitimate political goal, Kolers (2016) answers the question of ‘resilience of what to what’ by clarifying that it should regard ‘the resilience of reasonably just social systems to disruptions that could significantly reduce quality of life and undermine democracy, [and therefore the resilience of such systems] would seem to be of important moral concern’.

Now, taking into consideration our previous point on optionality in antifragility, a similar argument may also be made for antifragility: the goal of antifragility can be pursued for valuable systems, by endowing them with optionality and asymmetry of gains versus harms, in order to increase the chances for them to evolve favourably, as long as acceptable trade-offs with regard to potential individual fragility can be stipulated. Under this condition antifragility can become a legitimate normative goal.

When thinking of stipulating such trade-offs starting from the minimal definition of antifragility as the possibility of ‘gain from disorder’, the central controversial question in public policy and planning becomes what should be elected as the informational focus to define and measure that ‘gain’. This question can have a descriptive variant (what gain? at what level does the gain show up?, who gains?, at the expenses of whom or of what?), and a normative variant (who should gain? at what level should the gain show up?).

Here, for both variants of the question, we hold that the capability approach can fruitfully come to assistance and provide the adequate informational focus to define and measure the relevant ‘gain’ (Sen, 2009). First, because capabilities are tightly related to the idea of optionality: to extend someone’s capabilities is to extend one’s options to be able to choose, among those attainable, those functionings one values the most. Second, the capability approach points to a possible way to answer the question of ‘antifragility of what to what’, by working it out in a certain analogy with Rawls’ two principles of justice:

First principle: to the classical liberal individual rights and protections (which better be robust/resilient), we should add individual capabilities (whose extension better be antifragile);

Second principle: to pursue the goal of systemic antifragility compatible with the first principle.

Since with this the system would be made antifragile through extending capabilities by endowing individuals with optionality, this in part would mean to make individuals themselves antifragile. Overall, it seems to us possible to acknowledge that there is a favourable terrain for synergies, constructive interaction, positive feedbacks and ultimately beneficial trade-offs between our two principles.

We must admit there is a limit to such a Rawlsian analogy. Given that capabilities are differentiated and purpose-oriented, they are in principle not as universal as the basic rights and are not as purpose-neutral as the Rawlsian conception of the primary goods.⁷ So, there may arise the need to stipulate operational trade-offs when defining the relevant set of capabilities and, if necessary, which capabilities should be prioritised over which, and over what systemic antifragility. Indeed. But this observation pertains to the objections raised to the capability approach in general, so be we allowed not to wage this debate here (Brighouse and Robeyns, 2010), other than suggesting that these trade-offs may in planning be the proper object of what we called the plane of the shared vision, subject to public debate and democratic deliberation.

Ultimately, the need to stipulate level-relative trade-offs and to define relevant capabilities brings us to our claim that antifragility is only a partial goal for planning. We here follow Dupuy’s distinction between the objective analysis of the self-organising structure of social systems and the value judgement of what they do to individuals (Dupuy, 2014: 15). That in our case means to acknowledge that the question ‘what fragilises or antifragilises an urban system’ is not the same question as ‘what is just and what is unjust’. The second, normative question cannot be addressed and settled as a mere problem of expediency of action and policy under uncertainty and complexity. But although two different questions, they should not be answered independently. It is our core claim that these two questions in planning need to be addressed jointly, in order to make the answers we give to these questions normatively and operationally compatible one with the other.

Conclusion

We have sketched out what an antifragile planning could mean and discussed how its legitimacy as a public policy goal may be constructed.

Before ending, we want to ward off any exclusivism or fundamentalism of antifragility. From all we have said does not follow that cities should only be made of antifragile things and subsystems. Urban systems are composed of many components and subsystems: some fragile, some robust, some resilient and some antifragile. There are many sufficient and efficient reasons for that to be so, so antifragility is not the goal to pursue for everything, in every case and at any cost. There are fragile components and subsystems which at least in part own their raison d’être, value, importance or beauty, to their fragility; sometimes fragility is pursued for the charm and allure of the delicate, ephemeral, frangible, weak. Then there are objects and systems which are robust, and they better be designed and built that way, for reasons of functionality, economy, durability and simplicity. Then there are resilient systems, altogether functioning satisfactorily: in the strict sense, urban resilience is a desirable feature of valuable systems in response to adverse events, infrastructure failures, environmental disasters and social crises (Pickett et al., 2004). All in all, there are many objects and subsystems within urban systems which are fragile, robust, resilient and antifragile, and as such they are necessary and useful for their survival and functioning, and for human well-being.

Having said that, we hold that antifragility can be elected as a more general and superordinate principle for an urban system as a whole, and that then the system can and should within it include many components and subsystems which are, from case to case, made resilient, robust, fragile and antifragile. In our view, antifragility, like resilience, is an attempt to organically introduce within public policy the idea of complexity and of the ineliminable uncertainty of the future, coupled with care for future collective outcomes. In this sense, the two goals can coexist. Taking notice that the desirability of resilience and of antifragility may apply to different systems and scopes, it is of course valuable to pursue the resilience of systems to disruptions that could significantly reduce the quality of life and undermine democracy. But in general, also in the light of our core claim that resilience is a limit case of antifragility, for urban systems, we submit that antifragility is a more general and superordinate principle, and thus a worthwhile political goal.

There is perhaps one lingering question we need to address. Would antifragility of urban systems be possible without planning, without some form of governing or collective management of spatial transformations? The answer is, in a sense possibly yes: we cannot exclude that an unmitigated anarchy or purely market-based mechanisms would not at the end exhibit some form of antifragility at the systemic level. But ultimately, the answer to the question must be no, if we want to address Kolers’ preoccupation for the level-relative outcomes, for the antifragility of some capitalising on the fragility of others. In this sense, planning cannot avoid politico-ethical concerns. Only, our point is, it has to incorporate the awareness of complexity and uncertainty, and thus the goal of antifragility. So, we are returning here to our previous point of ‘what is antifragile’ and ‘what is just’ being two different questions. Two different questions that in planning need to be jointly and simultaneously addressed, so that the answers we ultimately give to these questions are compatible one with the other.

By discussing the notion of antifragility and comparing it with resilience, our primary purpose here was to point towards a way to broaden the general coordinates of the debate on planning under complexity and uncertainty. We hope that we laid out at least parts of the groundwork necessary to claim that antifragility is a contentful and legitimate partial goal for planning, and that therefore planning ought to incorporate antifragility into both theory and practice. We acknowledge that much theoretical and empirical groundwork has yet to be covered, especially on the contextualisation and application of the conceptual framework to different planning problems and trade-offs, for ours to ultimately prove to be a valuable contribution to the debate.

Footnotes

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by the grant for the research projects “Healthy Cities and Smart Territories” (2016/17) funded by Fondazione di Sardegna and the Autonomous Region of Sardinia, Italy.

Notes

Author biographies

Ivan Blečić is associate professor at the Department of Civil & Environmental Engineering and Architecture, University of Cagliari, Italy. His research interests include planning theory, evaluation methods and decision support for planning and territorial policy, urban economics, and urban modelling and simulation.

Arnaldo Cecchini, until 2018 full professor in Planning at the Faculty of Architecture in Alghero (Department of Architectture, Design and Urbanism, University of Sassari, Italy. He was the Dean of Faculty from 2011 to 2015. His research interests include techniques for urban analysis and modelling such as catastrophe theory, urban simulation, gaming simulation, cellular automata, scenario techniques and information systems for public participation.

References

Acemoglu

Robinson

(2012) Why Nations Fail: The Origins of Power, Prosperity, and Poverty. New York: Crown Publishing Group.

Ahern

(2011) From fail-safe to safe-to-fail: Sustainability and resilience in the new urban world. Landscape and Urban Planning 100(4): 341–343.

Albrechts

(2010) More of the same is not enough! How could strategic spatial planning be instrumental in dealing with the challenges ahead? Environment and Planning B: Planning and Design 37(6): 1115–1127.

Albrechts

Balducci

(2013) Practicing strategic planning: In search of critical features to explain the strategic character of plans. disP – The Planning Review 49(3): 16–27.

Alexander

Mazza

Moroni

(2012) Planning without plans? Nomocracy or teleocracy for social-spatial ordering. Progress in Planning 77(2): 37–87.

Alfasi

(2018) The coding turn in urban planning: Could it remedy the essential drawbacks of planning? Planning Theory 17(3): 375–395.

Alfasi

Portugali

(2007) Planning rules for a self-planned city. Planning Theory 6(2): 164–182.

Anderson

(1972) More is different. Science 177(4047): 393–396.

Ansar

Flyvbjerg

Budzier

, et al. (2017) Big is fragile. In: Flyvbjerg

(ed.) The Oxford Handbook of Megaproject Management. Oxford: Oxford University Press, 60–95.

10.

Basta

(2016) From justice in planning toward planning for justice: A capability approach. Planning Theory 15(2): 190–212.

11.

Batty

(2007) Cities and Complexity: Understanding Cities with Cellular Automata, Agent-Based Models, and Fractals. Cambridge, MA: MIT Press.

12.

Blečić

(2013) Costruzione degli scenari per la pianificazione. Milan: FrancoAngeli.

13.

Blečić

Cecchini

(2016) Verso Una Pianificazione Antifragile: Come Pensare al Futuro Senza Prevederlo. Milan: FrancoAngeli.

14.

Blečić

Cecchini

(2017) On the antifragility of cities and of their buildings. City, Territory and Architecture 4(1): 3.

15.

Blečić

Cecchini

Talu

(2013) The capability approach in urban quality of life and urban policies: Towards a conceptual framework. In: Serreli

(ed.) City Project and Public Space. Urban and Landscape Perspectives, vol. 14. Dordrecht: Springer.

16.

Blečić

Cecchini

Talu

(2018) Capability approach and urban planning: Fertile urban capabilities and quality of urban life of the most disadvantaged inhabitants. Archivio di Studi Urbani e Regionali 48(122): 34–52.

17.

Brand

(1994) How Buildings Learn: What Happens after They’re Built. New York: Viking Press.

18.

Brighouse

Robeyns

(2010) Measuring Justice: Primary Goods and Capabilities. Cambridge: Cambridge University Press.

19.

Carpenter

Walker

Anderies

, et al. (2001) From metaphor to measurement: Resilience of what to what? Ecosystems 4(8): 765–781.

20.

Chettiparamb

(2019) Responding to a complex world: Explorations in spatial planning. Planning Theory 18(4): 429–447.

21.

Cozzolino

(2018) Reconsidering urban spontaneity and flexibility after Jane Jacobs: How do they work under different kinds of planning conditions? Cosmos + Taxis 5(3): 12.

22.

Davoudi

Porter

(eds) (2012) Applying the resilience perspective to planning: Critical thoughts from theory and practice. Planning Theory & Practice 13(2): 299–333.

23.

Davoudi

Brooks

Mehmood

(2013) Evolutionary resilience and strategies for climate adaptation. Planning Practice and Research 28(3): 307–322.

24.

De Roo

Hillier

(2012) Complexity and Planning: Systems, Assemblages and Simulations. London: Routledge.

25.

Duit

(2016) Resilience thinking: Lessons for public administration. Public Administration 94(2): 364–380.

26.

Dupuy

J-P

(2014) Economy and the Future: A Crisis of Faith. East Lansing, MI: MSU Press.

27.

Ellerman

(2005) How do we grow? : Jane Jacobs on diversification and specialization. Challenge 48(3): 50–83.

28.

Flyvbjerg

(ed.) (2017) The Oxford Handbook of Megaproject Management (Oxford Handbooks). Oxford: Oxford University Press.

29.

Flyvbjerg

Bruzelius

Rothengatter

(2003) Megaprojects and Risk: An Anatomy of Ambition. Cambridge: Cambridge University Press.

30.

Fusco

Caglioni

Emsellem

, et al. (2017) Questions of uncertainty in geography. Environment and Planning A: Economy and Space 49(10): 2261–2280.

31.

Godet

(2001) Creating Futures: Scenario Planning as a Strategic Management Tool. Washington, DC: Brookings Institution Press.

32.

Goldman

(1964) Lindy’s Law. The New Republic. 150th edn, 13 June. Available at: https://www.gwern.net/docs/statistics/1964-goldman.pdf

33.

Hakim

(2014) Mediterranean Urbanism: Historic Urban / Building Rules and Processes. Berlin: Springer.

34.

Hilier

(2012) Baroque complexity: ‘If things were simple, word would have gotten round’. In: de Roo

Hillier

(eds) Complexity and Planning: Systems, Assemblages and Simulations. London: Routledge, 37–73.

35.

Ingram

Hong

(eds) (2012) Value Capture and Land Policies: Proceedings of the 2011 Land Policy Conference. Cambridge, MA: Lincoln Institute of Land Policy.

36.

Innes

Booher

(2010) Planning with Complexity: An Introduction to Collaborative Rationality for Public Policy. London: Routledge.

37.

Jacobs

(1961) The Death and Life of Great American Cities. New York: Random House.

38.

Johnson

(2010) Where Good Ideas Come from. New York: Riverhead Books.

39.

Kato

Ahern

(2008) ‘Learning by doing’: Adaptive planning as a strategy to address uncertainty in planning. Journal of Environmental Planning and Management 51(4): 543–559.

40.

Kolers

(2016) Resilience as a political ideal. Ethics, Policy and Environment 19(1): 91–107.

41.

Las Casas

Scorza

(2017) A renewed rational approach from liquid society towards anti-fragile planning. In: Gervasi

Murgante

Misra

, et al (eds) Computational Science and Its Applications – ICCSA2017. Lecture Notes in Computer Science. Cham: Springer International Publishing, 517–526.

42.

Mandelbrot

(1982) The Fractal Geometry of Nature. New York: W. H. Freeman and Company.

43.

Meerow

Newell

Stults

(2016) Defining urban resilience: A review. Landscape and Urban Planning 147: 38–49.

44.

Moroni

(2010) Rethinking the theory and practice of land-use regulation: Towards nomocracy. Planning Theory 9(2): 137–155.

45.

Moroni

(2015a) Complexity and the inherent limits of explanation and prediction: Urban codes for self-organising cities. Planning Theory 14(3): 248–267.

46.

Moroni

(2015b) Libertà e innovazione nella città sostenibile: ridurre lo spreco di energie umane. Roma: Carocci.

47.

Moroni

(2019) Constitutional and post-constitutional problems: Reconsidering the issues of public interest, agonistic pluralism and private property in planning. Planning Theory 18(1): 5–23.

48.

Moroni

Cozzolino

(2019) Action and the city. Emergence, complexity, planning. Cities 90: 42–51.

49.

Pickett

STA

Cadenasso

Grove

(2004) Resilient cities: Meaning, models, and metaphor for integrating the ecological, socio-economic, and planning realms. Landscape and Urban Planning 69(4): 369–384.

50.

Portugali

(2000) Self-Organization and the City (Springer Series in Synergetics). Berlin; Heidelberg: Springer-Verlag.

51.

Portugali

(2006) Complexity theory as a link between space and place. Environment and Planning A 38(4): 647–664.

52.

Portugali

(2008) Learning from paradoxes about prediction and planning in self-organizing cities. Planning Theory 7(3): 248–262.

53.

Portugali

(2012) Complexity theories of cities: Achievements, criticism and potentials. In: Portugali

Meyer

Stolk

, et al. (eds) Complexity Theories of Cities Have Come of Age. Berlin: Springer, 47–62.

54.

Portugali

Meyer

Stolk

, et al. (eds) (2012) Complexity Theories of Cities Have Come of Age: An Overview with Implications to Urban Planning and Design. Berlin; Heidelberg: Springer-Verlag.

55.

Rauws

(2017) Embracing uncertainty without abandoning planning. disP – The Planning Review 53(1): 32–45.

56.

Rawls

(1996) Political Liberalism. New York: Columbia University Press.

57.

Roggema

(2019) Design for disruption: Creating anti-fragile urban delta landscapes. Urban Planning 4(1): 113–122.

58.

Sandel

(2012) What Money Can’t Buy: The Moral Limits of Markets. London: Macmillan.

59.

Schumpeter

(1942) Capitalism, Socialism, and Democracy. New York: Harper & Brothers.

60.

Sen

(2009) The Idea of Justice. Cambridge, MA: Harvard University Press.

61.

Silva

(2016) Tactical urbanism: Towards an evolutionary cities’ approach? Environment and Planning B: Planning and Design 43(6): 1040–1051.

62.

Simmie

Martin

(2010) The economic resilience of regions: Towards an evolutionary approach. Cambridge Journal of Regions, Economy and Society 3(1): 27–43.

63.

Skrimizea

Haniotou

Parra

(2018) On the ‘complexity turn’ in planning: An adaptive rationale to navigate spaces and times of uncertainty. Planning Theory 18: 122–142.

64.

Swanstrom

(2008) Regional resilience: A critical examination of the ecological framework. Working Paper 2008-07, 25 April. Berkeley, CA: Institute of Urban and Regional Development (IURD), University of California.

65.

Taleb

(2012) Antifragile: Things That Gain from Disorder. 1st edn. New York: Random House.

66.

Wohl

(2018) Tactical urbanism as a means of testing relational processes in space: A complex systems perspective. Planning Theory 17(4): 472–493.

67.

Zellner

Campbell

(2015) Planning for deep-rooted problems: What can we learn from aligning complex systems and wicked problems? Planning Theory & Practice 16(4): 457–478.